No title

EDITORIAL 325 Editorial I nnovation matters – of course. Unless organizations change what they offer the world and t...

Author: Petra de Weerd-Nederhof and Olaf Fisscher (Editors)

72 downloads 846 Views 941KB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

EDITORIAL

325

Editorial

I

nnovation matters – of course. Unless organizations change what they offer the world and the ways they create and deliver it they could well find themselves in trouble. But although most organizations espouse the principle their underlying models of how innovation works are often limited – with the risk that they pay attention to some aspects but miss out on others. Smart firms constantly seek to extend and enrich the models that they used to drive the process. And in order to build better models we need to carry through research in the context of its application – innovation is above all a practice! Examples of emerging challenges that require us to develop our models of ‘good practice’ in innovation management include: • How to deal with an increasingly international game in which knowledge production is running at nearly US$1 trillion/year R&D spend (in the public and private sector) – and how to manage the knowledge flows around the world that converts this research into innovation? • How to mobilize and sustain high involvement from employees in the innovation process, which organizations of all shapes and sizes need to retain a competitive edge? • How to deal with the challenges of networked and ‘open’ innovation, especially around the ways in which we construct and manage inter-organizational networks? • How to deal not only with ‘steady-state’ innovation (doing what we do, but better) but also ‘discontinuous’ innovation in which radical shifts in technologies, markets or the political and competitive environment mean that the old recipes may not work and we need to learn new rules of the game? Challenges of this magnitude do not find answers in textbooks, but rather in the process of organizational learning. As Nelson and Winter pointed out, the process of creating what eventually become ‘innovation routines’ – and their associated structures, systems and procedures – begins with trial and error. So the research agenda for dealing with issues like these is essentially one built around shared © 2006 The Author Journal compilation © 2006 Blackwell Publishing

experimentation and capture of helpful lessons – a process of ‘probe and learn’. Innovation research needs a range of methodological approaches, but one valuable line in this context might be the creation of ‘learning laboratories’ – configurations in which researchers and practitioners can interact around emerging and challenging themes across the innovation agenda. One such laboratory in which new and emerging models have been discussed and explored is CINet – the Continuous Innovation Network (http://www.continuousinnovation.net) – an international group of researchers sharing common concerns about the effective management of innovation. Since its inception in the early 1990s, CINet has been a meeting place for practice-orientated research, and has shared this via several channels, including a regular conference. CINet is the third and final association with which Creativity and Innovation Management is now affiliated. In this issue of Creativity and Innovation Management we have included four of the papers nominated for the ‘best paper’ award at the 2005 conference in Brighton, UK. The guest editor of this special is John Bessant, who is also a member of our editorial board, and is now of the Innovation Studies Centre of Imperial College, London. In 2005 he was the local host and organizer of the CINet conference. Together, the four contributions in this special provide a good overview of the range of research and quality of work across this network – and highlight some key challenges around which further research is certainly needed. But they also indicate the ways in which partnership research engaging practitioners and researchers can take place. The article by Francis Jørgensen, Harry Boer and Bjørge Timenes Laugen looks at the question of how firms develop and sustain a culture of high involvement innovation. How do different organizations engage their employees in the task of innovation – and how to they embed this approach? The original model that they explore derived from research at the University of Brighton and was essentially normative – suggesting a linear evolution of capability. That model said little about the

Volume 15

Number 4

2006

doi:10.1111/j.1467-8691.2006.00411.x

326

CREATIVITY AND INNOVATION MANAGEMENT

ways in which implementation might take place – an issue which the article explores in some depth. Using case-study data the authors suggest much more variation – many roads lead to Rome in the implementation of continuous improvement – and provide a valuable extension of the model. The article also provides practical guidelines about how managers might approach the implementation task of how to sustain and enable continuous improvement. The article by Astrid Heidemann Lassen, Frank Gertsen and Jens Ove Riis draws upon five cases to explore three core propositions around the theme of entrepreneurship. Their concern is particularly with corporate entrepreneurship – the need to push the boundaries and stretch the organization beyond ‘straightforward’ exploitation innovation strategies towards more explorative kinds. In doing so, they contribute to the continuing debate around ‘discontinuous’ innovation and its management. The article by Rik Middel, Harry Boer and Olaf Fisscher takes as its theme the increasing inter-organizational concerns around innovation. We are bombarded in the twenty-first century with models of innovation which recognize that it is a multi-player game involving firms and inter-firm relationships, but there is still a lack of good research around the theme of inter-organizational innovation. This article contrasts the ideas of continuous improvement and collaborative innovation, and explores the role of different contexts through which such interactions can take place. Lastly, the paper by Rasmus Kaltoft, Harry Boer, Ross Chapman, Frank Gertsen, and Jacob S. Nielsen extends this inter-organizational innovation theme and offers an outline model for how collaborative improvement activity might take place. It outlines ten key factors emerging from their research, around which firms are exploring and developing their working models for such collaboration. None of these articles – or the other papers presented at the conference – pretend to offer ‘the’ definitive answer to the innovation challenges that they are exploring. But they do provide well-researched insights that help to light the way to dealing with these more effectively. They also offer practitioners structures, models and guidelines that help to codify the valuable experience gained through what are still experiments at the front-line of innovation management.

Additional contributions Next to this worthwhile special, in this last issue of the year we offer you a varied selec-

Volume 15

Number 4

2006

tion of five other contributions, by authors from all around the globe. The topics range from the influence of localized clusters on creativity of inventors through knowledge work in game development and decision-making instruments in innovation processes to a case description of the development of a training wrist computer, and, to begin, a competencebased model of initiatives for innovations. Katrin Talke, Sören Salomo and Nils Mensel are concerned with the initiation of innovation processes in new product development. They aim at shedding some light on the starting point of initiative formation by examining the initiative emergence process and competences facilitating initiative formation. Their competence-based model explains the occurrence of initiatives by drawing on literature from individual competence models, creativity and motivation theory. This model relates initiatives to individual competence arenas, including task-related, action-related and cognitive competences. Talke et al.’s research offers academic value by providing a more comprehensive understanding of the emergence of initiatives and by developing a series of research propositions, while also offering practical value as they suggest possible avenues for the support of initiative competences. Hans Heerkens’ contribution departs from the notion that most decision support methods elicit importance judgements but do not help innovators with the mental processes leading to the judgement. Seeing innovations as chains of non-routine decisions, Heerkens describes the pitfalls in the ‘importance assessment’. He outlines the use and effect of a few simple instruments which may provide betterfounded importance judgements that can be better communicated to other actors involved in innovation processes. In the next contribution, the case study of the Suunto t6 training wrist computer is used by Seppo Hänninen and Ilkka Kauranen to introduce and discuss a Multidimensional Product-Concept Model Enhancing Cross-Functional Knowledge Creation in the Product Innovation Process. Then, Sherwat Ibrahim, M. Hosein Fallah and Richard Reilly address the question of what contributes to the creative environment of localized clusters that increases innovation output, presenting the results of a study investigating the environment of localized clusters and how it affects the creativity of individual inventors working in those clusters. Their study supports earlier research on geographical clusters and also provides further insights into the attributes that contribute to innovation output of clusters. Peter Zackariasson, Alexander Styhre and Timothy L. Wilson’s contribution on phronesis © 2006 The Author Journal compilation © 2006 Blackwell Publishing

EDITORIAL

and creativity is also the closing contribution of this last issue of 2006. This article presents a study of the knowledge work involved in the development of video games, the success of which is based on the ability to create a sense of immersion for the gamers. In the case presented here, dedicated gamers were also preferred when hiring personnel to develop the games. Discussing the know-how of this specific group in terms of phronesis, the detailed and practical understanding of a particular field enables an understanding of the idiosyncratic competence of this group and its importance for the development process. The study points at a number of implications for innovation and creativity work, for instance, the emphasis on the type of know-how embodied by the notion of phronesis that is not easily represented and codified. In addition, the videogame development model suggested captures the open-ended nature of video game development, enabling a substantial degree of continuous adjustment the comments and suggestions provided by the gamers. At the very back of the issue you will find two book reviews written by Jamie Burton (University of Manchester) and Jeroen Kraaijenbrink (University of Twente), for: ‘Product and Services Management’ and ‘Knowledge Spillovers and Knowledge Management’, respectively, which we also hope

© 2006 The Author Journal compilation © 2006 Blackwell Publishing

327

you will enjoy reading. At the end of this volume we look forward to our first issue of 2007, with a mini special on creative spaces, put together by James Moultrie, Remko van der Lugt, Sebastian Jansen and Udo Ernst Haner, and a selection of articles based on presentations at the 9th ECCI in Lodz, Poland. A call for the 10th ECCI, taking place in Copenhagen, and for the 2007 R&D Management conference in Bremen, are also included in this issue. As has become our custom we also publish a list of reviewers, and we want to thank each and every one of them for their crucial work for the journal. Our appreciation also goes to the team at Blackwell’s, and specifically to Jeannette Visser-Groeneveld, who has done a terrific job in implementing Manuscript Central this year. By the time that you read this editorial, we would like to advise you to also check out our website, where we will announce the first ‘best paper’ award for a Creativity and Innovation Management contribution, which we hope will be the start of a good tradition. All the best for 2007, and may it be a creative and innovative year for all of us! John Bessant Petra de Weerd-Nederhof Olaf Fisscher

Volume 15

Number 4

2006

328


CI Implementation: An Empirical Test of the CI Maturity Model Frances Jørgensen, Harry Boer and Bjørge Timenes Laugen There are a number of tools available for organizations wishing to measure and subsequently develop Continuous Improvement (CI). In this article, we review and evaluate a well-accepted CI development model, namely the CI Maturity Model (Bessant and Caffyn, 1997), against data collected from the 2nd Continuous Improvement Network Survey and a number of empirical cases described in the literature. While the CI Maturity Model suggests that CI maturation ought to be a linear process, the findings in this article suggest that there are feasible alternatives for companies to develop CI capability.

Introduction ontinuous Improvement (CI) is the ‘planned, organized and systematic process of ongoing, incremental and companywide change of existing practices aimed at improving company performance’ (Boer et al., 2000). Based on Scientific Management, improved under the heading Industrial Engineering, exported to and perfected in Japan in the form of kaizen, CI has found its way to industry worldwide and is currently regarded as one of the cornerstones of good management. CI is a form of incremental innovation and the same holds for the process of developing CI capability, which is generally reported to take considerable (learning) time. However, the process by which this development occurs is not yet fully understood. According to Bessant and Caffyn (1997), CI matures through a relatively linear, incremental, five-stage learning process (see Figure 1) involving the accumulation of sets of certain behavioural routines (see Table 1). The model has a number of merits. For example, it can be used for auditing or evaluating the status of improvement at various levels in an organization and it has an inherent attractiveness for companies wishing to know which areas of improvement should be targeted in order to further develop the improvement process. Still, empirical studies of CI implementation do not fully support Bessant and Caffyn’s sug-

C

Volume 15

Number 4

doi:10.1111/j.1467-8691.2006.00404.x

2006

gestions, especially with regard to the linearity suggested in the model. For instance, Rijnders (2002), Savolainen (1999) and Jørgensen (2003) all describe situations in which CI maturation has followed a rather random and non-linear pattern. In addition, there may be support for desiring a non-linear development of CI, as some behaviours and activities underpinning this capability may take precedence over others. A simple explanation for the incongruence between the CI Maturity Model and the results of the cited implementation studies is that the former model is normative, while the latter publications are based on case studies of what is actually happening in companies implementing CI. Accepting this explanation would then suggest the need for an updated, empirically based model of CI maturity, preferably going beyond single or a limited set of case studies. Building such a model first and foremost requires a review and exploration of the CI Maturity Model and other related models in relation to the available empirical data, followed by the development of a number of hypotheses and/or critical questions in order to focus further research. The objective of this article is to conduct this first step in the development of an empirically based model of CI development. In the following section, we describe and discuss the CI Maturity Model, relating it to empirical studies found in the CI literature. Then we will present a set of © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

AN EMPIRICAL TEST OF THE CI MATURITY MODEL

hypotheses that will frame the empirical analysis of data collected through the 2nd Continuous Innovation Network (CINet) Survey conducted in 2004. We conclude the article with a discussion of the findings from this study and suggestions for further research needed to continue the development of an empirically based model of CI development.

The CI Maturity Model The CI Maturity Model (Bessant and Caffyn, 1997; Caffyn, 1999) describes the process of CI capability development as a five-stage process (see Figure 1): 1. ‘Natural’/background CI. There is no formal CI structure, problem solving is random, and the dominant mode of problem solving is by specialists. 2. Structured CI. There are formal attempts to create and sustain CI, and a formal problem solving process is used, supported by basic CI tools. CI is often parallel to operations. 3. Goal-oriented CI. All of stage 2, plus formal deployment of strategic goals and monitoring and measurement of CI against these goals. 4. Proactive/empowered CI. All of stage 3, plus the responsibility for CI is devolved to the problem solving units. 5. Strategic CI (the learning organization). CI has become a dominant way of life, involving everyone in the organization. Learning is automatically captured and shared. The CI Maturity Model assumes that successful CI implementation involves accumulating certain behavioural routines that represent organizational abilities. The core abilities and the key behaviours associated with the devel-

CI capability

Strategic CI Proactive/empowered CI Goal-oriented CI Structured CI Natural, ‘background’ CI

Time

Figure 1. The Five Stages of Maturity in the CI Maturity Model (based on Bessant and Caffyn, 1997; Caffyn, 1999) © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

329

opment of CI, as they are represented in the 2nd CINet Survey, are shown in Table 1. It is generally accepted that increases in CI maturity should correspond with improvements in operational performance, although empirical support for this relationship is lacking. Successful practice of these behavioural routines by the members of the organization reflects CI capability, defined by Caffyn (1999) as ‘the ability of an organization to gain strategic advantage by extending involvement in innovation to a significant proportion of its members’ (p. 1142). As members of the organization become more adept at incorporating these behavioural routines into their work processes, their CI maturity develops. Bessant and Francis (1999) describe the development from one maturity level to the next as a learning process that occurs over time. An assumption underlying the model is that this development occurs in a relatively linear fashion. In essence, this means that the organization must practice all of the behaviours associated with the first level of capabilities before focusing on the second level, and so on. A corollary to this assumption is that all of the capabilities are equally weighted, or of equal importance in relation to the development of CI maturity, and consequently, improved operational performance. For instance, the ‘ability to articulate and demonstrate CI values’ (Capability ‘F’) would not be more (or less) vital to improved operational performance than ‘the ability to move CI across organizational borders’ (Capability ‘D’). Kobayashi’s 20 Keys Model (1995), which was developed as a method of assessing and targeting improvement processes, is also founded on the premise that steady and gradual development on all capabilities is preferable to focusing on selected areas for improvement. In fact, Kobayashi (1995) explicitly emphasizes the interrelatedness of the improvement activities. He states that gaps of more than two developmental (i.e. maturity) levels across the behavioural groups will lead to decreases in operational performance. Neither model allows for the consideration that some capabilities or activities may be more vital to improved performance than others, or that some of the capabilities may actually be precursors of other capabilities, requiring development of the former before the latter can occur.

Normative Theory versus Reported Reality The CI Maturity Model is perhaps one of the most recognized normative frameworks for

Volume 15

Number 4

2006

330


Table 1. CI Capabilities and Key Behaviours (based on Bessant and Caffyn, 1997)

Volume 15

CI abilities

Corresponding behaviours

A: The ability to link CI activities to the strategic goals of the company Cronbach’s alpha: 0.909

Ideas and suggestions for improvement are responded to in a clearly defined and timely fashion – either implemented or otherwise dealt with Managers lead by example, becoming actively involved in the design and implementation of systematic ongoing improvement Managers support experimentation by not punishing mistakes, but by encouraging learning from them Managers support improvement processes by allocating sufficient time, money, space and other resources People (individuals/groups) initiate and carry through to completion, improvement activities – they participate in the process People make use of some formal problem finding and solving cycle People use appropriate tools and techniques to support their improvement activities People use measurement to shape the improvement process The organization recognizes in formal but not necessarily financial ways the contribution of employees to continuous improvement

B: The ability to strategically manage the development of CI Cronbach’s alpha: 0.868

Before embarking on initial investigation and before implementing a solution, individuals and groups assess the improvements they proposed against strategic objectives to ensure consistency Everyone understands what the company’s or their department’s strategy, goals and objectives are Improvement is an integral part of the individual’s or groups’ work, not a parallel activity Individuals and groups monitor/measure the results of their improvement activity and their impact on strategic or departmental objectives Individuals and groups use the organization’s strategy and objectives to focus and prioritize their improvement activities

C: The ability to generate sustained involvement in CI Cronbach’s alpha: 0.834

Individuals and groups are effectively working across internal (vertical and lateral) and external divisions at all levels People are oriented towards internal and external customers in their improvement activity Relevant improvement activities involve representatives from different operational levels Specific improvement projects are taking place with customers and/or suppliers

D: The ability to move CI across organizational boundaries Cronbach’s alpha: 0.837

Improvement activities and results are continually monitored and measured Ongoing assessment ensures that the organization’s processes, structure and systems consistently support and reinforce improvement activities Senior management make available sufficient resources (time, money, personnel) to support the continuing development of the company’s improvement system When a major organizational change is planned, its potential impact on the organization’s improvement system is assessed and adjustments are made as necessary

E: The ability to learn through CI activity Cronbach’s alpha: 0.625

Managers at all levels display leadership and active commitment to ongoing improvement When something goes wrong the natural reaction of people at all levels is to look for reasons why rather than to blame the individual(s) involved

F: The ability to articulate and demonstrate CI values Cronbach’s alpha: 0.871

Everyone learns from their experiences, both good and bad Individuals and groups at all levels share (make available) their learning from all work and improvement experiences Individuals seek out opportunities for learning/personal development (e.g. active experimentation, setting own learning objectives) Managers accept and, where necessary, act on all the learning that takes place People and teams ensure that their learning is incorporated into the organization by making use of the mechanisms provided for that The organization articulates and consolidates (captures and shares) the learning of individuals and groups

Number 4

2006

© 2006 The Authors Journal compilation © 2006 Blackwell Publishing


developing CI. According to the model, all of the CI capabilities are equally important for organizations wishing to improve their performance through CI implementation and, therefore, an organization should spread its efforts over development of all of the CI capabilities at approximately the same time. As many of the CI behaviours are interrelated, focusing on one area while neglecting others may sacrifice overall development. Experiences with actual CI implementations, however, indicate that CI development may occur in a much less orderly manner, with development of some CI capabilities taking precedence over others. Some organizations may have a logical justification for consciously prioritizing the development of one or more capabilities over others or this development may occur naturally due to certain conditions in a given organization. Based on a longitudinal study of CI implementation, Jørgensen (2003) suggests that an organization may have (good) reasons to develop specific CI capabilities before embarking on wider scale CI development. Specifically, she found that the existing culture and leadership styles may serve as barriers to team development and learning, and hence CI development in general. Thus, poor alignment of certain CI capabilities with others would need to be corrected before a wider CI initiative could be considered. In addition, Rijnders’ (2002) typology of CI implementation processes indicates that various organizational characteristics have a strong influence on the areas of focus that led to the most successful results from CI. Thus, the organizational context may determine which CI capabilities should be developed in which order. In no way do these studies, nor others described in the literature, provide the answer to how CI development does and should progress. What these studies indicate, however, is that there are questions as to how this development may occur and whether a linear development is preferable in all cases. In this article, we translate these questions into hypotheses, which we will subsequently test with data from the 2nd CINet Survey, as a first step in gaining a better understanding of CI development.

Hypotheses As mentioned above, the general underlying logic of the CI Maturity Model, as well as Kobayashi’s 20 Keys Model, is that operational performance will improve as the organization implements behaviours and/or activities associated with CI capabilities (or the 20 Keys, in the case of the latter). We will test this assumption according to the following hypothesis: © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

331

H0: Increases in CI maturity, measured in terms of the level of adoption of CI behaviours, will correspond to improvements in performance. The brief review of the empirical studies on CI implementation above suggests that if this null hypothesis is indeed accepted, it will be worthwhile to explore whether improved operational performance will only occur when CI develops relatively linearly. In other words, is improved operational performance dependent on relatively simultaneous development of all CI capabilities? To test this, we will examine whether there are any particular CI capabilities that are more likely than others to explain performance improvement. We hypothesize: H1: Only increases in all CI capabilities will correspond to improvements in performance. Acceptance of this hypothesis confirms the CI Maturity Model and, for that matter, the 20 Keys Model; rejection suggests there are more feasible ways of implementing CI successfully.

Methods Data Collection In order to test the hypotheses, we used data from the 2nd CINet survey. The survey is developed by an international research consortium, and comprises data from 543 manufacturing companies in 10 countries in Asia, Australia and Europe. The survey was conducted in 2004 and invited data from 2003 relating to a whole range of CI-related issues. To measure the level and use of CI behaviours, the respondents were asked to indicate to what extent they agreed or disagreed with 32 formulated statements related to CI activities. The statements are based on the constituent behaviours of Bessant and Caffyn (1997, pp. 19–20). All statements are measured on a 5point Likert-scale to indicate the degree to which the described behaviours are present in the organization.

Operationalization and Data Analysis We investigated the relationships between CI behaviours and improvement in performance in two steps. First, we performed a regression analysis with a variable indicating the respondents’ average total engagement in CI behaviours as the independent variable, and improvement of speed/cost, relationship and organizational performance as dependent variables. Similar to Bessant and Caffyn (1997), we grouped the CI behaviours into the six categories shown in Table 1, which also

Volume 15

Number 4

2006

332


reports the scale reliability of the respective behavioural groupings. Based on a factor analysis (see Table 2), we operationalized the performance variables as follows: • Speed/cost performance concerns on-time production and delivery, overall productivity and cost. • Relationship performance is associated with relations with customers, suppliers and other departments, customer satisfaction and quality conformance. • Organizational performance includes employee commitment and attitude towards change, employee skills and competencies, safety and working conditions, organization, communication and cooperation, administrative routines, and absenteeism. Second, we did a regression analysis with the three dimensions of performance as dependent variables, and the average occurrence of CI behaviours within each of the six categories as independent variables. In all the analyses we controlled for the influence from three contextual variables,

namely the importance of CI in the business unit, company size and production process type. We measured: • CI importance according to the responses to the CINet Survey item ‘How would you rate the overall importance of continuous improvement in your business unit? 1 = Vital, 2 = Of strategic importance, 3 = Of operational importance, 4 = Of minor importance, 5 = Not important’. • Company size according to the responses to the item ‘How many employees are there in the company?’ • Production process type according to the item ‘How would you describe the production system for your business unit’s most important product line? 1 = Project, 2 = Job shop, 3 = Batch, 4 = Line, 5 = Continuous’.

Results The results from the analyses are shown in Table 3. In general, the findings indicate that

Table 2. Factor Analysis of the Performance Indicators Factor

Speed/cost performance

Relationship performance

Organizational performance

Volume 15

Number 4

2006

Variable

Increased productivity Increased production volume Reduced cost Reduced lead times Improved delivery reliability Improved customer relations Higher customer satisfaction Improved supplier relations Improved quality conformance Improved relations between departments Improved safety and working conditions Increased employee commitment/attitude towards change Increased employee skills and competences Decreased absence Improved organization, cooperation and communication Improved administrative routines

Variable average

Factor loading

2,395 2,739 2,540 2,770 2,507 2,391 2,238 2,933 2,134 2,642

0.805 0.737 0.709 0.656 0.584 0.840 0.702 0.658 0.610 0.551

2,611

0.707

2,794

0.671

2,678

0.661

3,672 2,584

0.638 0.615

2,859

0.434

Factor average

Cronbach’s alpha

2,590

0.818

2,468

0.832

2,866

0.800



increased implementation of the CI behaviours is significantly related to higher improvement of performance.

A Group-by-group Analysis The strongest relationship between CI behaviours and the improvement of total performance is found for Group B behaviours, i.e. the ability to strategically manage the development of CI (Beta = 0.239, p < 0.01). The explanatory power of the model is also quite good (R2 = 0.152). Group B behaviours also have the strongest relationship with the improvement of speed/cost performance (Beta = 0.153, p < 0.001). The explanatory power of this model is somewhat lower than for the total performance improvement, however, still the highest among the six groups of CI behaviours (R2 = 0.117). Improving organizational performance is linked most strongly to and best explained by implementation of Group B behaviours (Beta = 0.210, p < 0.001, R2 = 0.126). Group C behaviours, i.e. the ability to sustain involvement of CI, have the weakest impact on performance improvement (Beta = 0.154, p < 0.01). This model also has the lowest explanatory power (R2 = 0.118) of the variance in performance. Further, Group C behaviours have the weakest relation to performance improvement in speed/cost (Beta = 0.092, p < 0.05, R2 = 0.102). Group C behaviours have the weakest link with improvement of relationship performance, and also the lowest explanatory power (Beta = 0.196, p < 0.001, R2 = 0.084). Similar to the other performance dimensions, Group C behaviours are linked most weakly to improving organizational performance, and also explain least of the performance variance (Beta = 0.109, p < 0.001, R2 = 0.094). Group D behaviours, i.e. the ability to move CI across organizational borders, have the strongest influence on improvement of relationship performance as well as the highest explanatory power of the variance in that performance area (Beta = 0.257, p < 0.001, R2 = 0.111). However, the figures are nearly the same as those for the total set of CI behaviours (Beta = 0.256, p < 0.001, R2 = 0.111) and the Group B (Beta = 0.254, p < 0.001, R2 = 0.110) and Group F behaviours (Beta = 0.246, p < 0.001, R2 = 0.106). The impact and explanatory power of the other three groups of CI behaviour, i.e. A (the ability to link CI activities to the strategic goals of the company), E (the ability to learn through CI activity) and F (the ability to articulate and demonstrate CI values), fall in between the extremes addressed above. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

333

Aggregate Analysis The analyses also reveal some interesting findings on a more aggregate level. For each of the six groups of CI behaviours as well as the total set of behaviours, the impact on relationship performance is strongest: all groups of CI behaviours have the highest Beta value for relationship performance, compared with the other performance groups. The explanatory power of the variance in performance is, however, lower for relationship performance than for the other dimensions. Group A, B, D and F behaviours have the highest explanatory power for organizational performance, while Group C and E behaviours explain most of the variance in speed/cost performance. This suggests that the adoption of CI behaviours first and foremost explains variance in organizational performance.

Influence from Control Variables In general, the variable indicating how important CI activities are considered in the company has a significant and positive relationship with all dimensions of performance improvement. Company size seems to be significantly related to improved speed/ cost performance, but does not affect the other performance dimensions. There are no significant relationships between the type of production process and any of the measures of performance.

Discussion The objective of this article was to test Bessant and Caffyn’s (1997) CI Maturity Model against empirical data derived from the results of the 2nd CINet Survey. The primary motivation for testing the model with a large dataset was to determine the degree to which this essentially normative model adequately represents what happens in actual CI implementations. Case studies (e.g. Rijnders, 2002; Savolainen, 1999; Jørgensen, 2003) suggest that companies implement the CI behaviours in a much less linear fashion than proposed in the CI Maturity Model. Furthermore, these studies suggest that various issues, for example those related to culture and leadership (Jørgensen, 2003), must be managed before embarking on CI development. Finally, these empirical studies suggest that specific characteristics of an organization may render some of the CI behaviours more important than others in terms of improving performance. In order to test the model relative to these case study-based findings as well as the theory-

Volume 15

Number 4

2006

334


Table 3. Regression Analyses Between Performance Dimensions (Dependent Variables), and CI Behaviours (Independent Variables) and Control Factors Dependent variables Overall performance

Speed/cost performance

Relationship performance

Organizational performance

Group A behaviours CI importance Employees Production process

0.108*** 0.295** −0.007 −0.083∗ R2 = 0.106

0.232*** 0.230*** 0.006 0.029 R2 = 0.099

0.178*** 0.297*** −0.028 −0.008 R2 = 0.113

Group B behaviours CI importance Employees Production process

0.153*** 0.291*** −0.005 −0.079∗ R2 = 0.117

0.254*** 0.219*** 0.012 0.036 R2 = 0.110

0.210*** 0.290*** −0.024 −0.002 R2 = 0.126

Group C behaviours CI importance Employees Production process

0.092** 0.293*** −0.009 −0.086∗∗ R2 = 0.102

0.196*** 0.226*** 0.002 0.022 R2 = 0.084

0.109** 0.292*** −0.028 −0.011 R2 = 0.094

Group D behaviours CI importance Employees Production process

0.114*** 0.294*** −0.011 −0.084∗ R2 = 0.107

0.257*** 0.227*** −0.002 0.025 R2 = 0.111

0.192*** 0.295 −0.034 −0.011 R2 = 0.118

Group E behaviours CI importance Employees Production process

0.132*** 0.297*** −0.007 −0.089∗∗ R2 = 0.111

0.230*** 0.230*** 0.009 0.019 R2 = 0.098

0.167*** 0.297*** −0.025 −0.015 R2 = 0.110

Group F activities CI importance Employees Production process

0.103*** 0.294*** −0.008 −0.089∗∗ R2 = 0.104

0.246*** 0.228*** 0.003 0.014 R2 = 0.106

0.187*** 0.296*** −0.030 −0.019 R2 = 0.116

Independent variables

Total set of CI behaviours CI importance Employees Production process

0.223*** 0.317*** −0.013 −0.029 R2 = 0.144

Note: The standardized coefficient Beta (numbers in italic) illustrates the relationships between the dependent variables, and independent and control variables. R2 presents the explanatory power of the regression models on the variance in performance. Significance levels: ***p < 0.01, **p < 0.05, *p < 0.1.

based literature, we formulated two hypotheses. First, we tested the hypothesis that increases in CI maturity, measured in terms of the level of adoption of CI behaviours, will correspond to improvements in performance. This

Volume 15

Number 4

2006

hypothesis is in fact central to all research and practice in the area of CI, as companies have adopted CI, and continue to do so, on the yet unsubstantiated premise that CI improves performance. Based on the results of the 2nd CINet © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


Survey, this hypothesis was accepted: adoption of CI has impact on all three performance categories distinguished in our analysis. Further analysis based on the second hypothesis, that only increases in all CI capabilities will correspond to improvements in performance, confirms, however, that the behaviours are not equally important and that specific behaviours have a stronger relationship with certain areas of performance improvement. These findings conflict with the assumptions of the CI Maturity Model as they suggest that different strategies for CI development may exist. Specifically, the analyses showed that behaviours associated with the ‘ability to strategically manage the development of CI’ (Group B) have the strongest influence on speed/cost performance (i.e. on-time production and delivery, overall productivity and cost) and organizational performance (i.e. employee commitment and attitude towards change, employee skills and competencies, safety and working conditions, organization, communication and cooperation, administrative routines, and absenteeism). An important consideration here is that the behaviours associated with this capability relate primarily to the degree to which employees understand the concepts and principles of CI and how they relate to the overall success of the company in which they work. This lends support to Jørgensen’s (2003) discussion on how lack of knowledge of CI principles, as well as the role of the individual employee and team in achieving organizational goals through CI, can serve as a serious barrier to successful CI. It will be necessary to conduct additional analyses of the relationship between specific behaviours within this group to determine whether this understanding of CI is more (or less) vital in terms of improved performance than the more strategically oriented behaviours within the same group. The behaviours associated with the ‘ability to move CI across organizational boundaries’ (Group D) are strongly related to, and also explain most of the variance in, improvement in relationship performance (i.e. performance associated with relations with customers, suppliers and other departments, customer satisfaction and quality conformance). Here, however, the impact and explanatory power of the total set and the Group B and F behaviours are quite similar. This suggests it does not really matter where CI implementation is started: the effects on relationship performance will be strong. The results of these analyses strongly suggest that CI maturity need not necessarily follow a linear progression in order to positively impact performance and that the development of certain capabilities may lead to improve© 2006 The Authors Journal compilation © 2006 Blackwell Publishing

335

ment of specific measures of performance. Therefore, the goals of the individual company could ostensibly be used to determine where CI development efforts should be targeted. By linking the goals solidly to specific CI efforts, it is possible that more emphasis would be put on CI, which in turn is shown to have a significant, positive relationship with all dimensions of performance improvement, according to the analysis of the control variables. On the other hand, the analyses presented here do not shed much light on the role of CI learning, leadership, and culture on improved performance. Although one of the six capabilities addresses CI learning (Group F), learning is also inherent in many of the other capabilities. Employees cannot be expected to, for example, use CI tools or to gain an understanding of the relevance of CI to the organization in the absence of a learning culture or other learning mechanisms. The same caution applies to leadership and the adoption of a CI culture, which are both assumed in many of the other groups of capabilities. The level of interrelatedness between learning, leadership, and culture and the other capabilities supports Jørgensen’s (2003) and Kaye and Anderson’s (1999) assertion that they be considered precursors or ‘energizers’ (Kaye and Anderson, 1999, p. 489), which need to be implemented prior to a CI development initiative, rather than behaviours that can be implemented alongside other CI behaviours. Developing a greater understanding of the specific CI behaviours and how they affect the overall CI implementation as well as performance will require further analyses of the individual behaviours within these groups.

Contributions and Directions for Future Research The development of our understanding of innovation processes such as those of developing CI capability seems to follow a similar pattern as we have witnessed in the innovation literature. The first models presented innovation as oversimplified (Pierce and Delbecq, 1977), neatly structured, stage-wise sequences of, for example, activities or decisions (Saren, 1984). The recognition that, as Schroeder et al. (1986) put it, ‘… an appreciation of the temporal sequence of activities that occur in developing and implementing new ideas is fundamental to the management of innovation’, led to the launch of a range of new research initiatives and, eventually, a whole range of much more accurate innovation process models (e.g. Boer and During, 2001; Van de Ven et al., 1989). Areas such as decision-

Volume 15

Number 4

2006

336


making theory (e.g. Cohen et al., 1972; Mintzberg et al., 1976), organizational change theory (e.g. Pettigrew, 1990), and new product development theory (e.g. Buijs, 2003) essentially went through the same process. Continuous Improvement process theory is only at the start of the same development process. The present article contributes to both the academic and managerial literature by taking the first steps in the development of a model of how CI progresses in an empirical context. The findings from this research, namely that CI development need not – and perhaps should not – progress in a linear fashion, emphasizes the limitations of an essentially linear model of CI development when planning CI initiatives. It is, however, far too premature to assume that companies should alter their strategies to accommodate the findings presented in this article. The appropriate strategy for any given organization may depend on factors not included in this study. Future research should be targeted at identifying additional factors (i.e. in addition to size, type of process, and rated importance of CI) that may have influence on which behaviours should be prioritized, in order to provide a more realistic picture of how companies can position their CI development, relative to their own situations.

References Bessant, J. and Caffyn, S. (1997) High involvement innovation, International Journal of Technology Management, 14(1), 7–28. Bessant, J. and Francis, D. (1999) Developing strategic continuous improvement capability, International Journal of Operations and Production Management, 19(11), 1106–19. Boer, H. and During, W.E. (2001) Innovation. What innovation? A comparison between product, process and organisational innovation, International Journal of Technology Management, 22(1–3), 83–107. Boer, H., Berger, A., Chapman, R. and Gertsen, F. (eds) (2000) CI Changes. From Suggestion Box to Organisational Learning, Ashgate, Aldershot.

Volume 15

Number 4

2006

Buijs, J. (2003) Modelling product innovation processes, from linear logic to circular chaos, Creativity and Innovation Management, 12(2), 76–93. Caffyn, S. (1999) Development of a continuous improvement self-assessment tool, International Journal of Operations & Production Management, 19(11), 1138–53. Cohen, M.D., March, J.G. and Olsen, J.P. (1972) A garbage can model of organizational choice, Administrative Science Quarterly, 17(1), 1–25. Jørgensen, F. (2003) ‘A journey through self-assessment, learning, and Continuous Improvement’, PhD dissertation, Uni.Print: Aalborg University. Kaye, M. and anderson, R. (1999) Continuous improvement: the ten essential criteria, International Journal of Quality & Reliability Management, 16(5), 485–506. Kobayashi, I. (1995) 20 Keys to Workplace Improvement, Productivity Press, Portland. Mintzberg, H., Raisinghani, D. and Theorêt, A. (1976) The structure of unstructured decision processes, Administrative Science Quarterly, 21, 246–75. Pettigrew, A.M. (1990) Longitudinal field on research on change: theory and practice, Organization Science, 1(3), 267–92. Pierce, J.L. and Delbecq, A.L. (1977) Organization structure, individual attitudes and innovation, Academy of Management Review, 2(1), 27–37. Rijnders, S. (2002) ‘Four Routes to Continuous Improvement: An empirical typology of CI implementation processes’, PhD dissertation, Twente University Press. Saren, M.A. (1984) A classification and review of models of the intra-firm innovation process, R&D Management, 14(1), 11–24. Savolainen, T. (1999) Cycles of continuous improvement: Realizing competitive advantages through quality, International Journal of Operations & Production Management, 19(11), 1203–22. Schroeder, R., Van De Ven, A.H., Scudder, G. and Polley, D. (1986) Managing innovation and change processes: findings from the Minnesota Innovation Research Program, Agribusiness, 2(4), 501–23. Van de Ven, A.H., Angle, H.L. and Poole, M.S. (1989) Research on the Management of Innovation: The Minnesota Studies, New York, Oxford University Press.



337

Frances Jørgensen (frances@production. aau.dk) is Associate Professor of Strategic Human Resource Management at the Center for Industrial Production at Aalborg University (Denmark). After obtaining a BSc in Experimental Psychology and an MA in Industrial and Organizational Psychology in the US, Frances was employed as an organizational psychologist working with diverse HRM projects, including team development and team building, management training and development, implementation of performance management systems, and change management. In 2001, she received a PhD in Organizational Change Management and has subsequently (co-)authored numerous articles on Continuous Improvement, Self-Assessment of Continuous Improvement, and change management. Harry Boer ([email protected]) is Professor of Organizational Design and Change at the Center for Industrial Production at Aalborg University (Denmark) and the Department of Business Administration at Stavanger University (Norway). He holds a BSc in Applied Mathematics and an MSc and PhD both in Management Engineering. He has (co-)authored numerous articles and several books on subjects such as Organization Theory, Flexible Automation, Manufacturing Strategy, and New Product Development and Continuous Improvement and Innovation. His research focuses on Continuous Innovation or, more precisely, the interaction between day-today operations, incremental change and radical innovation. Bjørge Timenes Laugen (bjorge.laugen@ uis.no) is an Associate Professor at the Department of Business Administration at Stavanger University (Norway). He received his MSc in Engineering in 2000, and his PhD in Innovation Management in 2006, both from Aalborg University, Denmark. His main research interest is the link between new product development, production, strategy, organizational development and continuous innovation. Dr Laugen is a board member of CINet (Continuous Innovation Network), a global network set up to bring together researchers and industrialists working in the area of Continuous Innovation.


Volume 15

Number 4

2006

338


Continuous Improvement and Collaborative Improvement: Similarities and Differences Rick Middel, Harry Boer and Olaf Fisscher A substantial body of theoretical and practical knowledge has been developed on continuous improvement. However, there is still a considerable lack of empirically grounded contributions and theories on collaborative improvement, that is, continuous improvement in an interorganizational setting. The CO-IMPROVE project investigated whether and how the concept of continuous improvement can be extended and transferred to such settings. The objective of this article is to evaluate the CO-IMPROVE research findings in view of existing theories on continuous innovation. The article investigates the similarities and differences between key components of continuous and collaborative improvement by assessing what is specific for continuous improvement, what for collaborative improvement, and where the two areas of application meet and overlap. The main conclusions are that there are many more similarities between continuous and collaborative improvement. The main differences relate to the role of hierarchy/market, trust, power and commitment to collaboration, all of which are related to differences between the settings in which continuous and collaborative improvement unfold.

Introduction

C

ontinuous improvement is the ‘planned, organised and systematic process of ongoing, incremental and company-wide change of existing work practices aimed at improving company performance’ (Boer et al., 2000). Continuous improvement has its earliest roots dating back to even before the Industrial Revolution started and scientific management was developed (Boer and Gieskes, 1999). The export of the concept from the United States to Japan and its development there, the influence of many other concepts, such as Quality Circles, Total Quality Management and Lean Production, and specific research on continuous improvement (Imai, 1986; Robinson, 1991; Bessant and Caffyn, 1997; De Lange-Ros, 1999) resulted in the development, exchange and dissemination of practically and theoretically relevant knowledge on the concept of continuous improvement (see, for example, Imai, 1986; Bessant and Caffyn, 1997, Boer et al., 2000). Today, continuous improvement is a consolidated concept in managerial theory and practice and widely regarded as vital in today’s business environments.

Volume 15

Number 4

doi:10.1111/j.1467-8691.2006.00407.x

2006

However, as competition is increasingly moving from the level of individual firms to that of networks of organizations, companies have to increasingly link their internal processes with external customers and suppliers (Ford et al., 2003). This includes not only operational processes like new product development and production, but also continuous improvement (Boer et al., 2000; Rijnders, 2002). In capturing the ‘state of the concept’, Boer et al. (2000) concluded that continuous improvement should no longer be restricted to intra-firm processes but increasingly applies to inter-firm processes as well. This idea provided the start for a three-year EU-funded research project CO-IMPROVE (Collaborative Improvement Tool for the Extended Manufacturing Enterprise, G1RD – CT2000 – 00299). The overall purpose of the project was to develop a tool supporting the implementation and operation of collaborative improvement within the extended manufacturing enterprise (EME). Collaborative improvement was defined as ‘a purposeful inter-company interactive process that focuses on continuous incremental innovation aimed at enhancing the EME overall performance’ © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

CONTINUOUS IMPROVEMENT AND COLLABORATIVE IMPROVEMENT

(Cagliano et al., 2005). There is still a considerable lack of empirically grounded theory on inter-organizational continuous improvement (Chapman and Corso, 2005). Consequently, it is a major challenge for practitioners as well as researchers to gain insight into and develop an understanding of the organization and management of the process of collaborative improvement (Boer and Gertsen, 2003). In COIMPROVE, an action research approach was adopted to address industrial improvement needs while creating knowledge and in-depth understanding of the process itself at the same time (Middel et al., 2006). The objective of this article is to evaluate the research findings in view of existing theories on continuous improvement. The research question addressed in this article is: RQ: What are the similarities and differences between continuous improvement and collaborative improvement? Thus, the article presents an attempt to discover what is specific for continuous improvement, what for collaborative improvement, and where theories on the two areas of application meet and overlap. This article draws on the continuous improvement and, emerging, collaborative improvement literature, and empirical findings obtained from a research setting involving an extended manufacturing enterprise consisting of a system integrator in the automotive industry and three of its suppliers in the Netherlands. The system integrator specializes in ‘motion control’ systems for the automotive, truck, marine, medical and agricultural markets. Supplier 1 is an assembly company of plastic precision parts for the automotive, medical and pharmaceutical industry. Supplier 2 produces fine mechanical parts for the high-tech industry. Supplier 3 is a producer of cylinder tubes for the automotive industry.

Continuous Improvement and Collaborative Improvement Vision and Commitment One of the prerequisites for continuous improvement success is a clear and agreed strategic framework including long-term and short-term targets and milestones (Imai, 1986, Bessant et al., 1994, Caffyn, 1998; Gieskes et al., 1999). The CI Maturity Model (Bessant et al., 1994; Bessant and Caffyn, 1997) describes continuous improvement in terms of key behaviours, which are essential for long-term success with continuous improvement. Two of the key behaviours are awareness and under© 2006 The Authors Journal compilation © 2006 Blackwell Publishing

339

standing of the organization’s aims and objectives, and the use of the organization’s strategic goals and objectives to focus and prioritize improvement activities (Bessant et al., 1994; Bessant and Caffyn, 1997, Caffyn, 1999). Further strategy-related components of continuous improvement are top management commitment and a long-term, company-wide perspective (Caffyn, 1998). For a company in the automotive industry today, the main challenge is to constantly monitor and adjust its cost structure to the standard in the industry. Continuous improvement and cost reduction are integrated and explicit in the system integrator’s policy and practices. The aim is to establish close cooperation and long-term agreements with a limited number of suppliers. As such, the system integrator looks for highly involved and dedicated partners that fully support the company in assembling and delivering to customers systems of top quality at agreed competitive prices by the promised delivery date. However, in a collaborative improvement context, companies should be motivated not only to pursuing their own goals, but also to improving the performance of the whole extended manufacturing enterprise through collaboration within the network. Thus, a shared and mutually understood vision is an important prerequisite for participating companies to fully exploit the opportunities within the collaboration. Before the start of CO-IMPROVE, improvement activities did take place in the extended manufacturing enterprise, but the companies lacked a shared vision on collaborative improvement and a sense of how to develop the concept. A case study, performed as part of CO-IMPROVE (see Middel et al., 2005a), indicated that, in fact, many improvement activities were ad hoc, problem-driven improvement projects, rather than collaborative, structural and proactive improvement processes. The activities were focused on product and process problems and driven by a performance measurement tool the system integrator used to asses its suppliers with regard to cost, quality and delivery performance. The suppliers suspected the collaborative improvement approach was a new way of imposing cost reduction and higher quality and delivery requirements. In order to resolve this, the system integrator put a lot of effort into introducing and explaining the concept and benefits of collaborative improvement. They convened a workshop, presented their vision on the CO-IMPROVE project, and invited the suppliers to reflect and comment on this, and to present their own goals and aims (and also started to act in a more directive fashion – see

Volume 15

Number 4

2006

340


‘Implementation Roles’ below). The discussion resulted in a common understanding of, and a shared vision on, collaborative improvement. These, in turn, enabled the partners jointly to focus, prioritize and initiate improvement activities, and disseminate knowledge, experiences and lessons learned as part of the collaborative improvement initiative (Kaltoft et al., 2003; Middel et al., 2004). Genuinely sharing a vision and, perhaps even more importantly, working together towards common goals, requires commitment. Each of the partners involved needs to facilitate and support the collaborative improvement process. Without that, projects will be stopped or performed poorly, learning possibilities will not be fully explored and exploited in the extended manufacturing enterprise, and new projects will not even be started. So, one of the prerequisites for both continuous improvement and collaborative improvement is a clear strategic framework. The main difference between continuous improvement and collaborative improvement concerns the ‘unit of analysis’ – the firm in the case of continuous improvement; the partnership in the case of collaborative improvement. Genuine commitment to that vision and, through that, the improvement process is another prerequisite for success. However, within a collaborative improvement setting, companies must also be externally committed to the relationships with their partners in the extended manufacturing enterprise.

Short-term Orientation versus Long-term Optimization If a company designs, organizes and implements its improvement activities to focus solely on short-term improvements, an effective continuous improvement system will never take root within the organization (Bessant, 1998). Organizations should balance top-down (planned, strategy-driven) and bottom-up (emergent, contributing to strategy development) improvement activities, as well as a short-term and long-term orientation. In CO-IMPROVE, an action learning approach was used and implemented through a cycle of workshops and face-to-face meetings. The aim of the workshops was to engage the companies in collaborative improvement activities, including diagnosing and factfinding, and implementation and evaluation of improvement actions. Moreover, the process of action learning emphasized the importance of a structured questioning and reflective process (Middel et al., 2005b). As described in the previous subsection, improvement initiatives had been based on

Volume 15

Number 4

2006

supplier assessments conducted by the system integrator. The initiatives had all the characteristics of a planned and systematic process with a short-term orientation aimed at solving immediate operational, that is, cost, quality and delivery, problems. However, when the partners began developing a common understanding of the concept and benefits of collaborative improvement, learning from their experiences, they also started considering improvement activities not only as a response to practical problems but also as a way to handle emerging, ‘creative’ improvement activities and as opportunities to develop a closer and long-term relationship with the system integrator. As a result, communication and knowledge/information exchange became more open, and the partners became increasingly better in balancing short-term, planned and assessment-driven improvement initiatives with long-term, emerging, creative and relational improvement initiatives. We conclude that a formal problem solving cycle is required for collaborative improvement, just as for continuous improvement, in order for an extended manufacturing enterprise to be able to fully exploit the improvement potential. Within the improvement cycle specific attention has to be paid to capturing and disseminating learning within and between the partners in order to stimulate and trigger reactive solutions and creative opportunities. Therefore, an effective and open mode of communication needs to be developed, both intra-organizationally as well as inter-organizationally, in order to share and transfer information, learning and knowledge between and within the partners. Another similarity is that companies need to find and manage a balance between a short-term and a long-term improvement perspective as well as between planned (top-down) and emerging (bottom-up) improvement initiatives.

The Belief in the Value of Small Improvements The successful implementation of continuous improvement requires (the development of) an underlying belief/assumption system that contains the core continuous improvement values (Bessant et al., 1995; Bessant and Caffyn, 1997; De Lange-Ros, 1999). According to Bessant et al. (1995), believing in the value of small step incremental innovation is a core enabler of continuous improvement. As explained previously, the suppliers initially hesitated to fully collaborate in the collaborative improvement process. In order to overcome this and increase their belief in the concept of collaborative improvement, in addition to trying to develop a shared vision and © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


goals, the system integrator promoted improvement initiatives that could be solved fairly easily and yield both operational and learning outcomes. Furthermore, it was considered important that the results of these projects should be shared in order to entice the partners to continue their collaboration. Finally, the companies would frequently present, discuss and reflect upon the different improvement projects. In effect, they started to learn from and with each other, which positively affected their belief in collaborative improvement. We conclude that believing in the value of small improvements is just as important in collaborative improvement as it is in continuous improvement. If resourced, organized and managed properly, continuous improvement helps companies to improve the performance of their internal processes and to learn and build continuous improvement competences. Collaborative improvement allows companies to improve the performance of intra- and interorganizational processes, while avoiding local optimization, to maintain and develop relationships, and to learn and build collaborative improvement competences.

Trust Specific research on the role of trust in continuous improvement programmes has not been reported, although we have found some indirect references to this factor. In Japan, for example, life-time employment was used to create trust among employees that they would not lose their job because of improvement initiatives. However, from the CO-IMPROVE research trust emerged as an important factor affecting collaborative improvement (Kaltoft et al., 2003). Trust can be defined as the belief of a firm that its partner in a relationship will act in the firm’s best interest in circumstances where that partner could take advantage or act opportunistically to gain at the firm’s expense (McCutcheon and Stuart, 2000). Trust in collaborative improvement processes can be an enabler, whereas a lack of trust can be a disabling factor in the process of collaborative improvement. Trust between organizations is built upon trust between individuals. Trust comes in various forms, namely cognitive trust (based on past experiences), affective trust (based on emotional bonds between individuals) and calculative trust (based on future perceptions or expectations) (McAllister, 1995; Morrow et al., 2004; Vieira, 2005). In our case study, all three forms of trust were present. The relationship between the system integrator and supplier 1 was based on cognitive trust, built on several collaborative © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

341

improvement initiatives successfully conducted in the past. The relationship between the system integrator and supplier 2 can be classified as affective trust. The sales representative had a very good operational and personal relationship with the purchaser of the system integrator. However, when the purchaser left the company, the trust of the supplier towards the system integrator was affected negatively in terms of their openness and willingness to share information. This shows that trust between companies is built upon the relationship between individuals, and also that trust is an enabling factor while lack of trust is a disabler. The relationship between supplier 3 and the system integrator was fairly new as business between the two companies was about to start up. In this respect, the two companies ‘trusted’ each other for the time being in order to develop business, which could be beneficial for both of them. This can be classified as calculative trust. So, organizational trust is built upon trust between people working together. Trust may take different forms, but whatever form dominates, it is a prerequisite for and enabler of collaborative improvement processes. Lack of trust has a negative effect on the collaborative improvement process.

Power and Decision Making A flattened hierarchy and decentralization of decision making have been put forward as important components of the continuous improvement process (Caffyn, 1998). The hierarchy within an organization strongly affects the decision-making power people have, as at least part of an individual’s power is embedded in the position of that individual in the organization (Fehse, 2002). Power also has a relational aspect, as it is enacted to influence the behaviour of others and/or to attain certain goals (Frost, 1987). The main difference between continuous improvement and collaborative improvement is that there are no hierarchical relationships between the partners in a collaborative improvement process. Power is not based on hierarchical position but on mutual dependency, and balancing of power in an extended manufacturing enterprise actually happens in ‘the marketplace’. In the Dutch extended manufacturing enterprise, the suppliers depended much more on the system integrator for their sales than the system integrator depended on the suppliers. The purchasing manager of the system integrator knew he could use this position to influence his partners so as to attain certain goals, but he deliberately did not do so, believing that this would influence the collab-

Volume 15

Number 4

2006

342


orative improvement process negatively. He preferred working on the premise that the decision-making process was a responsibility of the extended manufacturing enterprise and the collaborative improvement project team in which all the participants should have the opportunity to present, discuss and decide on the collaborative improvement projects, process and objectives. Before the CO-IMPROVE project, this was different. Based on regular supplier assessments, the system integrator selected the improvement projects and developed a specific improvement programme for each supplier. So, the system integrator did have and exercised the decision-making power in this process. At the beginning of the COIMPROVE project, this led the suppliers to continue acting reactively towards the system integrator and stay somewhat closed in their communication and information exchange. They perceived the CO-IMPROVE project as another way of imposing improvement initiatives. This affected the start-up phase negatively – there was no sense of urgency and a lack of activity. Only when the partners had developed a joint vision and the suppliers realized that the system integrator would not use its purchasing power, and that they were equal partners in the decision-making process, did the situation change gradually. We can conclude that decentralized decision making is as important for collaborative improvement as it is for continuous improvement. However, there is no formal hierarchy within collaborative improvement processes – market mechanisms such as purchasing power, rather than formal position, determine the relative power of the partners in an extended manufacturing enterprise. The Dutch system integrator had more power than its suppliers, but did not use its position – they rather relied on joint decision making.

Capabilities and Behaviours In the 1990s, the CIRCA team at Brighton University developed the CI Maturity Model based on practical, action-oriented research with a set of manufacturing companies (Caffyn, 1999). This model describes a set of behavioural routines which appear to be essential for long-term success with continuous improvement, and how these behaviours (should) develop over time (Bessant and Caffyn, 1997). Individuals and groups display the behaviours, and they are closely linked to a set of core continuous improvement abilities (Caffyn, 1999). Translated to a collaborative improvement setting, these abilities are as follows:

Volume 15

Number 4

2006

1. The partners in the extended manufacturing enterprise are guided by shared improvement and collaboration values. 2. Companies within the extended manufacturing enterprise use long-term goals and objectives to focus, prioritize and organize improvement activities. 3. People/companies within the extended manufacturing enterprise participate proactively in incremental improvement and their learning is captured and deployed. 4. People/companies within the extended manufacturing enterprise participate in implementing and facilitating improvement projects across the boundaries of the inter-company operations. 5. People/companies within the extended manufacturing enterprise constantly evaluate improvement projects and ensure that the outcomes are used to improve and monitor the collaborative improvement system at an extended manufacturing enterprise level. This set provided the basis for an assessment tool we used to measure the collaborative improvement maturity of the collaboration and for intervention purposes at the same time. The interventions were expected to affect the level of maturity to the extent that they would trigger a dialogue between the people involved, and help them develop a deeper understanding of the fundamental principles of collaborative improvement and an increased motivation to participate in subsequent improvement activities. At the start of the CO-IMPROVE project, the Dutch companies had particularly low scores on capturing and deploying evaluation-based learning from improvement projects and the overall collaborative improvement process. Before the research project started, the companies prioritized operational issues, rather than learning. The situation gradually improved over time due to the action learning approach. Especially at the workshops, explicit attention was given to reflecting on the collaborative improvement process, identifying experiences and lessons learned, and presenting them in plenum. Thus, the companies within the extended manufacturing enterprise became increasingly aware of the concept and benefits of collaborative improvement, recognized the importance of a structured process towards improvement and learning, and developed a setting for reflection and evaluation with a high degree of openness and trust. So, over time, the companies matured with regard to the collaborative improvement process, which was reflected in an increase in the score of the collaborative improvement maturity assessment on all five behaviours. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


We conclude that just as with continuous improvement, a set of behaviours appear to be important in collaborative improvement processes as well. If improvement efforts are successful, the maturity level of the collaboration will increase over time, provided that due attention is paid to developing behaviour 3 (learning) and behaviour 5 (evaluation). However, neither of these behaviours is easy to develop and sustain – it is much more convenient to focus on short-term operational issues than on long-term learning.

Implementation Roles The success of continuous improvement depends on the contribution and commitment of individuals and teams involved (Caffyn, 1997; Boer et al., 2000). Each actor plays one or more roles as part of the process. One of the important roles mentioned by Caffyn (1997) is the continuous improvement facilitator. Within the CO-IMPROVE project various roles were identified, which appeared to influence the progress and process of the collaborative improvement. Roles particularly important in the Dutch setting were: • Initiator: Starts activities, generates discussions and encourages participation. • Facilitator: Facilitates communication, moderates discussion, and encourages interaction and reflection. • Expert: Provides information, evaluates feasibility and anticipates constraints. • Gatekeeper: Provides contacts, and identifies and liaises with key sources of information. • Problem solver: Does the work, participates in activities and discussions, and reflects on experience and progress. As mentioned before, the initial approach adopted by the Dutch extended manufacturing enterprise did not produce much continuous improvement activity and learning. The system integrator therefore decided to change approach and become more active and directive, by starting activities, generating discussion and encouraging participation of all companies. This worked out well. More improvement initiatives were identified and selected, more collaborative improvement meetings took place, and all companies participated more actively (see also Kaltoft et al., 2004). The system integrator’s purchasers fulfilled the role of initiator by starting specific improvement initiatives with each of the suppliers. The purchasing manager fulfilled the roles of initiator and facilitator by creating momentum and speeding up the collaborative improvement initiatives. All individuals © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

343

involved in the improvement projects acted as problem solvers in the sense that they worked on the project, participated in discussions and reflected on the progress of the project and their own experiences. Over time, several other improvement projects were started in which not only the system integrator fulfilled the initiator role but also the suppliers. In a specific example, the sales engineer of one of the suppliers identified, during a factory tour, an opportunity for improving an existing product of the system integrator. He took the lead, wrote a proposal, calculated the expected operational outcomes and started to encourage the system integrator to participate in this project. As he proceeded, he contacted different people within his own company and the system integrator in order to receive all required information and discuss the feasibility and constraints of the project. As such, he fulfilled the roles of gatekeeper and expert. The latter role was also fulfilled by the purchasing manager of the system integrator, as he evaluated the feasibility and anticipated the constraints of the project for his own organization. If we compare our findings with the literature on continuous improvement, we conclude that the facilitator plays an important role in both continuous improvement and collaborative improvement. Our research points towards the importance of other roles as well, namely the initiator, expert, gatekeeper and problem solver roles. Although none of these roles are mentioned in the continuous improvement literature, they are in the more general innovation literature. Boer and During (2001), whose work is partly based on Roberts and Fusfeld (1981) (see also Schon, 1963; Frohman, 1978; Maidique, 1980; Galbraith, 1982) mention: idea generator (= initiator), scout (= expert), gatekeeper and problem solver. So, they are likely to be of importance in continuous improvement processes as well. Finally, we essentially confirmed Roberts and Fusfeld’s (1981) findings, namely that: • some roles can be fulfilled by more than one person in order for the project to be successful • individuals/companies may occasionally fulfil more than one critical function • the roles that individuals/companies play may change over time.

Contributions Contribution to Theory As companies increasingly link their internal processes with external customers and suppli-

Volume 15

Number 4

2006

344


ers, new approaches must be developed to enhance and continuously improve the business performance not only of the partners involved but also of the collaboration as a whole (Kaltoft et al., 2003). Therefore the concept of continuous improvement must be developed for, and applied in, inter-organizational settings as well (Boer et al., 2000; Rijnders, 2002), in the form of collaborative improvement. In this article we report research aimed at identifying similarities and differences between continuous improvement and collaborative improvement. A system integrator in the automotive industry with three suppliers provided the empirical setting for this research. Table 1 summarizes the findings presented and discussed in this article. There are a lot of similarities but also some differences between continuous improvement and collaborative improvement. Improvement, whether it takes place within an organization or is undertaken jointly by different

organizations, is a human activity first of all. It requires vision, sense of direction, commitment, understanding, communication, development, evaluation, learning, planning and creativity, resources, organization, management. In that sense there are no differences between continuous improvement and collaborative improvement. The differences we found are all related to differences in setting. Continuous improvement takes place within a hierarchy, collaborative improvement in ‘the marketplace’. Continuous improvement prospers in a flattened hierarchy (Caffyn, 1998) in which individuals and teams are empowered to make and carry through improvement decisions. In the strict sense of the word, hierarchy is not a construct that can appropriately be used to describe the interaction between firms. Power can. But an individual’s positional power is not the same as an organization’s purchasing power.

Table 1. Similarities and Differences between Continuous Improvement and Collaborative Improvement Plays a role in continuous improvement and collaborative improvement:

Plays a role in continuous improvement, not collaborative improvement: Plays a role in collaborative improvement, not continuous improvement

• Clear strategic framework/vision and goals, shared between all participants • Communication and understanding of strategy to all participants • Intra-organizational commitment • Commitment to training and personnel development • Balance between long-term and short-term improvement perspectives • Formal problem solving cycle • Capturing and transfer of learning • Effective and open communication and information sharing • Balance between top-down and bottom-up improvement initiatives • Shared belief in the benefits of small step improvements • Empowerment/decentralization of decision making • The facilitator role • The initiator, expert, gatekeeper and problem solver roles1 • Wide range of tools that can be used and applied in the process • Flattened hierarchy • Position-based power • Benefit sharing • Power based on market-related dependencies, including purchasing power • Trust (between organizations) • Commitment to collaboration

Note 1 Not mentioned in the continuous improvement literature but very likely to play a role in intra-firm improvement processes as well as in collaborative improvement processes.

Volume 15

Number 4

2006



Benefit sharing is a key factor in collaborative improvement, but does not logically play a role in continuous improvement as there is only one beneficiary in that case, the organization. Sharing of benefits has a positive influence on commitment to and trust in the partners, two other variables which do not play a major role in continuous improvement, but play a key role in affecting the behaviour of collaborative organizations towards each other. In fact, it seems that trust and commitment (to the collaboration) are what a flat hierarchy with empowered individuals/teams is in an organization – they shape the environment in which improvement can prosper. Two defining features that are not specifically discussed in the continuous improvement and the emerging collaborative improvement literature, but which are very important in the relationships between individuals and companies, are time and dynamics. Both the experienced past, the perceived present and the anticipated future affect the behaviour in a relationship (see also Ford et al., 2003), through the development of trust, continuous improvement/collaborative improvement capabilities, and commitment. However, none of these factors, or the other factors listed above, are stable – they change over time, affecting each other in a continuous process with ups and downs, a process we are only beginning to understand.

Further Research The results reported here seem to make logical sense. However, there are some weaknesses related to method. We performed action research involving an extended manufacturing enterprise active in the automotive industry and located in the Netherlands. Depth together with longitudinality is an advantage, the narrow empirical basis a weakness. Research of collaborative improvement in extended manufacturing enterprises in other countries and/or industries, or representing different countries (cultures) and/or time zones (temporal separation) may produce additional factors and different similarities and differences between continuous improvement and collaborative improvement and thus add to and/or refine our findings. Furthermore, we focused our research on similarities and differences between continuous improvement and collaborative improvement. An area we left unexplored is the interaction between the two. Continuous improvement requires capabilities which can and should mature over time and so does collaborative improvement. However, does a specific maturity in continuous improvement © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

345

positively influence the collaborative improvement maturity? In other words, do companies with a high continuous improvement maturity perform better in, and achieve a higher level of, maturity with regard to collaborative improvement more easily than companies with a lower continuous improvement maturity? Is there a relationship between continuous improvement and the collaborative improvement maturity, and what are its effects?

Contribution to Management In spite of the weaknesses indicated above and the many questions for further research, there are a few lessons managers and other practitioners involved in collaborative improvement efforts can learn from the research presented here. One key lesson is that collaborative improvement resembles continuous improvement a great deal. Experiences with continuous improvement, which many companies have today, and the ‘hints’ collected in the upper part of Table 1 are very useful for managers engaging on behalf of their company in a collaborative improvement process together with their customers and/or suppliers. A second key lesson, however, is that collaborative improvement is not the same as continuous improvement. Positional power cannot be used, and purchasing power need not be used. Without trust and commitment, nothing substantial will happen. And benefit sharing is a must. The problem is that none of the partners can enforce any of these factors on the other partners. Leading by example, like the system integrator in the Dutch case, may provide the optimum route to success.

References Bessant, J. (1998) Developing continuous improvement capability, International Journal of Innovation Management, 2, 409–29. Bessant, J. and Caffyn, S. (1997) High-involvement innovation through continuous improvement, International Journal of Technology Management, 14(1), 7–28. Bessant, J., Caffyn, S. and Gilbert, J. (1994) ‘Mobilising continuous improvement for strategic advantage’, in Platts, K.W., Gregory, M.J. and Neely, A. (eds), Operations Strategy and Performance, Manufacturing Engineering Group, University of Cambridge, Cambridge, pp. 175–80. Bessant, J., Caffyn, S. and Gilbert, J. (1995) ‘Continuous improvement: the European dimension’, in Draaijer, D., Boer, H. and Krabbendam, K. (eds), Management and New Production Systems, Proceedings of the 2nd International Conference of the European Operations Management Association, Enschede, University of Twente, pp. 31–40.

Volume 15

Number 4

2006

346


Boer, H. and During, W.E. (2001) Innovation. What innovation? A comparison between product, process and organisational innovation, International Journal of Technology Management, 22(1–3), 83–107. Boer, H. and Gertsen, F. (2003) From continuous improvement to continuous innovation, a (retro)(per)spective, International Journal of Technology Management, 26(8), 805–27. Boer, H. and Gieskes, J.F.B. (1999) Editorial: ‘Continuous improvement: from idea to reality’, International Journal of Operations & Production Management, 19(11), 1102–5. Boer, H., Berger, A., Chapman, R. and Gertsen, F. (2000) CI Changes: From Suggestion Box to Organisational Learning, Continuous Improvement in Europe and Australia, Ashgate, Aldershot. Caffyn, S. (1997), Extending continuous improvement to the new product development process, R&D Management, 27(3), 253–67. Caffyn, S.J. (1998) ‘The scope for application of continuous improvement to the process of new product development’, PhD Thesis, University of Brighton. Caffyn, S. (1999) Development of a continuous improvement self-assessment tool, International Journal of Operations & Production Management, 19(11), 1138–53. Cagliano, R., Caniato, F., Corso, M. and Spina, G. (2005) Implementing collaborative improvement: lessons from an action research process, International Journal of Production Planning & Control, 16(4), 345–55. Chapman, R.L. and Corso, M. (2005) Introductory paper: From continuous improvement to collaborative innovation: the next challenge in supply chain management, International Journal of Production Planning & Control, 16(4), 339–44. De Lange-Ros, D.J. (1999) ‘Continuous improvement in teams. The (mis)fit between improvement and operational activities of improvement teams’, PhD Thesis, Enschede, Print Partner Ipskamp. Fehse, K.I.A. (2002) ‘The role of organizational politics in the implementation of information systems, three cases in a hospital context’, PhD Thesis, Enschede, Print Partners Ipskamp. Ford, D., Gadde, L., Håkansson, H. and Snehota, I. (2003) Managing Business Relationships, John Wiley & Sons, Chichester. Frohman, A.L. (1978) The performance of innovation: managerial roles, California Management Review, 20(3), 5–12. Frost, P.J. (1987) ‘Power, politics and influence’, in Jablin, F.M., Putnam, L.L., Roberts, K.H. and Porter, L.W. (eds), Handbook of Organizational Communication: An Interdisciplinary Perspective, SAGA Publications, California. Galbraith, J.R. (1982) Designing the innovating organization, Organizational Dynamics, Winter, 5– 25. Gieskes, J.F.B., Boer, H., Baudet, F.C.M. & Seferis, K. (1999) CI and performance: a CUTE approach, International Journal of Operations & Production Management, 19(11), 1120–37. Imai, M. (1986) Kaizen. The Key to Japan’s Competitive Success, McGraw-Hill, London. Kaltoft, R., Boer, H., Corso, M., Gertsen, F., Gieskes, J.F.B., Middel, H.G.A. and Nielsen, J.S. (2003)

Volume 15

Number 4

2006

‘Factors affecting the development of collaborative improvement with strategic suppliers’, in Spina, G. et al. (eds), Proceedings of the EUROMA and POMS Conference 2003 – The Challenges of Integrating Research and Practice, Vol. 3, pp. 601–10. Kaltoft, R., Boer, H., Caniato, F., Gertsen, F., Middel, H.G.A. and Nielsen, J.S. (2004) ‘Implementing collaborative improvement – one approach fits all?’ in van Wassenhove, L.N. et al. (eds), Proceedings of the 11th Annual International EUROMA Conference, Operations Management as a Change Agent, INSEAD Business School, Fontainebleau, France, Vol. I, pp. 333–42. McAllister, D.J. (1995) Affect- and cognition-based trust as foundations for interpersonal cooperation in organizations, Academy of Management Journal, 38(1), 24–43. McCutcheon, D. and Stuart, F.I. (2000) Issues in the choice of supplier alliance partners, Journal of Operations Management, 18, 279–301. Maidique, M.A. (1980) Entrepreneurs, champions, and technological innovation, Sloan Management Review, 21 (Winter), 59–76. Middel, H.G.A., Fisscher, O.A.M. and Groen, A.J. (2004) Managing and organising collaborative improvement: A system integrator perspective, Proceedings of the 5th International CINet Conference, University of Western Sydney, Sydney (on CD-Rom). Middel, H.G.A., Gieskes, J.F.B. and Fisscher, O.A.M. (2005a) Driving collaborative improvement processes, Production Planning and Control, 16(4), 368–77. Middel, H.G.A., Brennan, L., Coghlan, D. and Coughlan, P. (2005b) ‘The application of action learning and action research in collaborative improvement within the extended manufacturing enterprise’, in Kotzab, H., Seuring, S., Müller, M. and Reiner, G., Research Methodologies in Supply Chain Management, Physica Verlag, Heidelberg, pp. 365–80. Middel, R., Coghan, D., Coughan, P., Beennan, L. and McNichols, T. (2006) Action Research in Collaborative Improvement, International Journal of Technology Management, 33(1), 67–91. Morrow, J.L., Henson, M.H. and Pearson, A.L. (2004) The cognitive and affective antecedents of general trust within cooperative organizations, Journal of Managerial Issues, 16(1), 48–64. Rijnders, S. (2002) ‘Four routes to continuous improvement, An empirical typology of CI implementation processes’, PhD Thesis, Enschede, Twente University Press. Roberts, E.B. and Fusfeld, A.R. (1981) Staffing the innovative technology-based organization, Sloan Management Review, 22(3), 19–34. Robinson, A. (1991) Continuous Improvement in Operations: A Systemic Approach to Waste Reduction, Productivity Press, Cambridge, MA. Schon, D.A. (1963) Champions for radical new inventions, Harvard Business Review, March– April, 77–86. Vieira, L.M. (2005) ‘Trust within global chains’, in Demeter, K. (ed.), Proceedings of the 12th Annual International EUROMA Conference, Operations and Global Competitiveness, Budapest, Hungary, pp. 525–32. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


347

Rick Middel ([email protected]) is a PhD student at the Faculty of Business, Public Administration and Technology at the University of Twente. He holds an MSc in Industrial Engineering and Management at the University of Twente (The Netherlands). His research interest is in Continuous Improvement and Collaborative Improvement. Harry Boer ([email protected]) is Professor of Organizational Design and Change at the Center for Industrial Production at Aalborg University (Denmark) and the Department of Business Administration at Stavanger University (Norway). He holds a BSc in Applied Mathematics and an MSc and PhD both in Management Engineering. He has (co-)authored numerous articles and several books on subjects such as Organization Theory, Flexible Automation, Manufacturing Strategy, and New Product Development and Continuous Improvement and Innovation. His research focuses on Continuous Innovation or, more precisely, the interaction between day-today operations, incremental change and radical innovation. Olaf A.M. Fisscher (o.a.m.fisscher@ utwente.nl) is Professor of Organization Studies and Business Ethics at the University of Twente. He holds a master degree in Industrial Engineering Management. He obtained his PhD in Social Sciences on the management and organization of R&D laboratories. His research is focused on Organizing for Innovation and organizing for Corporate Social Responsibility. He is coeditor of Creativity and Innovation Management.


Volume 15

Number 4

2006

348


Collaborative Improvement – Interplay but not a Game Rasmus Kaltoft, Harry Boer, Ross Chapman, Frank Gertsen and Jacob S. Nielsen Collaboration literally means working together. Collaborative improvement is an extension of continuous improvement and can be defined as a purposeful inter-company interactive process that focuses on continuous incremental innovation aimed at enhancing the collaboration’s overall performance. Developing collaborative improvement is a protracted and difficult process. Previous research has identified a number of factors affecting that process and suggested that it is not so much the individual factors, but rather their interplay that determines the successful development of collaborative improvement. This article reports research aimed at developing a deeper understanding of that interplay. Ten relationships between ten factors are presented and discussed. It appears that vision, approach, trust and commercial reality are the strongest factors. These factors are, however, influenced by, or affect the other factors, notably national culture, partner characteristics and competences, the use of power, individual behaviour and commitment. The way this interplay develops varies from case to case and has great influence on the development of collaborative improvement.

Introduction

I

n order to survive in an increasingly globalized, competitive marketplace, companies today must build and rely upon close relationships with customers and suppliers (Venkatesan, 1992; Quinn and Hilmer, 1994; Quinn, 1999). De Maio and Maggiore (1992) summarize the evolution of customer-supplier relationships in four main stages: vertical integration (1920–1950), use of the market (1960–1970), operational and logistical integration with suppliers (1970–1980), and comakership based on, amongst other things, technological integration (today). In effect, the battlefield of competition is moving from the level of individual firms to that of collaborations between firms (Gomes-Casseres, 1994; Rice and Hoppe, 2001; Busby and Fan, 1993; Stock et al., 2000), including supply chains and networks, and extended manufacturing enterprises (Frohlich and Westbrook, 2001). Consequently, new approaches must be developed not only to enhance the business performance of such collaborations, but also to support continuous performance improvement. The EU-funded project CO-IMPROVE (Collaborative Improvement Tool for the Extended

Volume 15

Number 4

doi:10.1111/j.1467-8691.2006.00405.x

2006

Manufacturing Enterprise) addressed that need, focusing on the learning required to enhance collaborative improvement. The project involved four universities from Denmark, Ireland, Italy and The Netherlands, two software vendors in Greece and Sweden, and three extended manufacturing enterprises consisting of three systems integrators (companies that integrate components provided by suppliers) located in Denmark, Italy and The Netherlands, respectively, and three to four suppliers each, located in these countries and, in the Italian and Dutch cases, in Austria and Germany as well. In total, the project involved eleven dyadic relationships between systems integrators and suppliers. Developing collaborative improvement appears to be a protracted and difficult process. In CO-IMPROVE, we identified a number of factors affecting that process (Cagliano et al., 2002; Kaltoft et al., 2004a, 2004b; Middel et al., 2002; Nielsen et al., 2004). Our tentative conclusion (Boer et al., 2005) was that it is not so much the individual factors, but rather their interplay that determines the successful development of collaborative improvement. This article reports further analysis aimed at developing a deeper understanding of that interplay. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

COLLABORATIVE IMPROVEMENT

349

Research Problem Collaboration literally means working together (Huxham, 1996; Jordan and Michel, 2000). According to Winer and Ray (1994), ‘collaboration is a mutual beneficial and welldefined relationship entered into by two or more organizations to achieve results they are more likely to achieve together than alone’. In the CO-IMPROVE project, we defined the improvement part of a collaborative relationship, or collaborative improvement, as ‘a purposeful inter-company interactive process that focuses on continuous incremental innovation aimed at enhancing the [collaboration’s] overall performance’ (Cagliano et al., 2002; Coughlan et al., 2002; Middel et al., 2002). According to Bessant and Caffyn (1997), the problem with (intra-firm) continuous improvement is that it is a simple and attractive concept at first sight, which actually appears to be difficult to design, implement and develop successfully. Implementing collaborative initiatives across disparate members of supply networks may be even more difficult. Previous research has identified the influence of: • vision (e.g. DiBella and Nevis, 1998) • individual behaviour, including commitment (e.g. Monczka et al., 1998; Mohr and Spekman, 1994; Moore, 1998), political behaviour (e.g. Pfeffer, 1992), and opportunism (e.g. Williamson, 1981) • power (e.g. Cook, 1977; Pfeffer, 1992; Buchanan and Badham, 1999)

• trust, and its counterparts contract and safeguards (e.g. Williamson, 1981; Sako, 1992; Vangen and Huxham, 2003; Kumar, 1996; Child and Faulkner, 1998; McCutcheon and Stuart, 2000) • competence (e.g. Jansen et al., 1999) • partner characteristics such as strategy, structure and size (e.g. Mintzberg, 1983; Daft, 2001) • national culture (Olivari et al., 2002). In the CO-IMPROVE project we confirmed the influence of these factors and identified three additional factors: • approach to the process of developing collaborative improvement • communication • commercial reality. Seven of the factors are endogenous to the process of collaborative improvement and three are exogenous; see Figure 1 (Boer et al., 2005), and Kaltoft et al. (2004a, 2004b), Nielsen et al. (2004) and Boer et al. (2005) for further details on the individual factors. However, on the basis of their analysis, Boer et al. (2005) conclude that none of the factors have decisive influence and suggest that it is the interplay between the factors that determines the development of the collaboration. The open question, however, is exactly how the factors interact. Hence, the research question central to the present article: ‘How does the interplay between key factors in a collaborative improvement process affect the success of the collaboration?’

Culture Approach Communication Partner characteristics Vision Commercial reality

Trust

Competence Individual behaviour

Collaborative improvement progress

Power

Figure 1. Factors Affecting the Development of Collaborative Improvement © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

Volume 15

Number 4

2006

350


Empirical Setting and Research Method

Empirical Findings

The empirical data underpinning the analysis presented in this article was obtained from action research in the Danish extended manufacturing enterprise, comprising a systems integrator and three suppliers and, thus, three of the eleven dyads involved in COIMPROVE. Basic information on the extended manufacturing enterprise is summarized in Figure 2. The three suppliers were represented in the collaborative improvement team by their top managers, two of them owners of the company. Three purchasers and their purchasing manager represented the system integrator. Finally, two PhD students and two senior researchers from Aalborg University, Denmark, were involved in the team, as action researchers. The central methodology in the project was action research (Coughlan and Coghlan, 2002) by the researchers working closely together with the industrial partners, and action learning by the extended manufacturing enterprise. Thus, theoretical insight is mostly derived from naturally occurring data (Marshall and Rossman, 1999). The researchers had many roles. They facilitated, supported, asked the ‘right’ questions, educated, mediated and were also directly involved as participants on the same level as the practitioners (for further details, see Coughlan et al., 2002).

We expressed success in terms of (a) the number of collaborative improvement projects started and completed, and (b) the development of the relationship between the partners involved in a dyad. The first factor can be quantified; the second can only be assessed qualitatively. The participants in CO-IMPROVE Denmark initiated 15 improvement projects, most of which were completed successfully. One project was suspended, pending the implementation of a new Enterprise Resource Planning system by the system integrator. The kind of projects initiated varied from improvement of quality and delivery performance, to more resource demanding projects such as rolling out the systems integrator’s Total Productive Maintenance/Management and lean manufacturing programmes into the supply chain. Dyads 2 and 3 experienced major dips in the process towards a collaborative improvement relationship. After completion of COIMPROVE, the practitioners and academics jointly concluded that dyad 2 ended up at the same level or perhaps even slightly worse compared to the level before CO-IMPROVE. Dyad 3 improved the relationship only marginally. Dyad 1 experienced a steady improvement of the relationship to a much more mature collaboration level. The question is how the interplay of key factors presented previously affected the success of the collaboration in the three dyads. In the next subsection we will address each of the ten relationships depicted in Figure 1.

Success

Product category: Mobil hydraulics Employees: > 7500

SI

D

d1 ya

Supplier 1 Product category: Foundry products Employees: 250

Dy ad

Dya d 2

Product category: Metal parts Employees: 80

3

Product category: Metal parts Employees: 65

Supplier 3

Supplier 2

Figure 2. Overview of the Empirical Setting in the Danish Extended Manufacturing Enterprise

Volume 15

Number 4

2006



The (Role of) Interplay Interplay between Approach and Culture, Vision and Communication The collaborative improvement process experienced a radical change of approach approximately six months into the process (see Kaltoft et al., 2004a). The initial approach focused on the project level. After six months, focus shifted to the relational and extended manufacturing enterprise level. Culture Æ approach: The Danish meeting and decision-making culture is very much based on dialogue, breaking out of meetings into working groups but presenting the results in plenum. The Danish practitioners’ attitude towards academia tends to be ‘we know better, don’t come and tell us how to do what we have done for many years’ and, therefore, when academia interacts with Danish practitioners, improvement suggestions have to be presented with care and well into the intervention process. The trade culture between the companies involved in this project used to be quite closed, but all companies, and particularly the system integrator, had a strong desire to open up the relationships by sharing information previously not shared. It was decided to commence CO-IMPROVE with a workshop approach, involving three dyads working in dyadic groups but presenting in plenum (i.e. at the level of the extended manufacturing enterprise). The initial focus was on improvements at the level of the individual dyads. We labelled this approach ‘bottom-up learningby-doing’, meaning that the extended manufacturing enterprise would initially focus on operational level problems and gradually move into more strategic and conceptual areas. Vision ´ approach: The companies that joined CO-IMPROVE all lacked an understanding of and a vision on collaborative improvement. Looking at the process in retrospect, it was very beneficial that the initial approach was focused on the operational level and tangible improvement projects, which the companies could easily relate to. However, the researchers’ intention slowly to direct focus towards the strategic/ conceptual level of collaborative improvement, thus further developing the operational as well as the relationship level, was not fulfilled due to lack of vision on the practitioners’ side. Six months into the project, the project team decided to change approach to less frequent meetings while at the same time developing improvement projects that involved more participants and pursued greater impact. The © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

351

project team started to act as a steering committee. The workshops no longer involved working with improvement projects but focused more on steering the projects as well as equipping the steering committee with the knowledge and competencies needed to develop the relationship level. The new approach gradually developed the practitioners’ understanding of collaborative improvement and at the very last workshop (at which the topics were relationship vision and evaluation of the process as a whole) it became clear that a joint vision on the relationships had indeed developed. Approach Æ communication (content): Two key factors interact with communication in slightly different ways: approach interacts with the content, whereas individual behaviour interacts with the process, of communication. The communication before CO-IMPROVE mainly concerned fire-fighting. This changed in the course of time. During the workshops, improvement projects with long-term perspectives were identified and implemented within the extended manufacturing enterprise. Consequently, communication started to include long-term aspects, e.g. moving from ‘how can we improve the delivery level of a particular product’ to ‘how can we improve and sustain the overall quality level by implementing kanban’. Having said this, firefighting still absorbed a lot of the communication agenda. After the intervention addressed above, firefighting gradually disappeared from the workshop agendas to be replaced by much more strategic topics such as vision, collaboration, the future of the relationships, and skills and competence development. The communication between the workshops also changed slowly towards these same topics, but due to the organizational set-up (such as the purchaser handling quality problems), operational problem solving, including fire-fighting still dominated day-to-day interactions. Interplay between Communication and Competence Competence Æ communication (content and process): The initial competencies of the participants, as described under the previous heading, were not sufficient to successfully and efficiently set up and handle improvement projects. Consequently, communication focused on basic improvement problems (e.g. quality or delivery improvement) and at times frustration was evident. After the new approach described above had been implemented, the participants slowly increased

Volume 15

Number 4

2006

352


their project management and continuous improvement skills. Towards the end of the CO-IMPROVE project, the partners’ collaborative improvement maturity had clearly increased. This manifested itself in the topics discussed and improved communication between the participants. Interplay between Partner Characteristics and Collaboration Progress Partner characteristics Æ collaboration progress: The characteristics of the COIMPROVE partners did not change during the process and cannot therefore explain any of the progress made. However, what did change was the impact of improvement projects. Initially, improvement projects would typically involve one or two representatives from the supplier, one purchaser from the system integrator and one researcher. Further into the collaboration, especially after executing the new approach, ‘new’ project members were involved in specific improvement projects. These members might not have had any knowledge of the project CO-IMPROVE but exercised indirect influence on the progress of the collaboration. Interplay between Trust and Commercial Reality This section is divided into two parts, one describing dyads 1 and 3 and the other describing dyad 2 since the former two dyads displayed quite similar interplay in terms of trust and commercial reality. For the dyad numbers, see Figure 2. Trust is also interplaying with individual behaviour, which will be described in the following section. Commercial reality Æ trust Dyads 1 and 3: Commercial reality had a positive effect on trust between the participants in dyads 1 and 3. This was due to the system integrator’s increased success in the marketplace and that company’s decision to demonstrate trust in the suppliers and allow them to grow with the system integrator. Supplier 3 received additional turnover because the system integrator decided to phase out a foreign supplier and give supplier 3 the opportunity to produce these components, new to that supplier. This decision increased the supplier’s faith in its strategic position with the system integrator. After the completion of CO-IMPROVE, supplier 3 offered to buy a new machine able to produce a large part of the new components. The supplier was determined to make the purchase beneficial for all partners and proposed a three-year contract guaranteeing the components to ‘belong’ to the supplier for

Volume 15

Number 4

2006

three years and guaranteeing a set low price for the system integrator during the same period. This was unusual because all other contracts had a duration of one year. Nevertheless, the system integrator decided to sign the contract and this increased trust between the partners further. Supplier 1 experienced a steady growth with the system integrator. They had developed a healthy trusting relationship, which continued to improve throughout the CO-IMPROVE project. When the system integrator was successful in the search for potential suppliers in the Chinese market, the two Danish suppliers were informed that this would not affect their position because their prices were competitive. Also their close geographical proximity was valued by the system integrator. When this was proven to the suppliers, especially supplier 1, trust in the system integrator increased. Dyad 2: The system integrator announced at an early stage of CO-IMPROVE that they were in the process of looking for an additional supplier for the foundry components delivered by supplier 2. This was devastating news to that supplier since the system integrator purchased more than 50% of the supplier’s turnover and wanted to reduce this by 80%. (Commercial) reality was that it took almost two years (until half a year after CO-IMPROVE ended) before turnover was actually removed. However, the fact that the announcement came at an early stage and was not executed immediately, created an uncertain situation for the supplier, which decreased significantly its trust in the system integrator. Interplay between Individual Behaviour and Vision, Power and Trust, and Communication Political and opportunistic behaviour plays a core role in the development of collaborative improvement. It is influenced by vision, or lack thereof. Above all, however, individual behaviour has a dynamic relationship with trust and power in that it ‘produces’ changes in these factors, while also being affected by them at the same time. Vision Æ individual behaviour: Initially, suppliers 2 and 3 lacked a vision on collaboration and relationship development. Supplier 1 was quite visionary in terms of its relationship with the system integrator, but they had not expressed this directly to the system integrator. The system integrator wanted to establish a strategic relationship with the suppliers rolling out its production philosophy to them, but had difficulties expressing that long-term vision (Nielsen et al., 2004; Boer et al., 2005). © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


This created a fair amount of uncertainty and, in effect, a lot of political behaviour, as none of the participants understood where system integrator was heading. Appearing interested but acting reluctantly was a dominant aspect of the suppliers’ behaviour. A clear example of this form of political behaviour took place at a meeting at which the system integrator offered supplier 2 to become their competence centre in foundry technology. The supplier declined the offer on the spot because they were afraid that this was a subterfuge for industrial espionage by the system integrator. Furthermore, the supplier was unwilling to make such long-term commitments anyway. Supplier 2 even engaged in what is best characterized as opportunistic behaviour (‘we don’t need them, they need us’). In dyad 1 the supplier’s vision on collaboration was not fully realized because the system integrator did not share the same vision and the companies had not expressed any actual collaborative vision to each other since they did not know what it implied and how to establish such a relationship. However, this did not create political behaviour as the development of the relationship was already heading towards the level of a full partnership (see Bhote, 1989 for definition). After 17 months of collaboration in CO-IMPROVE, all companies had a better grasp of what a collaborative improvement relationship implied and they discussed the future of each dyad in these terms. Power ´ individual behaviour ´ trust: Even though the system integrator was seeking a collaborative improvement relationship with its suppliers, the company did not find it easy to change its own behaviour. This became most evident when it was time for the partners to select the first improvement projects. The system integrator immediately put itself in charge of the process and did not leave any of the selection to the suppliers. This immediately created a political response from suppliers 2 and 3 – they acted reluctantly but tried to appear interested, because it illustrated to them that the system integrator was still ‘the big bad company’ that made the important decisions and could not be trusted. In fact, power-based behaviour was strongest in dyad 2. At the start of the project, the system integrator replaced a purchaser, who had a good social relationship with the supplier, established over many years (see Kaltoft et al., 2004b), with a purchaser who was very focused on improving operational performance rather than building up a social relationship. The relationship suffered from that as the system integrator’s decision to appoint © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

353

a new purchaser reduced the trust between, and created a great deal of political behaviour from, both parties, including attempts to apply pressure to get the partner to act in the desired direction with the partner trying to counteract and avoid this pressure. One example that eventually ended in some sort of success concerns the way quality problems were handled. Whenever a quality problem occurred, a lot of time was spent on quarrelling about whether it was in fact a problem, how serious it was and what the consequence should be. Eventually, an improvement project was initiated to establish when something should be considered as a quality problem and when a process inspection procedure should be commenced. The result, from numerous discussions with different members in both organizations, was a very clear definition of quality and defects, which allowed the partners to determine very easily if a quality problem had occurred. In spite of this, the trust level of supplier 2, and to a lesser extent supplier 3, was low and stayed low. Supplier 2’s lack of vision and commercial reality led the system integrator to taking away but also occasionally giving back turnover. This created political behaviour in the form of the supplier appearing interested but acting reluctantly to the system integrator’s clearly expressed interest in developing a strategic relationship with that supplier. This, in turn, caused the system integrator never fully to trust supplier 2 either, and in effect the system integrator continued looking for possibilities to replace that supplier. Six months after completion of the project, supplier 2 told us: ‘CO-IMPROVE has made us realize with great certainty that we cannot trust the system integrator’. Supplier 1 and the system integrator never had this kind of relationship. Before COIMPROVE, the trust level between supplier 1 and the system integrator was high and only got higher during the project. In this dyad, (negative) political behaviour did not occur at all. Individual behaviour Æ communication: The fact that some of the participants engaged in political and/or opportunistic behaviour while others did not, had a considerable effect on trust and the use of power, but also on the communication process. None of the participants in dyad 1 displayed overt political or opportunistic behaviour, and this created a healthy collaborative environment with hardly any use of power, a high level of trust between the partners and a sound way of communicating. Dyad 2 experienced a lot of political behaviour from all part-

Volume 15

Number 4

2006

354


ners, and this affected the purchaser’s opinion to the extent that he openly admitted that he did not like the two key representatives of the supplier. As a result the purchaser adopted a power position and was determined to find an alternative supplier. Trust also suffered because of this – both partners mistrusted each other regarding long-term commitments. This was openly admitted by the supplier and expressed through the actions of the system integrator. Whether it was due to mistrust or part of a political game is difficult to say, but at an early stage of the project the supplier would not give the researcher assigned to this dyad access to its manufacturing facilities. They believed the research actually to be industrial espionage. The system integrator exercised pressure and reasoned with the supplier to get the researcher into the company, as the system integrator saw a major advantage in having a neutral person involved in the relationship. The communication process suffered dearly from this situation. Communication was strictly business-related, focused on transactional aspects (Kaltoft et al., 2004b), and could even become quite aggressive at times. Referring to language inappropriate in a business relationship, the supplier said: ‘The situation got so intense sometimes that things were said that should never have been said’. As trust increased, the use of power and political games decreased in dyad 3. The system integrator communicated very openly from the beginning, but had difficulties learning not to exercise its power all the time. The supplier was hesitant, closed and expressed mistrust at first, but learned to communicate at a level of openness not previously shown by this supplier.

Discussion According to Bessant and Caffyn (1997), attempts to implement continuous improvement often fail due to unanticipated difficulties in taking an organization through such a complex change process. It is obvious that implementing collaborative improvement, i.e. inter-organizational continuous improvement, is at least as difficult if not more difficult, as the two or more organizations involved have to manage not only internal but also external issues. This view is supported and exemplified by Barratt (2004): implementing internal change requires a company fully to understand internal processes, and perhaps not many companies do so (Frankel et al., 2002). In a collaborative improvement process, internal as well as external processes must be identified and comprehended.

Volume 15

Number 4

2006

In the literature, a range of factors influencing the development of collaboration between firms have been identified, some endogenous, others exogenous to the collaboration. Previous research (Boer et al., 2005) has confirmed the importance of these factors, identified additional factors not previously published, and suggested that it is not so much the individual factors but rather their interplay, which affects the development of collaborative improvement. The analysis presented in this article confirms this tentative conclusion. We found that the interplay between the factors was important for all three dyads, albeit in a detrimental manner in some cases and beneficial in others. The findings reported in the previous section are summarized in Table 1. The factors with the greatest impact on each other and the other factors and, indeed, the collaboration process, seem to be vision, approach, trust and commercial reality. Several authors have identified the need for a joint objective or vision, and the consequences of the lack thereof in an inter-organizational relationship. Ackermann et al. (2003) label it conflicting goals, Medlin and Quester (2001) call it future orientation, and Huxham (2003) characterizes it as common aims or vision. Our findings show that lack of or, more precisely, poor communication, of a vision induces opportunism and political behaviour, which, in turn, reduces openness of communication. Furthermore, lack of vision leads to an intuitive, culture-driven choice of a consequently onesided and non-sustainable approach towards the development of collaborative improvement. A successful implementation process requires understanding and direction, joint improvement activity and learning, and a genuine willingness to collaborate based on trust and commitment (Kaltoft et al., 2004a). Such an approach may, however, induce the opposite effects, i.e. reduced trust and increased opportunism and political behaviour, especially if one of the partners in the dyad is not genuinely committed to the collaboration but would rather continue relying on an arm’s length relationship (Kaltoft et al., 2004b). Commercial reality may have a similar effect and affect trust and, through that, commitment positively or negatively and, if the latter is the case, induce even more political behaviour.

Conclusion Theoretical Contribution and Further Research The research question underpinning this article was: ‘How does the interplay between key © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


355

Table 1. Interplay Between Factors Affecting the Development of Collaborative Improvement Culture → approach Vision ↔ approach

Approach → communication

Competence → communication

Partner characteristics → collaboration progress

Commercial reality → trust

Vision → individual behaviour Power ↔ individual behaviour

Individual behaviour ↔ trust

Individual behaviour → communication

Culture affects the initial approach which, however, may not be sustainable. Lack of vision allows culture to dominate. Moving to a longer-term competence-based approach requires and enhances the development of a joint vision. A learning-by-doing approach invites operational level communication. A longer-term competencebased approach invites communication about strategic issues. Strategic insight and change competencies are a prerequisite for the partners to get the communication between them beyond operational level issues. Company size and a bureaucratic structure affect decisiveness negatively. Legal ownership combined with a relatively small and flexible organisation enhances decision-making speed. Commercial reality may be perceived as: • A challenge → more trust. • A threat → less trust. A joint vision reduces political behaviour and opportunism and increases commitment. Power distance induces political behaviour and opportunism. Shown commitment and less political behaviour and opportunism reduces (perceived) power distance. Political behaviour and opportunism lead to mistrust, and vice versa. Shown commitment and trust reinforce each other. Political behaviour and opportunism reduce openness. Commitment increases openness.

factors in a collaborative improvement process affect the success of the collaboration?’ The ten factors identified by Boer et al. (2005) do indeed interplay with each other, and that interplay strongly influences the process towards collaborative improvement. The factors and their interplay may be beneficial to, and enhance the success of, the collaboration, but they may also cause failure and deteriorate the collaborative relationship. Table 1 summarizes the main findings reported in this article. The findings are based on longitudinal action research of three dyads. This represents some strengths, in particular with regard to depth and insight into the dynamics involved, but also some weaknesses, as only four companies all from one country and one type of industry were involved. Each of the sentences on the right-hand side of the © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

table therefore actually present a hypothesis, which should be subjected to further investigation. More specifically, further research is needed to establish if the set of factors we identified is complete, if all factors are relevant, and if they interact in the same way in other empirical settings – collaborations in other cultures in which, for example, power distance is greater than in Denmark, international collaborations with greater language, time and/or cultural differences, and collaborations involving companies that have more experience with continuous improvement and/or interorganizational collaboration.

Managerial Implications For comparable settings, some (tentative) managerial implications can be inferred from

Volume 15

Number 4

2006

356


our analysis. In so doing it is important to recognize that some of the factors are more manageable than others, especially vision, approach and competence. The other seven factors are more difficult to manage directly (trust, individual behaviour, power, communication, commercial reality), if at all (culture, partner characteristics), but they can be influenced indirectly. Companies and their managers involved in efforts to establish a collaborative improvement relationship should therefore focus on developing a clear and shared vision and a suitable set of improvement, collaboration and change competences as early in the process as possible. Combined with actually undertaking and, especially, learning from joint improvement activities, vision and competence provide the basis for, what seems to be, a successful approach, towards developing collaborative improvement: understanding and sense of direction, joint improvement activity and learning, and a genuine willingness to collaborate. Such an approach helps in developing the right competences, through training and learning; commitment, rather than opportunism, political behaviour, power games and mistrust; and sufficient robustness to cope with, sometimes awkward, commercial reality.

References Ackermann, F., Franco, L.A., Gallupe, B. and Parent, M. (2003) SGSS for multi-organizational collaboration: reflections on process and content, Strathclyde Business School, Research Paper No. 2003/16. Barratt, M. (2004) Understanding the meaning of collaboration in the supply chain, Supply Chain Management: An International Journal, 9(1), 30–42. Bessant, S. and Caffyn, S. (1997) High-involvement innovation through continuous improvement, International Journal of Technology Management, 14(1), 7–28. Bhote, K.R. (1989) Strategic Supply Management: A Blueprint for Revitalizing the Manufacturer-supplier Partnership, AMACOM, New York. Boer, H., Gertsen, F., Kaltoft, R. and Nielsen, J.S. (2005) Factors affecting the development of collaborative improvement with strategic suppliers, Production Planning & Control, 16(4), 356–67. Buchanan, D. and Badham, R. (1999) Power, Politics and Organizational Change. Winning the Turf Game, Sage, London. Busby, J.S. and Fan I.S. (1993) The extended manufacturing enterprise: its nature and its needs, International Journal of Technology Management, 8(5–6), 294–308. Cagliano, R., Caniato, F., Corso, M. and Spina, L. (2002) Fostering collaborative improvement in extended manufacturing enterprise: a preliminary theory, 4th International CINet Conference, Espoo, Finland, 15–18 September.

Volume 15

Number 4

2006

Child, J. and Faulkner, D. (1998) Strategies of Cooperation: Managing Alliances, Networks, and Joint Ventures, Oxford University Press, Oxford. Cook, K.S. (1977) Exchange and power in networks of inter-organizational relationships, Sociological Quarterly, 18, 62–82. Coughlan, P. and Coghlan, D. (2002), Action research for operations management, International Journal of Operations & Production Management, 22(2), 220–40. Coughlan, P., Coghlan, D., Brennan, L., Lombard, F., McNichols, T. and Nolan, R. (2002) Action learning in the extended manufacturing enterprise: towards an advanced implementation of the NALP approach, 4th International CINet Conference, Espoo, Finland, 15–18 September. Daft, R.L. (2001) Organization Theory and Design, West, St. Paul,. De Maio, M. and Maggiore, E. (1992) Organizzare per innovare – Rapporti evoluti clienti fornitori, Etas Libri (in Italian). DiBella, A.J. and Nevis, E.C. (1998) How Organizations Learn: An Integrated Strategy for Building Learning Capability, Jossey-Bass, San Francisco. Frankel, R., Goldsby, T.J. and Whipple, J.M. (2002) Grocery industry collaboration in the wake of ECR, International Journal of Logistics Management, 13(1), 57–72. Frohlich, M.T. and Westbrook, R.R. (2001) Arcs of integration: an international study of supply chain strategies, Journal of Operations Management, 19, 185–200. Gomes-Casseres, B. (1994) Group versus group: how alliances networks compete, Harvard Business Review, 72(4), 62–74. Huxham, C. (ed.) (1996) Creating Collaborative Advantage, Sage Publishers, London. Huxham, C. (2003) Theorizing collaborative practice, Public Management Review, 5(3), 401–23. Jansen, W., Steenbakkers, W. and Jägers, H. (1999), Electronic commerce and virtual organisations, Electronic Journal of Organizational Virtualness, 2(3), 53–66. Jordan, J. and Michel, F. (2000) Next Generation Manufacturing, John Wiley and Sons, New York. Kaltoft, R., Boer, H., Caniato, F., Gertsen, F., Middel, R. and Nielsen, J.S. (2004a) Implementing collaborative improvement – top-down, bottom-up, or both? 5th International CINet Conference, Sydney, Australia, 22–25 September. Kaltoft, R., Boer, H., Gertsen, F. and Nielsen, J.S. (2004b) Balancing the social and transactional aspects of a collaborative improvement relationship, 5th International CINet Conference, Sydney, Australia, 22–25 September. Kumar, N. (1996) The power of trust in manufacturer–retailer relationships, Harvard Business Review, 74(6), 92–106. McCutcheon, D. and Stuart, F.I. (2000) Issues in the choice of supplier alliance partners, Journal of Operations Management, 18, 279–301. Marshall, C. and Rossman, G. (1999) Designing Qualitative Research, Sage Publications, California. Medlin, C. and Quester, P.G. (2001) A collaborative interest model of relational coordination: examining relational norms as actor bonds, 17th IMP conference, Oslo, Norway, 9–11 September. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


Middel, R., Gieskes, J. and Fisscher, O. (2002) Driving collaborative improvement processes, 4th International CINet Conference, Espoo, Finland, 15–18 September. Mintzberg, H. (1983) Structures in Fives, Prentice Hall, Englewood Cliffs. Mohr, J. and Spekman, R. (1994) Characteristics of partnership success: partnership attributes, communication behaviour and conflict resolution techniques, Strategic Management Journal, 15, 135– 52. Monczka, R.M., Petersen, K.J., Handfield, R.B. and Ragatz, G.L. (1998) Success factors in strategic supplier alliances: the buying company perspective, Decision Sciences, 29(3), 553–77. Moore, K.R. (1998) Trust and relationship commitment in logistics alliances: a buyer perspective, International Journal of Purchasing and Materials Management, 34(1), 24–37. Nielsen, J.S., Boer, H., Gertsen, F. and Kaltoft, R. (2004) The influence of power, trust and political behaviour in the process of collaborative improvement, 5th International CINet Conference, Sydney, Australia, 22–25 September. Olivari, P., Smeds, R. and Corso, M. (2002) Continuous learning in product development: a crosscultural comparison between Finland and Italy, 4th International CINet Conference, Espoo, Finland, 15–18 September. Pfeffer, J. (1992) Managing with Power, Harvard Business School Press, Boston. Quinn, J.B. (1999) Strategic outsourcing: leveraging knowledge capabilities, Sloan Management Review, 40, 9–36. Quinn, J.B. and Hilmer, E. (1994) Strategic outsourcing, Sloan Management Review, 35, 43–55. Rice, J.B. and Hoppe, R.M. (2001) Supply chain versus supply chain: the hype and the reality, Supply Chain Management Review, September–October, 46–54. Sako, M. (1992) Prices, Quality and Trust: Inter-firm Relations in Britain and Japan, Cambridge University Press, Cambridge. Stock, G.N., Greis, N.P. and Kasarda, J.D. (2000) Enterprise logistics and supply chain structure: the role of fit, Journal of Operations Management, 18, 531–47. Vangen, S. and Huxham, C. (2003) Building trust in inter-organizational collaboration, EURAM Conference, Milan, Italy, 3–5 April. Venkatesan, R. (1992) Sourcing: to make or not to make, Harvard Business Review, 70(6), pp. 98–107. Williamson, O.E. (1981) The economics of organizations: the transaction cost approach, American Journal of Sociology, 87(3), 548–77. Winer, M. and Ray, K. (1994) Collaboration Handbook; Creating, Sustaining and Enjoying the Journey, Amherst H. Wilder Foundation, Saint Paul, MN.


357

Dr Rasmus Kaltoft ([email protected]) holds an MSc Degree in Industrial Engineering from Aalborg University (2001), and finished his PhD study at the Center for Industrial Production, also at Aalborg University. Dr Kaltoft has published articles in the fields of Supply Chain Management, Change Management, Collaborative Improvement and Management Gaming. During his PhD studies, Dr Kaltoft was a member of the CO-IMPROVE team. His main interests are collaborative improvement, continuous improvement, and supplier/customer relationships. Dr Kaltoft is currently employed as a senior consultant at Capacent. Harry Boer ([email protected]) is Professor of Organizational Design and Change at the Center for Industrial Production at Aalborg University (Denmark) and the Department of Business Administration at Stavanger University (Norway). He holds a BSc in Applied Mathematics and an MSc and PhD both in Management Engineering. He has (co-)authored numerous articles and several books on subjects such as Organization Theory, Flexible Automation, Manufacturing Strategy, and New Product Development and Continuous Improvement and Innovation. His research focuses on Continuous Innovation or, more precisely, the interaction between day-today operations, incremental change and radical innovation. Professor Ross Chapman (r.chapman@ uws.edu.au) is the Associate Dean (Research) for the College of Business and the Acting Director of the University Centre for Industry and Innovation Studies (CInIS) at the University of Western Sydney, Sydney, Australia. He has worked in private industry in technical, QC/QA and R&D Management positions, and has taught and researched predominantly in the areas of Quality Management; Continuous Improvement; Innovation and Technology Management; and Operations Management. Ross is author or co-author of three books and over 80 refereed journal and conference papers in the above areas; is a Regional Editor or Editorial Board Member for several international journals, and is a Founder of the Board of the Continuous Innovation Network (CINet). Frank Gertsen (fgertsen@production. aau.dk) is Professor of Innovation Man-

Volume 15

Number 4

2006

358


agement at the Center for Industrial Production at Aalborg University in Denmark. He is head of the PhD program in Mechanical Engineering at the Faculty of Engineering, Science and Medicine. He is on the interim board of the Center for Innovation and Product Development (CIPU). He is member of the CINet, the IMSS, the Academy of Management, and the EurOMA. He has been a visiting scholar at Stanford University and Hong Kong City University. He has edited several international. journals and is on the editorial board of Creativity and Innovation Management. He has written more than 100 research papers and book contributions. His research interest is in the area of innovation and change and his current project involvement includes Discontinuous Innovation Project, Continuous Innovation, Radical Corporate Entrepreneurship project. He currently teaches innovation management at PhD, masters, and executive MBA levels. Jacob S. Nielsen ([email protected]) is an assistant professor at the Center for Industrial Production at Aalborg University. He holds an MSc in Business Studies and successfully defended his PhD thesis entitled ‘Collaborative Lea(r)ning’ in March 2006 at Aalborg University. His research focuses on lean manufacturing and collaborative improvement and learning in interorganizational settings.

Volume 15

Number 4

2006


CORPORATE ENTREPRENEURSHIP AND RADICAL INNOVATION

359

The Nexus of Corporate Entrepreneurship and Radical Innovation Astrid Heidemann Lassen, Frank Gertsen and Jens Ove Riis This paper explores the linkage between the entrepreneurial orientation of established firms and the development of radical innovation. Through five case studies in firms involved in radical innovation, three propositions are developed, suggesting that proactiveness, risktaking and autonomy stimulate the development of radical innovation, whereas competitive aggressiveness does not necessarily do so, as radical innovations are directed towards the creation of entirely new arenas of business, where existing competitors are not present.

Introduction

U

nderstanding innovative processes has become a central theme in much literature (e.g. Miles and Snow, 1978; Miller, 1983; March, 1991; Tidd et al., 1997). Within the field of innovation, radical innovation has increasingly captured the attention of researchers (Green et al., 1995; Gilbert, 2002; Darroch and McNaughton, 2002; Chandy et al., 2003, Sorescu et al., 2003). This is due to the perception that such innovations serve as the basis of new technological trajectories and paradigms and are an important part of the process of creative destruction in which extant techniques and approaches are replaced by new technologies and products (Lynn et al., 1996, McDermott and O’Connor, 2002). In fact, breakthrough ideas almost by definition, serve as the basis of future technologies, products, services and industries (Tushman and Anderson, 1986). Also within the field of entrepreneurship an interest in radical innovations has been found in the exploration of how to define entrepreneurship. The economist who, above all others in the 20th century, emphasized the centrality of entrepreneurship and innovation in the understanding of the dynamics of industrial and economic change in the capitalist system was Schumpeter, whose earlier work, in particular, places the entrepreneur at the centre of ‘creative destruction’ (Schumpeter, 1934). After Schumpeter’s work, most researchers © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

have accepted his identification of entrepreneurship with innovation (Stopford and Baden-Fuller, 1994) although his work is not exclusive regarding the context in which the entrepreneur works. Especially in the corporate entrepreneurship literature (entrepreneurship within existing organizations) there is a considerable overlap between focus on innovation and entrepreneurship. As the need to understand and develop corporate entrepreneurship increases (Stevenson and Jarillo, 1990; Dess et al., 2003), so does the need to understand the linkages between entrepreneurship and radical innovation. However, scholarly studies have to a large extent segregated the two theoretical fields, focusing on either entrepreneurship or innovation as independent processes, thereby limiting their utility and application in a corporate context (Baum et al., 2001; Chesbrough, 2003). The aim of this paper is to explicitly address this linkage through conceptual discussion of the two areas. Based on this understanding, empirical research through exploratory and inductive case studies is analysed. In order to do so, corporate entrepreneurship is addressed through the entrepreneurial orientation framework (Lumpkin and Dess, 1996), and the fundamental reasoning, in line with the thoughts of both Kirchhoff (1991) and Schumpeter (1934), that entrepreneurial activity often lies at the core of radical or ‘breakthrough’ innovations that lead to wealth creation, is adopted.

Volume 15

Number 4

2006

doi:10.1111/j.1467-8691.2006.00406.x

360


Furthermore, focus is placed on technology as the source of radical innovation. As such the research question of this paper is: RQ: ‘How does corporate entrepreneurship in terms of an entrepreneurial orientation influence radical innovation in existing organizations?’ In order to explore the linkage between radical innovation and corporate entrepreneurship, a literature review has initially been conducted to create an understanding of the theoretical focus areas of the two fields. Then a description of the methodology of the empirical research is presented. This leads to a presentation of the empirical findings, on the basis of which three specific propositions on the linkage between corporate entrepreneurship and radical innovation are developed. These propositions are discussed in answer to the proposed research question. Finally, an overall conclusion on the research is reached, including suggestions for further research.

Theoretical Background In this section a theoretical understanding of the concepts of radical innovation and corporate entrepreneurship is presented. Based on that, an initial framework is created, through which the empirical data is analysed.

Radical Innovation Innovativeness ranges from incremental to radical. Incremental innovation is critical to sustaining and enhancing shares in mainstream markets (Baden-Fuller and Pitt, 1996) and focuses on improving existing products and services to satisfy ever-changing customer demands (Bessant, 2003). Radical breakthroughs, in contrast, serve as the basis for future technologies, products, services and industries (Christensen, 1997; Hamel, 2000; Abetti, 2000). Terms such as ‘disruptive’, ‘radical’, ‘non-linear’, ‘discontinuous’, ‘breakthrough’, and ‘paradigm-shifting’ have all been used to describe what in essence is breaking away from the customary, creating entirely new possibilities for growth. Radical innovation can, however, be defined along different dimensions, and some of the liveliest debates in both industry and academia have centered on this. The term ‘radical innovation’ is thus applied to describe innovation that is highly revolutionary or discontinuous, and represents a new paradigm that can generate new wealth whilst transforming or displacing some parts or all of an established market (Christensen, 1997).

Volume 15

Number 4

2006

For the purpose of this paper, radical innovation is defined following the definition of O’Connor and Ayers (2005), as the commercialization of products or technologies that have a strong impact on (a) the market, in terms of offering wholly new benefits, and (b) the firm, in terms of generating new business. To specify the market impact in more detail, Leifer et al.’s (2000) definition is also considered, namely that a radical innovation must entail at least one of the following: 1. world’s-first performance features 2. significant (5–10 times) improvement in known features or 3. significant (30–50%) reduction in cost. Thus, the key to radical innovation is the amount of new value added through exploration and exploitation of new opportunities. Resonating Schumpeter’s (1934) observations, innovation can furthermore be perceived in terms of type of innovation. Hence, the term refers to both the introduction of a new product, process, technology, system, technique, resource, and capability to the firm or its markets. In this article focus is placed on technology as the source of major change, while being aware that the sources can be of different natures, such as sea-changes in the market, policy, regulatory regimes, ‘unthinkable’ events (9/11), or business model innovation (Tidd et al., 2005). A recent popular example of the latter is the so-called ‘blue ocean’ thinking (Kim and Mauborgne, 2004), which is a synthetic re-mix of the customer-values offered, which appears new in the face of the customers yet minimizes costs and becomes uncontested by competitors.

Corporate Entrepreneurship While the literature on corporate entrepreneurship is diverse and ranges from considerations on the role of the individual (Pinchot, 1985) to corporate venturing and corporate entrepreneurship’s macroeconomic effects (Zahra, 1991, 1993), corporate entrepreneurship is in this paper considered ‘an entrepreneurial orientation, which permeates an organization’s outlook and operations leading to a variety of outcome’ (Lumpkin and Dess, 1996). This definition is based on Miller and Friesen’s (1983) categorization of innovative strategy making. Covin and Slevin (1986, 1991) expanded this concept, called it ‘entrepreneurial posture’, and phrased three now widely-accepted characteristics of firm level entrepreneurship: 1. innovativeness © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


2. proactiveness and 3. risk-taking. Lumpkin and Dess (1996) renamed the concept the ‘entrepreneurial orientation’ and further extended the concept by identifying two additional dimensions: 4. autonomy and 5. competitive aggressiveness. By limiting the focus of the paper to entrepreneurial orientation, this work is positioned within a perspective on corporate entrepreneurship emphasizing not merely the formation of new firms, but rather the process of exploring and exploiting opportunities demonstrated by the entrepreneurial organization (Quinn, 1979; Schollhammer, 1981; Burgelman, 1983; Drucker, 1985; Rule and Irwin, 1988; Stevenson and Jarillo, 1990; Kuratko et al., 1993; Stopford and Baden-Fuller, 1994). This includes innovative activities and orientations aimed at the development of new products, services, technologies, administrative techniques, strategies and competitive postures (Miller, 1983; Covin and Slevin, 1986, 1989, 1991; Morris and Paul, 1987; Schafer, 1990; Stevenson and Jarillo, 1990; Naman and Slevin, 1993; Lumpkin and Dess, 1996). Innovativeness is referred to, in wide Schumpeterian terms in the entrepreneurial orientation construct, as a firm’s tendency to engage in and support new ideas, novelty, experimentation, and creative processes that may result in new products, services, or technological processes. It is recognized that innovativeness can vary in degree of radicalism, but essentially it refers to a willingness to depart from existing technologies or practices and venture beyond the current state of the art (Kimberly, 1981). The common denominator amongst all entrepreneurial firms is in this paper considered to be that they innovate. As such, innovation can be perceived to be a focal point of corporate entrepreneurship. Stopford and Baden-Fuller (1994) underline this when they write that ‘most authors accept that all types of entrepreneurship are based on innovations.’ So, without innovation there is no corporate entrepreneurship regardless of the presence of other dimensions termed entrepreneurial in the literature. Following this logic, the label entrepreneurial should not be applied to firms that are not innovative. In the following sections innovativeness is as such not addressed as a separate dimension, but is considered the primary dependent variable reflected upon through the other dimensions. Proactiveness is related to the act of initiative in the entrepreneurial process, and to acting in anticipation of future problems, needs © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

361

or changes. For example Lieberman and Montgomery (1988) emphasized the importance of first-mover advantage as the best strategy for capitalizing on a market opportunity. Thus, proactiveness by anticipating and pursuing new opportunities and by participating in emerging markets has become associated with entrepreneurship. The proactiveness dimension of entrepreneurial orientation therefore most closely resembles the ideas suggested by Miles and Snow (1978) of a prospector type, about which they stated: the Prospector’s prime capability is that of finding and exploiting new products and market opportunities . . . Prospectors are frequently the creators of change in their respective industries. Change is one of the major tools used by the Prospector to gain an edge over competitors. Risk-taking is a quality that is often used to describe entrepreneurship. It can have various meanings, depending on the context in which it is applied. In this paper, risk is seen in the context of strategy, which according to Baird and Thomas (1985, pp. 231–2) can take three forms: (a) venturing into the unknown (b) committing a relatively large portion of assets; and (c) borrowing heavily. Venturing into the unknown involves a sense of uncertainty and unfamiliarity that may apply generally to different types of risk. Financial risk refers to a risk-return trade-off, and was used by Miller and Friesen (1978) when they defined risk taking as ‘the degree to which managers are willing to make large and risky resource commitments – i.e., those which have a reasonable chance of costly failures’. Autonomy is an essential element of entrepreneurship, and is traditionally seen through the formation of new and independent ventures. However, in an organizational context autonomy refers to the independent actions of an individual or a team in bringing forth an idea or a vision and carrying it through to completion (Lumpkin and Dess, 1996). As such their actions are taken free of rigid restrictions and where the individual or team throughout the process remains free to act independently, to make key decisions, and to proceed. Fostering autonomy in an organizational context may involve flattening hierarchies and delegating authority to operating units, and draws on the idea of Burns and Stalker’s organic management systems, promoting decentralization, low formalization, dynamic behaviour, learning and flexibility in structures and processes of the organization

Volume 15

Number 4

2006

362


(Burns and Stalker, 1961; Drucker, 1988; Nonaka, 1991; Senge, 1994). Competitive aggressiveness refers to a firm’s predisposition to directly and intensely challenge its competitors to achieve entry into a new market or improve the present position in an existing market; that is, a strong focus on outperforming competitors. According to Lumpkin and Dess (1996) competitive aggressiveness is characterized by responsiveness, and a willingness to be unconventional rather than rely on traditional methods of competing. The propensity of the firm to analyse and target a competitor’s weaknesses (MacMillan and Jones, 1984), as well as the tendency to be first-to-market with new product offerings on a frequent basis (Porter, 1985), are also considered elements of competitive aggressiveness.

Methods The methods used in order to explore these questions are of an inductive nature, being qualitative case studies in five firms. Qualitative case studies were chosen as research design in order to explore how an entrepreneurial orientation influences radical innovation in a real-life context (Yin, 1994). A purposive sample of firms was selected to provide information-rich cases, and the criteria for selection were that: • the firms were highly experienced with entrepreneurship and innovation • the firms were located within a high-tech industry • the firms had engaged in technology-based radical innovation • the firms were of different sizes and age (year of establishment). Five cases were selected in order to achieve some robustness in the inductive findings despite the variety of the firm context. An overview of the cases/firms is presented in Table 2. Determining the degree of innovativeness in each case, the cases were evaluated in relation to the definitions of radical innovation

by O’Connor and Ayers (2005) and Leifer et al. (2000). As appears in Table 1, all cases complied with the definition by O’Connor and Ayers (2005) of offering wholly new benefits to the market and generating new business within the firms. Adding Leifer et al.’s (2000) definition, it also became evident that all cases presented significant improvements in known features. However, not all cases were characterized by introducing ‘world’s-first’ performance features or significant cost reductions. Where ‘world’s-first’ performance features were not introduced, significant cost reductions were revealed, which allows for entirely new market possibilities. Regarding involvement in high-tech industry, the firms were located within different sectors, but were all involved in high-tech development and manufacturing. The sectors they represent are healthcare, audio, wireless technology, and heating (see Table 2). The criteria for differences in size and age were adopted to reflect on the definition of corporate entrepreneurial processes taking place in organizations regardless of size (Antoncic and Hisrich, 2003). These criteria were met by the sample ranging from 32 to 20,250 employees and from 5 to 84 years since establishment (see Table 2). This allowed for considerations on the effect of size and age on the linkage between corporate entrepreneurship and radical innovation. The data was collected using qualitative semi-structured interviews with senior managers, middle managers, project leaders and R&D professionals in order to examine their perception of and experiences with corporate entrepreneurship and radical innovation, and the factors that contribute to their development and integration. Conversations were recorded and subsequently transcribed and discussed with the interviewees, before synthesizing into the frame of a case. By carrying out interviews at several different levels of the organization, the research seeks to gain differing views of the linkage between corporate entrepreneurship and radical innovation. Furthermore, the case studies were conducted fol-

Table 1. Degree of Innovation in Cases Indication of radical innovation

Volume 15

Cases

A

B

C

D

E

Offering wholly new benefits to the market Generating new business within the organization ‘World’s-first’ performance features Significant (5–10 times) improvement in known features Significant (30–50%) reduction in cost

• • • •

• • • • •

• • • •

• •

• •

• •

• •

Number 4

2006



363

Table 2. Overview of Cases Case

Characteristics of the firm

Innovation

A

Established in 1976. Profile: Develops, manufactures and markets professional audio products. Size: 185 people working in 15 different countries. World leader Established in 1999 as a spin-off from wellestablished firm. Profile: Audio power conversion. Size: 35 employees Established in 1922. Profile: Healthcare firm focused on diabetes care. Size: Employs approximately 20,250. World leader Established in 2001 as a division of a British firm. Profile: The wireless communication industry. Size: 32 employees in the Danish division Established in 1933. Profile: Refrigeration and aircondition, Heating and Motion Control (case within Heating and Water division). Size: 17,500 employees. Market leader

Unique technology within digital signal processing

B

C

D

E

lowing the same protocol in order to ensure the reliability of the results as well as to reflect the validity of the differences that might emerge (Yin, 1994). Relevant documentation was additionally provided by the respondents both before and after the interviews. This included strategic documentation, product development roadmaps and funding proposals. This data has been used to cross-reference findings from the interviews and to provide additional historical background to the case studies. Quotations appearing in this research are translated from Danish into English. In Table 2, a short description of the cases is given, stating firm facts and the innovations around which the cases evolved.

Findings Table 3 illustrates selected statements made by respondents from the five cases reflecting on the dimensions of an entrepreneurial orientation: proactiveness, risk-taking, autonomy, and competitive aggressiveness, in relation to the development of radical innovation. There is a general consensus amongst the interviewees that the four dimensions of entrepreneurial orientation are highly related to the development of radical innovation. The empirical data shows differences in the types of difficulties experiences by the firms, depending on the size of the organization, but © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

Unique audio power conversion technology Monitor of intracellular events and protein translocation in real time Front edge silicon IP for wireless terminals Radical rethinking of CO2 sensoring technology

essentially the same factors are highlighted as being important.

Discussion and Development of Propositions As seen through the literature review, both theories on radical innovation and corporate entrepreneurship depart from an emphasis on the disturbance of status quo in the market or industry and focus on exploration of the unknown as opposed to exploitation of the already known (March, 1991). The empirical data also showed that a positive sense of a synergetic linkage between the two constructs is created. On the basis of the findings, three propositions on the linkage between corporate entrepreneurship and radical innovation are developed in this section.

Proactiveness and Risk-taking An orientation towards proactivity and change encompasses inherent risks of low profitability and overextension of resources, thereby closely relating the term proactiveness to risk. Risk-taking is often referred to as highly related to uncertainty, lack of planning, and unmanageable actions (Osborne, 1995; Sarasvathy, 2001). Through the case studies, an interesting perspective on proactivity and

Volume 15

Number 4

2006

364


Table 3. Selected Statements Reflecting on the Dimensions of Entrepreneurial Orientation and Radical Innovation

Volume 15

Proactiveness

Risk-taking

Autonomy

Competitive aggressiveness

A

’Some say that “luck follows the crazy”, but I strongly believe that “luck follows the ones who seek it”. So we need to “seek” continuously, and this will create possibilities of riding on entirely new waves and gaining a unique market position’. (CEO)

‘Radical innovation is really high-risk, and it is important not to loose sight of ‘cash-cow’ products in that process. But if your mind-set is that it is necessary to create a balance between the two, the need for more risk-willingness in radical innovative projects becomes more manageable’. (R&D professional)

‘We have been very used to breakthrough ideas surging automatically, because of some highly innovative individuals in the organization. But, now we also try to stimulate a culture of “small intrapreneurs” in the R&D department, meaning that we encourage people who find it fascinating to spot both technological and market opportunities to do so’. (Project Manager)

‘We work very “blue ocean-like”, meaning that of course we know our markets and the competitors who operate there. But when it comes to radical innovation, we haven’t directly targeted any existing competitors – instead we have focused on creating new markets, where we would automatically be leading. Of course some of the existing competitors have been defeated based on this, but it has not been our primary objective to do so’. (CEO)

B

‘We always have Masters students and PhD students in our organizations. They work on topics where we can see possibilities, but don’t have the time to focus ourselves. This way we can afford to continuously explore new areas and spot new opportunities in spite of our small size’. (CEO)

‘Of course management said no the first few times – it was an entirely new area for the firm. But, because they kept an open mind we were able to show them the perspectives in the project, and they chose to take the risk even though they didn’t normally do so’. (CEO)

‘Our parent firm went from being very traditional and mainly focused on their internal ability to innovate, to being open towards exploring new possibilities through establishment of spinoff ventures. In this way, the openness towards change and willingness to finance a new venture made the development of our technology possible. We needed to detach ourselves a bit from the traditional ways of our parent firm to be able to grow’. (CEO)

‘Audio-power is applicable is so many different places, and in this sense our market is very large. However, what we think our technology will do is to revolutionize and change many of the traditional markets. Therefore we don’t directly compete with existing players in these markets, but focus on convincing customers of the potential of our technology and in this way create our own markets’. (CEO)

Number 4

2006



365

Table 3. conitnued Proactiveness

Risk-taking

Autonomy


C

’The management is very proactive in certain areas. For example they create internal competitions, where great ideas can win funding to become established projects in the organization. But it is difficult in such a big organization; the individual idea is easily lost amongst many’. (R&D professional)

‘Risk-willingness is incredibly important for the more radical innovative projects. Especially in a large organization where business potential is measured very much on the relationship between development expenses and potential earnings and how close to the core business the new technology is. We make billions here, so a potential ROI of millions is not always enough’. (Innovation Manager)

‘I think it is really worth a thought or two that the project was realized only because (x) was so persistent on the fact that it needed to become a separate venture in order to survive. Besides being involved in the development of the technology, he is also a great politician, and I think that’s a skill many intrapreneurs need’. (R&D professional)

’There are only 3–4 direct competitors to our firm on a global basis because we are so large. So radically new projects are normally not considered in relation to being competitive aggressive, but more as possibilities of taking us in a direction where our competitors are not’. (Innovation Manager)

D

‘At the moment we work on exploiting all the possibilities our technology can create for a very large customer. But we know that at the same time it is important to keep on focusing on other things as well. Therefore we still allocate resources to making prototypes of technologies that, for example, can do the same thing in a very different way, or something entirely different’. (CEO)

‘A few years ago there was a serious decline in our main market in Asia due to SARS. So we basically didn’t make any money at a time when we were geared for growing. But, our employees all believed so much in our idea that they were willing to accept a salary cut of 10 percent for a while until we got back on our feet. That I think shows that our organizational culture builds on a certain riskwillingness’. (R&D professional)

‘In my former job, I had increasingly become manager rather than developer, and I didn’t like it – not with my personality. So I talked to a few of my colleagues about a joint idea we had, and we very quickly decided to start on our own’. (CEO)

‘We realized that many of our dreams could come true by becoming a separate venture from a large British corporation, instead of competing with them. We still had the opportunity to create something unique and offer it to the world’. (CEO)


Volume 15

Number 4

2006

366


Table 3. conitnued

E

Proactiveness

Risk-taking

Autonomy


‘We know already that several of our main markets will stagnate in the future, and we are therefore continuously on the lookout for new areas. This project is one of the potential areas’. (Director of Technical Business Development)

‘We work with the market of tomorrow and therefore our patience is much higher than in other places in the organization. The way the projects are evaluated is not ROI potential from the early phases, but the progress made and the knowledge created at certain milestones. This way we try to create an entirely different culture and environment here’. (Innovation Manager)

‘We have a large venture department, which supports internal intrapreneurs. We try to educate our employees to come up with great new ideas and have the courage to do something about them. I think that the more activity we have in this region the more different competences our firm will have to draw on. In this sense, not all ideas developed internally should necessarily stay here. Independent ventures with the flexibility of a small firm could be a much better place to develop them’. (Innovation Manager)

‘We try to look entirely outside of what is normally done, and do something different. Competitors aren’t doing it either’. (R&D professional)

risk-taking appeared, suggesting the two terms can be considered jointly. All five case studies depict firms that can be categorized as prospectors (Miles and Snow, 1978), creating changes in the industry and gaining first-mover advantages by participating in emerging markets. As can be seen from the selected quotations, the firms acted proactively in anticipation of future problems, needs or changes by seeking out new opportunities. In cases A–D the proactiveness is highly related to their general sense of need for continuous exploration of new opportunities, while in case E the exploration is connected to the active realization of a possible future problem in terms of stagnation of main markets. The development of radical innovation was in all cases directly connected to this proactive exploration of new opportunities in terms of new markets, new possibilities of applications, or new technologies. The CEO in case A expresses this proactive attitude very eloquently: ‘… I strongly believe that “luck fol-

Volume 15

Number 4

2006

lows the ones who seek it”. So we need to “seek” continuously, and this will create possibilities of riding on entirely new waves and gaining a unique market position’. A common characteristic of the proactive approach to radical innovation in the cases is an ever present focus on flexibility, responsiveness, open-minded questioning and identifying new opportunities. The cases show that flexibility here is characterized by more than merely the ability to react quickly to changes, but also comprises the ability to incorporate change as a continuous consideration in the organization. As such, change is not met with resistance, but is perceived as a natural process, and proactiveness is stimulated on an ongoing basis. Proactiveness being a central factor in the successful development of radical innovation in the case studies is expressed through several different actions and considerations. Examples of such actions include: all cases showed a continuous re-evaluation of possible © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


new ways of applying existing technology; case D and E allocated resources to prototyping of new technology that may essentially threaten the existing technologies of the firm; and case B used cooperation with university graduate students as an affordable way of exploring new areas while also exploiting existing opportunities in spite of limited resources. The actions taken by the case firms which could be interpreted as risk-taking (venturing into the unknown and committing large portions of assets to such projects), were largely perceived by the firms as dependent on their ability to manage risks rather than on a willingness to take risks. As such, cases A and E show how learning from one radically innovative project to another is essential. Through willingness to continuously allocate resources to the creation of radical innovations an understanding of how to handle the radical innovation process is created. This generates a high expertise in how to interpret different risks, making the decision process easier as well as creating a high tolerance for risk and avoiding a negative atmosphere of anxiety connected to uncertainty. This knowledge on how managing the risks involved in radical innovation is also expressed through the approach to evaluation of the projects, which differed from conventional evaluation criteria. A venture-manager in case E expressed this as: ‘We work with the market of tomorrow and therefore our patience is much higher. The way the projects are evaluated is not ROI potential from the early phases, but the progress made and the knowledge created by certain milestones …’. In case firms A and B, innovation projects were, furthermore, often financed in cooperation with other firms or research institutions as a way to reduce the financial risks. Examples of strategies to cope with risk include: continuous allocation of resources to exploration of new knowledge; accumulating knowledge on how to interpret and handle risk in radical innovation; creating separate evaluation criteria based on progress in knowledge creation rather than ROI potential; and shared knowledge and shared risks through cooperation with external partners. Based on the above discussion, it is relevant to argue that an emphasis on proactiveness and risk-taking stimulates radical innovation. Hence, the following proposition and the nexus between corporate entrepreneurship and radical innovation is stated as: Proposition 1: A firm emphasizing an entrepreneurial orientation in terms of proactiveness and risk-taking stimulates radical innovation. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

367

Autonomy An emphasis on autonomy and its importance for the development of radical innovation in the case studies was evident in several different areas; for example, the organizational framing of the radical innovation, the focus on the individual, and the management styles. Establishment of autonomous spin-off firms in cases A, B and C was a means by which the most suitable organizational frame for the development and exploitation of the radical innovation was created. The spin-off firms typically inherited the most suitable routines from the parent firm but were also free to create independent processes specifically aimed at the radical innovation. Through openness to the establishment of such new organizational constellations a high degree of flexibility surrounding the radical innovation was obtained even when the parent firm itself could not change quickly. This flexible approach has also enabled the continuous re-evaluation of possible new applications for existing technology. It was furthermore found in all the case studies that the individual’s orientation towards autonomy is an essential factor in the development of a radical innovation. In cases A, B and D, self-organized groups are a preferred working method. Through these groups a dynamic atmosphere and individual ownership is created. The individuals in the groups are continuously oriented towards innovation, a crosssectional understanding of innovation in the organization, and a sense of joint responsibility for the work. As such, the role of the individuals is essential in relation to the way new ideas are brought into existence, the way understanding for and commitment to the ideas is created in the organization, the way the radical innovation is developed, and the way communication across the organizations takes place. In all five cases this is, for example, seen through a concentration of projects around motivating key individuals who have insight into the project throughout the entire process. As such, it was clear to see that individuals who were specifically motivating, knowledgeable or in other ways stood out from the crowd were the focal point of the development of radical innovation projects. In case C this was expressed by an R&D professional as: ‘I think it is really worth a thought or two that the project was realized only because (x) was so persistent on the fact that it needed to become a separate venture in order to survive. Besides being involved in the development of the technology, he is also a great politician, and I think that’s a skill many intrapreneurs need’.

Volume 15

Number 4

2006

368


In order to ensure a successful outcome of a radical innovation process, top-management commitment is also stressed as a determining factor (Kanter, 1984; Caffyn, 1999; Simon et al., 2003; Mikkelsen, 2005). This commitment could encompass control, prioritising, protection, resource allocation, active involvement, etc. In all the case studies, a visible commitment from the top-management was highlighted as a very important factor, and in particular prioritizing and resource allocation were visible signs of such commitment. Additionally, positive attention and freedom to make independent decisions rather than detailed management from the top level were highlighted, underpinning the need for management to allow for a certain degree of autonomy in radically innovative projects. In cases A, B and E such commitment was present, whereas cases C and D stressed its importance while also outlining the difficulties in achieving such commitment. In cases C and D the success of the radical innovation was instead highly related to the drive of specifically motivated individuals. Based on the above observation, it is likely that a corporate environment emphasizing autonomy is an essential factor in the development of radical innovation. Hence, a second proposition on the nexus between corporate entrepreneurship and radical innovation is: Proposition 2: A firm emphasizing an entrepreneurial orientation in terms of autonomy stimulates radical innovation.

Competitive Aggressiveness As already touched upon the case firms portray characteristics in line with the prospector type (Miles and Snow, 1978), which is reflected through their proactive behaviour, frequent creation of change and flexible business models. This also illustrates a high degree of responsiveness towards potential opportunities in the market. Interlinked with responsiveness was also a high tendency to be first-tomarket with new products or technologies. As the cases evolved around the development and implementation of radical innovations, being first-to-market in this connection was a given. However, taking a broader look at the case firms, this tendency could be found both in terms of the radical innovation had incremental improvements introduced by the firms. In several of the cases unconventional approaches were used in order to create the best possible conditions for the development and implementation of radical innovation. This often provided a competitive advantage. However, looking at the strategic consider-

Volume 15

Number 4

2006

ations behind these unconventional approaches, focus was to a greater extent on how to develop the best circumstances for the exploration and exploitation of the innovation than specifically on how to gain the best competitive advantage. This observation was even further elucidated, when considering the propensity of the firms to analyse and target a competitor’s weaknesses (MacMillan and Jones, 1984). The cases showed that all the firms had a very good understanding of their competitors and the market the innovation was to be implemented in, but also that their focus was on the market the innovation was going to create rather than on how to compete in the existing market. It was found that the understanding of existing players’ weaknesses was good, but the focus on specifically targeting these weaknesses was low, as the intent was to create entirely new markets. The CEO in case B expressed this in the following way: ‘Audiopower is applicable is so many different places, and in this sense our market is very large. However what we think our technology will do is to revolutionize and change many of the traditional markets. Therefore we don’t directly compete with existing players in these markets, but focus on convincing customers of the potential of our technology and in this way create our own markets’. Hence, it is relevant to stress the proactive intent of the firm rather than competitive aggressiveness, in relation to radical innovation, as this refers to how a firm relates to market opportunities in the process of new entry; it does so by seizing initiative and acting opportunistically in order to shape the environment, through influencing trends and creating new demands. Competitive aggressiveness, in contrast, refers to how firms relate to competitors, that is, how firms respond to trends and demand that already exists in the market. Based on the above, it is argued that a third proposition on the linkage between corporate entrepreneurship and radical innovation is relevant. Proposition 3: A firm which does not emphasize an entrepreneurial orientation in terms of competitive aggressiveness has increased likelihood of stimulating radical innovation.

Conclusion The pursuit of creative new solutions to the challenges facing the firms of today means that understanding different aspects of innovation has increasingly become a focus of attention in both theory and practice. The idea © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


of innovation has evolved within several different fields of research, however, often without the establishment of clear conceptual connections between theories developed in the respective fields. Within the field of entrepreneurship research it is seen that, though progressing into theories on corporate entrepreneurship, innovation is merely addressed as one factor amongst others without clearly establishing a link to innovation theory. Within innovation theory considerations on entrepreneurship are most often only mentioned in connection with Schumpeter’s founding thoughts on the effect of entrepreneurship and innovation on the macroeconomic state of society. The intention of this paper was to start discussing the link between corporate entrepreneurship and radical innovation in order to clarify the interrelatedness between the two constructs. Radical innovation is in this connection argued to share conceptual foundations with corporate entrepreneurship as both constructs are based on a focus on departing from the customary and bringing something new into existence. Positioned within the stream of entrepreneurship research, which focuses on innovation as the common denominator among all firms that could be described as entrepreneurial, this paper focused on the entrepreneurial orientation of the firm, illustrated through innovativeness, proactiveness, risk-taking, autonomy and competitive aggressiveness. The research question to which answers were sought was: ‘How does an entrepreneurial orientation influence radical innovation in existing organizations?’ Through case studies in five firms involved in radical innovation, factors of an entrepreneurial orientation were discussed in order to understand which influence these had had on the development of the radical innovation. It was found that a firm’s emphasis on proactiveness and risk-taking played a very influential role in the development of radical innovation. Proactive behaviour was either seen in relation to a general focus on continuous innovation as a means of growth, or in relation to active anticipation of future problems in existing markets, and involved exploration of new opportunities in terms of new markets, new possibilities of applications, or new technologies. Openness towards change and flexibility in all facets of the organization, from resource allocation to technology and administrative systems, were found to facilitate the proactive behaviour and risk-taking. Seemingly high-risk ventures were often undertaken, and as the firms learned how to manage the radical innovation process the © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

369

actual risk-taking was diminished. Risk-taking thus became more a matter of risk management. Emphasizing autonomy was also found to be an essential contributor to the development of radical innovation in the cases. Autonomy was seen in the areas of the organizational framing of the radical innovation, the focus on the individual, and the management styles. Spin-off ventures were in several cases created to establish a highly flexible and autonomous frame for the innovation to be fully exploited. A focus on independent individuals allowed for decentralization through self-management and responsibility given to the individual/ group of individuals. As such, autonomy was also reflected at the level of management, where resource allocation, freedom to explore and positive attention were significant factors. Discussing the proposed linkage between competitive aggressiveness and the development of radical innovation, it was found that only a few elements of a competitive aggressive strategy were true in the cases. The strategies followed reflected responsiveness, a tendency to be first-to-market, and a willingness to be unconventional. However, the propensity of the firms to analyse and target a competitor’s weaknesses was not very influential. It was found that rather than relating to competitors, trends and demands in already existing markets, the case firms were engaged in creating new markets, that is, displacing old competitors through the creation of new competitive arenas, where no direct competitors were yet to be found. Through discussion of the empirical case studies it was hence found that the entrepreneurial orientation of the firm is highly influential on the development of radical innovation, and three propositions on this relationship were formulated: 1. A firm emphasizing an entrepreneurial orientation in terms of pro-activeness and risktaking stimulates radical innovation 2. A firm emphasizing an entrepreneurial orientation in terms of autonomy stimulates radical innovation 3. A firm which does not emphasize an entrepreneurial orientation in terms of competitive aggressiveness has increased likelihood of stimulating radical innovation. Evidence of the nexus between the constructs of corporate entrepreneurship and radical innovation was also found, providing reasons to continue the exploration of this area, in order to create a more solid understanding of the interrelatedness as well as the differences between the fields of entrepreneurship and innovation.

Volume 15

Number 4

2006

370


References Abetti, Pier A. (2000) Critical success factors for radical technological innovation: a five case study, Creativity and Innovation Management, 9(4), 208– 21. Antoncic, B. and Hisrich, R.D. (2003) Clarifying the intrapreneurship concept, Journal of Small Business and Enterprise Development, 10(1), 18–25. Baden-Fuller, C. and Pitt, M. (1996) Strategic Innovation, Routledge, London. Baird, I.S. and Thomas, H. (1985) Toward a contingency model of strategic risk taking. Academy of Management Review, 10, 230–43. Baum, J.R., Locke, E.A. and Smith, K.G. (2001) A multidimensional model of venture growth, Academy of Management Journal, 44(2), 292–303. Bessant, J. (2003) High-Involvement Innovation: Building and Sustaining Competitive Advantage Through Continuous Change, Wiley, Chihcester. Burgelman, R.A. (1983) A process model of internal corporate venturing in the diversified major firm, Administrative Science Quarterly, 28(2), 39–55. Burns, T. and Stalker, G.M. (1961) ‘Mechanistic and organic systems’, in The Management of Innovation, Tavistock Publications, London, pp. 119–25. Caffyn, S. (1999) Development of a continuous improvement self-assessment tool, International Journal of Operations and Production Management, 19(11), 1138–53. Chandy, R.K., Prabhu, J. and Antia, K. (2003) What will the future bring? Dominance, technology expectations, and radical innovation, Journal of Marketing, 67(3), 1–18. Chesbrough, H.W. (2003) The logic of open innovation: managing intellectual property, California Management Review, 45(3), 33–58. Christensen, Clayton M. (1997) The Innovators Dilemma: When new technologies cause great firms to fail, Harvard Business School Press, Boston, MA. Covin, J.G. and Slevin, D.P. (1986) ‘The development and testing of an organizational-level entrepreneurship scale’, in Ronstadt R., Hornaday, J., Peterson, R. and Vesper, K.H. (eds.), Frontiers of Entrepreneurship Research, Babson College, Wellesley, MA, pp. 628–39. Covin, J.G. and Slevin, D.P. (1989) Strategic management of small firms in hostile and benign environments, Strategic Management Journal, 10, 75–87. Covin. J.G. and Slevin, D.P. (1991) A conceptual model of entrepreneurship as firm behavior, Entrepreneurship: Theory and Practice, 16(1), 7– 24. Darroch, J. and McNaughton, R. (2002) Examining the link between knowledge management practices and types of innovation, Journal of Intellectual Capital, 3(3), 210–22. Dess, G.G., Ireland, R.D., Zahra, S.A., Floyd, S.W., Janney, J.J. and Lane, P.J. (2003) Emerging issues in corporate entrepreneurship, Journal of Management, 29(3), 351–78. Drucker, P.F. (1985) Innovation and Entrepreneurship, Harper & Row, New York. Drucker, P.F. (1988) The Coming of the New Organization, Harvard Business Review, Jan–Feb, 45–53.

Volume 15

Number 4

2006

Gilbert, C. and Bower, J.L. (2002) Disruptive Change – When Trying Harder is Part of the Problem, Harvard Business Review, Spotlight: Practical Strategy, May, 94–101. Green, S., Gavin, M. and Smith, L. (1995) Assessing a multidimensional measure of radical technological innovation, IEEE Transactions on Engineering Management, 42(3), 203–14. Hamel, Gary (2000) Leading the Revolution, HBS Press (e-book). Kanter, R.M. (1984) The Change Masterss, Touchstone, Simon & Schuster, New York. Kim, W.C. and Mauborgne, R. (2004) Blue Ocean Strategy: How to Create Uncontested Market Space and Make Competition Irrelevant, HBS Press (e-book). Kimberly, J.R. (1981) ‘Managerial Innovation’, in Nystrom, P.C. and Starbuck, W.H. (eds.), Handbook of Organizational Design, vol. 1, pp. 84–104, Oxford University Press, New York. Kirchhoff, B. (1991) Entrepreneurship’s contribution to economics, Entrepreneurship Theory and Practice, 16(2), 93–112. Kuratko, D.F., Hornsby, J.S., Naffziger, D.W. and Montagno, R.V. (1993) Implement entrepreneurial thinking in established organizations, SAM Advanced Management Journal, 58(1), 28– 33, 39. Leifer, R., O’Connor, G.C. and McDermott, C.M. (2000) Radical Innovation – How Mature Companies can Outsmart Upstarts, Harvard Business School Press. Lieberman, M. and Montgomery, D. (1988) Firstmover advantages, Strategic Management Journal (Special Issue), 9, 41–58. Lumpkin, G.T. and Dess, G.G. (1996) Clarifying the entrepreneurial orientation construct and linking it to performance, Academy of Management Review, 21(3), 135–72. Lynn G.S., Morone, J.G. and Paulson, A.S. (1996) Marketing and discontinuous innovation: the probe and learn process, California Management Review, 38(3), 8–37. MacMillan, I.C. and Jones, P.E. (1984) Designing organizations to compete, Journal of Business Strategy, 4(4), 11–26. March, J.G. (1991) Exploration and exploitation in organisational learning, Organization Science, 2(1), 71–87. McDermott, C.M. and O’Connor, G.C. (2002) Managing radical innovation: an overview of emergent strategy issues, Journal of Product Innovation Management, 19, 424–38. Mikkelsen, H. and Riis, J.O. (2005) Ledelse af projektmylderet, Børsens Forlag, Copenhagen. Miles, R.E. and Snow, C.C. (1978) Organizational Strategy, Structure, and Process, McGraw-Hill, New York. Miller, D. (1983) The correlates of entrepreneurship in three types of firms, Management Science, 29(7), 770–91. Miller, D. and Friesen, P.H. (1978) Archetypes of strategy formulation, Management Science, 24, 921–33. Miller, D. and Friesen, P.H. (1983) Strategy-making and environment, Strategic Management Journal, 4(3), 221–31. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


Morris, M.H. and Paul, G.W. (1987) The relationship between entrepreneurship and marketing in established firms, Journal of Business Venturing, 2(3), 247–59. Naman, J.L. and Slevin, D.P. (1993) Entrepreneurship and the concept of fit: A model and empirical tests, Strategic Management Journal, 14, 137–53. Nonaka, I. (1991) The knowledge-creating company, Harvard Business Review, 79(1), 96–104. O’Connor, G. and Ayers, A.D. (2005) Building a radical innovation competency, Industrial Research Institute, Inc., Journal of Research and Technology Management, 48(1), 309–34. Osborne, R.L. (1995) The essence of entrepreneurial success, Management Decision, 33(7), 4–9. Pinchot, G. (1985) Intrapreneuring: Why you Don’t Have to Leave the Corporation to Become an Entrepreneur, Harper & Row, New York. Porter, M.E. (1985) Competitive Advantage, Free Press, New York. Quinn, J.B. (1979) Technological innovation, entrepreneurship, and strategy, Sloan Management Review, 20(3), 15–23. Rule, E.G. and Irwin, D.W. (1988) Fostering intrapreneurship: the new competitive edge, Journal of Business Strategy, 9(3), 44–47. Sarasvathy, Saras D. (2001) What makes entrepreneurs entrepreneurial? submitted to Harvard Business Review. Schafer, D.S. (1990) Level of entrepreneurship and scanning source usage in very small businesses, Entrepreneurship Theory and Practice, 15(2), 19– 31. Schollhammer, H. (1981) The efficacy of internal corporate entrepreneurship strategies, Frontiers of Entrepreneurship Research, Babson College, Wellesley, MA, pp. 451–6. Schumpeter, J. (1934) The Theory of Economic Development, Harvard University Press, Cambridge, MA. Senge, P. (1994) The Fifth Discipline: The Art and Practice of Learning Organization, Doubleday Currency, New York. Simon, E.S., McKeough, D.T., Ayers, A.D., Rinehart, E. and Alexia, B. (2003) How do you best organize for radical innovation? Research-Technology Management, 46(5), 17–20. Sorescu, A.B., Chandy, R.K. and Prabhu, J.C. (2003), Sources and financial consequences of radical innovation: insights from pharmaceuticals, Journal of Marketing, 67(4), 82–102. Stevenson, H.H. and Jarillo, J.C. (1990) A paradigm of entrepreneurship: Entrepreneurial management, Strategic Management Journal, 11, 17– 27. Stopford, J.M. and Baden-Fuller, C. (1994) Creating corporate entrepreneurship, Strategic Management Journal, 15(7), 521–36. Tidd, J., Bessant, J. and Pavitt, K. (1997) Managing Innovation – Integrating Technological, Market and Organizational Change, 1st edn, John Wiley and Sons, Chichester. Tidd, J., Bessant, J. and Pavitt, K. (2005) Managing Innovation – Integrating Technological, Market and Organizational Change, 3rd edn, John Wiley and Sons, Chichester. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

371

Tushman, M.L. and anderson, P. (1986) Technological discontinuities and organizational environments, Administrative Science Quarterly, 31(3), 439–65. Yin, R.K. (1994) Case Study Research: Design and Methods, Sage Publications, London. Zahra, S.A. (1991) Predictors and financial outcomes of corporate entrepreneurship: an exploratory study, Journal of Business Venturing, 6(4), 259–85. Zahra, S.A. (1993) A conceptual model of entrepreneurship as firm behavior: A critique and extension, Entrepreneurship: Theory and Practice, 17(4), 5–21.

Astrid Heidemann Lassen ([email protected]) is currently in the process of completing her PhD at the Center for Industrial Production, Aalborg University, Denmark. Her research interests lie within the field of corporate entrepreneurship with a strong emphasis on the intersections with radical innovation, knowledge and strategic management. Astrid Heidemann Lassen has presented papers on this matter at several international conferences, has written book contributions, and publications in international journals are forthcoming. Astrid Heidemann Lassen teaches corporate entrepreneurship at Masters level. She is also President of the PhD association at the Faculty of Engineering, Science and Medicine at Aalborg University, member of the Board of the International Doctoral School at AAU, and National Coordinator for the Danish National PhD Network. Frank Gertsen (fgertsen@production. aau.dk) is Professor of Innovation Management at the Center for Industrial Production at Aalborg University in Denmark. He is head of the PhD program in Mechanical Engineering at the Faculty of Engineering, Science and Medicine. He is on the interim board of Center for Innovation and Product Development (CIPU). He is member of the CINet, the IMSS, the Academy of Management, and the EurOMA. He has been a visiting scholar at Stanford University and Hong Kong City University. He has edited several international journals and is on the editorial board of Creativity and Innovation Management. He has written more than 100 research papers and book contributions. His research interest is in the area of innovation and change and his current project involvement includes Discontinuous Innovation Project, Continuous Innovation, Radical Corporate Entrepreneurship project. He currently teaches innovation management at PhD, masters, and executive MBA levels.

Volume 15

Number 4

2006

372


Jens Ove Riis ([email protected]) is Professor of Industrial Management Systems, Aalborg University, Denmark. Professor Riis holds a Masters Degree in Mechanical Engineering with specialization in Industrial Engineering from the Technical University of Denmark (1964), and a PhD in Operations Research from the University of Pennsylvania, USA (1968). Professor Riis has published articles and books in the field of Project Management, Industrial Management, Design of Production Management Systems, and Management of Technology, and is now Deputy Director of a new research Center for Industrial Production at Aalborg University. From its start, Professor Riis has been the director of the post-experience Masters Program on Management of Technology that emphasizes the integration of business development, appropriate technology development and application, and organizational design and development.

Volume 15

Number 4

2006


COMPETENCE-BASED MODEL OF INITIATIVES

373

A Competence-Based Model of Initiatives for Innovations Katrin Talke, Sören Salomo and Nils Mensel The initiation of the innovation process has recently produced increased interest among the new product development scholars. While most extant research has focused on the management of opportunity recognition, of the development of new product concepts and of concept selection, little research has examined the initiative emergence process. With this article the authors intend to shed some light on the starting point of initiative formation by outlining a structured and comprehensive concept of the initiative emergence process; investigating individual competences facilitating initiative emergence, including task-related, action-related and cognitive competences; and developing a competence-based model for helping explain the occurrence of initiatives. For future research, the authors develop propositions on the impact that different competency constellations have on initiatives along their emergence process, and on factors that determine how to support competences for initiatives adequately.

Introduction

I

n innovation management research a vast number of studies have been published exploring how to design the innovation process optimally. Contrary to later phases, the initiation of the innovation process has only recently received considerable research attention (Reid & de Brentani, 2004, p. 171). Research regularly takes for granted the initial idea leading to the development of a new product (Zaltman, Duncan & Holbeck, 1984, p. 59). Literature dealing with the fuzzy front end mostly concerns aspects on the management of opportunity recognition, of the development of new product concepts and of concept selection (Kim & Wilemon, 2002; Khurana & Rosenthal, 1997; Moenaert et al., 1995; Reid & de Brentani, 2004; Reinertsen, 1999; Rice et al., 2001). Thereby, innovation management, and the management of the fuzzy front end in particular, is mostly reduced to the question of how novelties can be managed and diffused both within the innovating firm and the market. This view neglects the fact that managing innovations is not a mere problem of diffusion, but also of emergence (Gemünden, 2001, p. 413). Thus, an important gap is left regarding the emergence of initiatives. With this article we intend to shed some light on the starting point of initiative formation by examining the © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

initiative emergence process. Building on this process, we investigate competences facilitating initiative formation and we develop a competence-based model for initiatives. We also suggest ways to improve competences for initiatives according to individual and initiative process requirements. In doing so, we aim at providing a better understanding of the emergence of initiatives for both academics and practitioners.

Relevance of Initiatives and Research Questions The idea-development and idea-selection stages are often referred to as the ‘fuzzy front end’ of new product development (NPD) (Cooper & Kleinschmidt, 1986; Khurana & Rosenthal, 1998; Smith & Reinertsen, 1991). This characterization already implies difficulties in managing the initiative phase with traditional co-ordination instruments such as budgets, targets or time schedules (Smith & Reinertsen, 1991, pp. 43ff). The term ‘initiative’ stands for the impulse, stimulating activities that lead to new or improved products, processes or services. Thus, the initiative phase is a fundamental stage of the overall innovation process, as both the innovation is triggered and important parameters are set for a successful NPD process.

Volume 15

Number 4

2006

doi:10.1111/j.1467-8691.2006.00402.x

374


The relevance of initiatives manifests in several respects:

Volume 15

tive phase by illustrating relevant sub-phases, and describing management activities within these sub-phases, such as aspects of the management of opportunity recognition, of the development of new product concepts and of concept selection (Khurana and Rosenthal 1997; Kim and Wilemon 2002; Massey, Montoya-Weiss and O’Driscoll, 2002; Moenaert et al., 1995; Reid and de Brentani 2004; Reinertsen 1999; Rice et al. 2001). Yet, the question of which individual predispositions are requisite for initiative emergence has not been addressed in the literature so far. This question is of specific relevance as individual competences can be seen as background factors determining the outcome of any management activity, such as improving communication (Moenaert et al., 1995), speeding up decisionmaking (Reinertsen, 1999) or improving problem solving (Massey, Montoya-Weiss & O’Driscoll, 2002), conducted in the initiative phase. Responding to this research need, we intend to contribute to the innovation management literature by providing insights into the process of initiative emergence, individual competences relevant for initiative emergence and by proposing ways of supporting initiative capacity by enhancing competences for initiatives. Therefore, building on existing research we commence defining initiatives along their specific properties. These elements are relevant for understanding the process of initiative emergence and thus for managing this critical starting point of each innovation process. Next, we suggest a comprehensive conceptual framework for understanding individual competences fostering initiative emergence. This competence-based initiative model integrates different research directions including creativity concepts, motivation theory and individual competence concepts. Building on this model, we then suggest ways to improve competences for initiatives according to individual and initiative process requirements. We conclude by discussing the implications for theory and practice and identify avenues for future research.

(a) initiatives are requisite for innovation emergence. Hence, as performance needs innovation (Foster & Kaplan, 2001), initiatives become a precondition for sustainable corporate performance; (b) deficiencies in idea development and idea selection are dominant factors explaining innovation failure (Geschka & SchwarzGeschka, 2000, p. 102; Khurana & Rosenthal, 1998, p. 57). When deficient ideas serve as basis for the R&D process there is little chance of the outcome leading to technical or commercial success. Even technically good products founded on ideas irrelevant for the market will probably cause missed opportunities, wasted resources and a loss in credibility (Rubenstein, 1994, p. 653); (c) initiatives determine to a large extent the characteristics of innovations (Salomo & Mensel, 2001). Ground-breaking ideas will presumably lead to radical innovations (Reid & de Brentani, 2004, p. 177), ideas brought up by lead users or other market parties will probably lead to more marketdriven innovations compared to ideas developed by scientific research partners (Mensel, 2004, p. 110); (d) initiatives create the input to the innovation portfolio of firms. Hence, they determine the deal flow. Without a sufficient number of different initiatives, no adequate risk-return spread in innovation portfolios becomes possible (Cooper, Edgett & Kleinschmidt, 1999, p. 335). Thus, firms depend on numerous initiatives that vary in their characteristics in order to develop successful innovation portfolios, (e) because of the increasing innovation speed an effective and efficient initiative phase is gaining importance for corporate success (Gupta, Brockhoff & Weisefeld, 1992; Smith & Reinertsen, 1991, pp. 43ff). Research shows that the initiative phase on average may take as long as the subsequent decision process (Gemünden, 2001, p. 414), so that precious time elapses before organizations start developing the actual innovation (Gupta & Wilemon, 1990).

Defining Initiatives: Elements and Initiative Process

From these arguments it can be concluded that the success of innovation, both in terms of its development within the firms and its market acceptance, is to a substantial extent determined during the initiative phase and the resulting type of initiative. Existent fuzzy front-end research concentrates primarily on the anatomy of the initia-

Initiatives represent the initial impulse for further innovation activities (Birkinshaw, 2000; Hauschildt 1969, p. 734). Stimuli are required for tasks occurring irregularly without an organizationally determined starting point (O’Connor & Rice, 2001, p. 103). In most cases initiatives emerge from creative efforts of a single person, stemming from an active, self-

Number 4

2006



authorized behaviour, exceeding the formally expected work efforts (Frese et al., 1996, p. 38). As initiatives encompass a multitude of decisions on the topics of idea-generation and idea-selection, they are typically seen as a separate phase of the innovation process (Koen et al., 2001, p. 46). The innovation process is started as soon as the initiative is officially announced in the firm. This goes along with resources being more or less officially allocated to solving remaining issues of uncertainty reduction in terms of further pursuing the innovation project (Kim & Wilemon 2002, p. 271). The requirements for the emergence of initiatives can be described by means of five criteria (Hauschildt, 2004, p. 291 ff.). The first criterion concerns the dependence of an initiative to an individual: ‘Opportunity recognition for (radical) innovation is highly dependent on individual initiative and capacity, rather than routine practices and procedures of the firm’ (O’Connor & Rice, 2001, p. 103). This implies that the initiator’s curiosity, sensitivity, attention, power of observation and perception determines whrether or not an initiative is taken, correctly evaluated and announced in a suitable manner (Hauschildt, 2004, p. 292). In many cases, groups are referred to as initiators of an idea. While mostly one group member can be identified as originator of the idea, others seize the suggestion and go on advancing it (Rice et al., 2001, p. 410). The idea is then modified, expanded and enhanced, so that in the end it cannot be attributed to a specific individual any more. Second, initiatives emerge when a discrepancy between expectation and reality is recognized. Generally, it is assumed that crises, discontentment, tensions or other forms of external pressure resulting from an actual situation or an anticipated development stimulate individuals to come up with new ideas (Van De Ven, 1986, p. 595). When a problem is observed in a given situation, a course of action is triggered aimed at finding a solution in order to eliminate the experienced discrepancy or to take advantage of it. As soon as the problem is resolved and the situation is perceived as satisfactory, the process ends (Baker et al., 1980, p. 36). An initiative can arise from an alteration of the situation, from the emergence of new alternatives, the setting of new targets or changes in existing targets (Koen et al., 2001, p. 47; Hauschildt, 2004, pp. 292 ff.). The third criterion is linked to the initiator’s will to take action. An initiative is started only if the perceived discrepancy between an actual and a desired situation stimulates the initiator’s will to commence activities aiming at closing the gap (Nijhof, Krabbendam & © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

375

Looise, 2002, p. 682). A feeling of responsibility is prerequisite for the initiator’s will to take action. In a narrow sense, the responsibility for starting an initiative is bound to the individual job description or mandate. A broader understanding of responsibility for initiatives allows employees to propose ideas that do not immediately fall under their sphere of competence (Hauschildt, 1969, p. 737). A fourth criterion represents the official announcement of the idea. Therefore, the idea has to be made known to an organizational unit with competences to allocate resources for further pursuing the new project. Thereby, the idea is taken from the individual to the organizational level. If the initiator has the means to test his idea, the recognition of a problem and its official announcement can happen at different points in time. The particular case in which employees engage in unscheduled, unauthorized R&D activities, is often called ‘bootlegging’ (Augsdorfer, 1996, pp. 19 ff.; Knight, 1967, p. 493). Though this action initiates the problem-solving process, a ‘real’ organizational initiative is not started, as the official announcement to an authorized organizational unit is omitted. The fifth element concerns the end of the initiative process, which is manifested in a decision about the continuation of the initiative. Again, this decision is made by an organizational unit with competences to assign resources for further pursuing the new project (Hauschildt, 2004, p. 225). In a selection process alternative initiatives are evaluated and the one that matches the given criteria best is singled out. Carefully filtering the existing initiatives is of central importance, as not every initiative represents a valid alternative for the respective organization (Rice et al., 2001, p. 414 ff.; Singh, 2000, p. 18;). Only initiatives matching the firm’s strategy and resources should then be pursued (Cooper & Kleinschmidt, 1995). Therefore, initiatives are always linked with an evaluation and selection process, where the ‘raw material’ becomes an opportunity. When the process is successfully run through, resources are assigned and the initiative is realized within the scope of a project. These five constituting elements can be seen as subsequent process steps. By defining initiatives as a process, a better understanding of the starting and end point of the ‘fuzzy front end’ can be obtained. While the beginning seems to be easy to define but difficult to observe, the end seems to be difficult to define but more easy to observe. Considering the five elements will help to clarify the organizational requirements for the emergence of initiatives. Moreover, they provide a basis for a subse-

Volume 15

Number 4

2006

376


quent description of essential competences for initiators. By means of these necessary competences organizational measures can be determined, which enhance the existing competences and thereby improve the number and quality of initiatives.

Competences for Initiatives As initiatives are always connected to individuals (Hauschildt, 2004, p. 292) personal competences relevant for the emergence of new ideas can be defined (Bharadwaj & Menon, 2000, p. 425). A variety of competence models can be found in management literature (Cheetham & Chivers, 1996; Mansfield & Mathews, 1985; Salomo, Brinckmann & Gemünden, 2004; Schon, 1983; Schroder, 1989). But none of these models explicitly targets the initiative phase of the innovation process. Acknowledging the basic idea of these models that occurrence and nature of behaviour is dependent on individual competences, we aim at developing a comprehensive competence model that is adequate for the initiative phase. We refer to models of creativity, because a recent study by Mensel (2004, pp. 70 ff.) implies the existence of strong analogies between creativity and initiative. In socio-psychological literature creativity is seen as the production of novel, useful ideas or problem solutions, which refers to the process of idea generation or problem solving as much as to the actual idea or solution (Amabile, 1983; Sternberg, 1988). Following Basadur (1995, p. 63) ‘you can’t solve a problem without finding a problem first’ and finding the problem by posing the right questions is valued by Henle (1974, p. 23) as ‘the most creative part of the whole process’. As problem recognition and problem solving are central elements for the emergence of initiatives, an orientation on creativity models seems to be appropriate for framing competences for initiatives. Models containing parameters of creativity are provided by Jones (1987) and Lovelace (1986). Jones’ model builds on information processing theory, explaining individual creativity as combination of short-term memory information and long-term memory strategies. Perception barriers serve as an input filter, while values and self-image work as an output filter (Jones, 1987, p. 5). Jones’ model focuses on personal factors, while contextual factors are not explicitly regarded. Still, Jones (1987, p. 3) considers the possibility of lowering barriers by individual training measures. Lovelace’s model builds on motivation theory and regards creativity as function of motivation

Volume 15

Number 4

2006

and ability (Lovelace, 1986, p. 166). He assumes that managerial and extra organizational interventions determine how well individual needs are fulfilled, and thus the level of motivation. While Lovelace considers motivational issues in detail (1986, pp. 168 ff.), ability is not further specified and, hence, treated as constant. Both models focus on personal factors responsible for the emergence of creativity, but omit the creative process as such, so that it remains a ‘black box’. Models describing the creativity process along different stages are presented by Wallas (1926) and by Basadur, Graen and Green (1982). Basadur, Graen and Green (1982, p. 45) design a three-stage model, distinguishing between the phases of problem finding, problem solving and solution implementation. For each phase Basadur, Graen and Green (1982, p. 44) assume two sub-phases, where ideas are developed in an open, divergent process (ideation) and evaluated and selected in a convergent process (evaluation). Consequently, the authors propose that training measures should promote ideation and evaluation separately (Basadur, Graen & Green, 1982, p. 47). In an empirical study the effectiveness of such training measures could be largely confirmed (Basadur, Graen & Green 1982, p. 65). A socio-psychological model of creativity containing both parameters of creativity and a description of how these factors integrate into a creative process is presented by Amabile (1983, 1990, 1997). As Amabile’s model regards both aspects, it can be seen as integration of the predecessor models and as the most comprehensive model. Hence, we use Amabile’s approach as the basis for building our model. Amabile (1990, p. 82) describes the process of creativity emergence as encompassing five stages of problem recognition, of solution preparation, generation and evaluation, and of result assessment. She argues that the emergence of creativity in an organization is significantly influenced by the extent of creativityrelevant skills possessed by its employees (Amabile, 1988, p. 126). These skills include the three major components expertise, creativity skills and task motivation, which become relevant to a varying degree during the different stages of the creativity process. Amabile (1988, p. 137) proposes that each of these skills is necessary for some level of creativity to emerge, and the higher the level of each skill, the higher the overall level of creativity. In later works, Amabile (1997) suggests that for improving creative output, these creativityrelevant skills should be developed, sustained and enhanced through informal and formal mechanisms such as work environment, training and education. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

Initiative

ntio ce Ac eten mp

co

T co ask m -re pe la te te nc d e


Cognitive competence

Figure 1. Competences Critical for the Emergence of Initiatives

We use Amabile’s (1983, 1990, 1997) considerations concerning both the process perspective on creativity emergence and the individual factors facilitating creativity and transfer them to the initiative phase. Using the analogy to the creativity-relevant skills we suggest three competence fields critical for the emergence of initiatives: ‘action-competence for initiatives’, ‘task-related competence for initiatives’ and ‘cognitive competence for initiatives’. According to the assumption of Amabile’s creativity model, the simultaneous occurrence of all three competences forms the basis for the emergence of initiatives as visualized in Figure 1. The action competence for initiatives is regarded as the preliminary premise for initiating initiatives. Action competence is understood as a person’s will to take action, make decisions and take risk in a given situation (Jensen & Schnack, 1997, p. 165). The presence of a person’s will to take action finally determines whether anything will be done at all. The main elements constituting this competence are task-related motivation, commitment and visions (Fien & Skoien, 2002, p. 272; Salomo, 2000, p. 86). In socio-psychological research, the motivational element of actioncompetence is given specific attention. Taskrelated motivation is not only viewed as a trigger for any creative process. The presence of motivation is also seen as potential surrogate, to make up for deficiencies of the other two competences. It is argued that a highly motivated person is likely to draw skills from other domains, or to apply great effort to acquiring necessary skills in the target domain (Amabile, 1997, p. 44). Hence, initiatives will emerge given a sufficient level of motivation. Depending on the source, intrinsic and extrinsic motivation can be differentiated (Kruglanski, 1975, pp. 387 ff.). While intrinsic motivation has its source internally arising © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

377

from self-stimulation, extrinsic motivation is induced by external influences (De Charms, 1968, p. 328). The task-related competence of initiatives is the second requirement for initiatives. It comprises the know-how and abilities acquired for a specific task by experience, education or training (Katz, 1974; Salomo, 2000, p. 85). The task-related knowledge available to a person in a given situation is viewed by Amabile as ‘the set of cognitive pathways that may be followed for solving a given problem or doing a given task’ (1997, p. 42). Rarely, ideas emerge from a pure thought-process, but require a basic understanding of the respective subject in terms of expertise. Hence, the proposition that motivation or action competence may act as a potential surrogate for task-related competence may hold true for routine tasks. In the case of initiatives for innovation we suggest that some form of task competence must be available at the level of the initiating individual. Related to a specific task one distinguishes between strategic and operational knowledge (Armbrecht et al., 2001, p. 31). Strategic knowledge encompasses know-how about business objectives and plans relevant for developing long-term company programmes. Operational knowledge means functional expertise is important for framing a solution for the initial initiative’s problem. The cognitive competence is the third condition for the emergence of initiatives. This competence is the strongest representation of intellectual aspects required for initiative emergence (Messick, 1984, p. 125). It describes a person’s ability to think ‘off the beaten track’, to apply novel methods and techniques for problem exploration and to proceed in a systematic manner when evaluating different problem-solving options. Hence, the cognitive competence for initiatives determines the initiator’s potential to recognize a problem or a chance as well as the worthiness of changing the situation. Therefore, an initiator not only has to think creative but also to give thoughtprovoking impulses to accentuate problems that lead to initiatives for innovations. Essentially, a creative and an analytical cognitive style can be differentiated (Sadler-Smith, 1998, p. 191). By thinking creatively, information is processed broadly, with the risk of neglecting single elements of the problem. Thinking creatively allows a move from traditional to novel approaches (Kirton, 1976, pp. 622 ff.). By thinking analytically, information is processed in chunks, putting attention to the overall context in danger. This cognitive style is applicable for evaluating the economic and technological adequacy of the initiative (Sadler-Smith, 1998, p. 191).

Volume 15

Number 4

2006

378


Building on assumptions in socio-psychology, we suggest that the likelihood of an initiative emerging is maximized when action competence, task-related and cognitive competence for initiatives synchronously prevail (Amabile, 1983, 1990; Mensel, 2004, p. 322). Summing up these considerations, we put forward the following research proposition: P1: Simultaneous existence of action competence, task-related and cognitive competence for initiatives leads to the highest likelihood of initiatives occurring.

Competences for Initiatives and the Initiative Process For a better understanding of how the three competences for initiatives become relevant during initiative emergence, we add a process perspective to the static view. As a result of changing requirements during the course of the initiative phase the relevance of requisite competences is very likely to shift. We suggest that intrinsic motivation, operational knowhow and creative thinking are fundamental for early initiative generation, while for later initiative selection extrinsic motivation, strategic knowledge and an analytical cognitive style gain importance. The shifting relevance of competence configurations is shown in Figure 2. For the generation of innovative initiatives the initiator’s intrinsic, task-related motivation must be extensive, as the activity is not standardized (Hennessey & Amabile, 1988, p. 17). People are also most likely to produce innovative and useful ideas once they can identify with a task and exhibit a high degree of intrinsic motivation (Amabile, 1997, p. 44). For coming up with an innovative idea, ‘outof-the-box’ thinking is imperative and a creative cognitive style also brings forward the development of a comprehensive solution. In order to proficiently work out the idea’s

Generation of Initiatives

intrinsic

details the existence of operational knowledge becomes a necessary condition. This operational knowledge will typically encompass both technical proficiency and market-related expertise for the target work domain. The selection of initiatives does not only follow the results of technology focused feasibility studies. In addition, the innovation’s overall fit with business objectives and plans must be assessed, potential risks evaluated and budgets approved. Hence, for evaluating initiatives, strategic expertise gains importance (Massey, Montoya-Weiss & O’Driscoll, 2002, p. 40). Companies typically face a situation in which not just one idea, but rather a large number of ideas is proposed. Additionally, initiatives for innovations are characterized by an inherent complexity because of their newness to the company and, or to the market (Danneels & Kleinschmidt, 2001). Hence, the decision process of selecting the most promising initiative should proceed in a structured manner that makes analytic thinking specifically relevant. In the selection phase extrinsic motivation is beneficial because, unlike initiative generation, the activity is clearly defined (Amabile, 1993, p. 188). Moreover, for an objective evaluation a certain distance vis-à-vis the initiative is important. These reflections lead to the following research proposition: P2a: During the initiative process the relevance of requisite competence configurations is shifting. P2b: Effectiveness of initiative generation will improve with a specific competence configuration at the individual level, which is characterized by intrinsic motivation, operational know-how and a creative cognitive style. P2c: Effectiveness of initiative selection will improve with a specific competence configuration at the individual level, which is characterized by extrinsic motivation, strategic knowledge and an analytical cognitive style.

Generation Selection

extrinsic

Action-competence Action competence

operational

strategic

Task-related competence

creative

Selection of Initiatives

analytical

Cognitivecompetence

Figure 2. Process Perspective of Initiative Competences

Volume 15

Number 4

2006



We argue, as is commonly assumed, that all competences relevant for initiative emergence can be developed, sustained and enhanced through formal and informal organizational instruments, such as training and education. We have already suggested that all competences have to prevail simultaneously, each option at the adequate time, in order to maximize the likelihood of initiatives emerging. Drawing from Amabile’s model of creativity we further assume that the competence fields are multiplicatively connected. Therefore, the organizational effort for supporting initiatives will experience a non-linear relationship with the number of initiatives. Putting emphasis on a single competence will lead to decreased marginal benefits of these activities with regard to the number and quality of initiatives generated. Organizational instruments for systematically enhancing initiatives can be developed on the basis of this concept for initiative competences. Training the competences of potential initiators in a targeted manner will lead to increased initiatives, both in number and quality. Initiative competences can be developed through various organizational instruments, targeting the individual skills of potential initiators. In the following section we present a range of personnel development methods, proved suitable in practice.

Supporting Initiative Capacity by Enhancing Competences for Initiatives Our model of competence-based initiatives can be utilized for systematically enhancing initiatives for innovations in corporations. Instruments for developing initiatives can be classified according to their impact on the three competence fields. When describing the competences, we introduced alternative options that are of varying relevance for the initiative generation and selection phase. These alternatives can serve as starting point for organizational development methods. In what follows we present central instruments for initiative enhancement discussed in theory and practice. However, we do not intend to provide a complete enumeration, but suggest selected instruments, which have been proved by empirical research to be of specific relevance.

Action Competence for Initiatives Instruments for developing this competence can be differentiated if intrinsic or extrinsic © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

379

motivation is primarily targeted (Chen, Ford & Farris, 1999, p. 48). Empirical studies show that the intrinsic motivation of employees is encouraged best when the opportunity is given to work with competent colleagues, to operate in an intellectually appealing workplace, to work on challenging tasks, to initiate new activities and projects and to have the freedom to follow up their own ideas (Chen, Ford & Farris, 1999, p. 52). For extrinsic motivators a distinction between material and nonmaterial incentives can be made. Material incentives include both bonuses and non-cash benefits. Apart from fixed or variable payment forms, a motivational role is played by additional resources that help employees to better accomplish their job, for example, attendance to seminars, access to relevant literature, exchanging views with experts (Markham & Aiman-Smith, 2001, p. 49). Non-material incentives encompass instruments such as appreciation, praise, feedback, all acknowledging individual competences (Amabile, 1997, p. 45). Job guarantees and other measures securing the social status also fall under this category (Neubeiser, 1998, p. 130). In their empirical analysis Chen, Ford and Farris (1999, pp. 52 ff.) find that intrinsic motivators best support the effectiveness of innovation-oriented R&D employees. Transferred to the initiative phase it can be argued that specifically fostering intrinsic motivation, for example by demanding tasks with high degrees of freedom or collaborations with other experts, enhances quantity and quality of initiatives (Staudt, 1989, p. 367). Measures to boost intrinsic motivation produce a stronger effect for the generation of initiatives contrary to extrinsic motivators (Amabile, 1998, p. 79). For the initiative selection phase, extrinsic motivation gains relevance, particularly when the tasks are limited in time and clearly defined (Kohn, 1994, p. 16).

Task-Related Competence for Initiatives Problem recognition requires a combination of individual observation and knowledge. Thus, the likelihood of an initiative emerging depends on the potential initiator’s existing expertise. With knowledge management instruments, stimulating both the generation of new knowledge and the utilization of existing knowledge, task-related competence can be developed (Schmiedel-Blumenthal, 2001, p. 125). These instruments aim at conveying business objectives, extending task-related knowledge, providing search instruments for internal and external information, documenting experiences, results and new expertise,

Volume 15

Number 4

2006

380


and fostering a learning-oriented corporate culture (Armbrecht et al., 2001, pp. 328ff.). Basically, the probability for an initiative to emerge should increase with the initiator’s degree of task-related knowledge. Of relevance is not only in-depth expertise for the respective problem, but also knowledge elements suitable for coming up with new ideas for a possible solution. The more disconnected the combined knowledge elements, the higher the innovative potential of the resulting initiative (Mensel, 2004, p. 146). Knowledge management instruments to support this innovative potential of initiatives have to permit employees to access domain-unspecific knowledge, as in semi-formal knowledge networks such as communities of practice (Schneider, 2004; Wenger, 1998).

Cognitive Competence for Initiatives Generally, intellectual abilities of employees have to be regarded as a given condition. Changing the cognitive style from creative to analytical or vice versa will only work in the context of long-term training. By means of different techniques the individual problemsolving repertoire can be expanded so that the thinking process will exceed the barriers of the dominant cognitive style (Mensel, 2004, p. 162). Introducing creativity techniques will support the initiative-generation process (Osborn, 1966; Rickards, Aldridge & Gaston, 1988). Depending on the cognitive style to be supported, a distinction can be made between inductive and systematic-analytical techniques (McFadzean, 1998, p. 311). Systematic-analytical creativity techniques cope with problems by splitting them up into fragments. The aim is an improved problem comprehension, leading to the creation of new alternatives. Examples for these techniques are the morphological box (Zwicky, 1971) and systematic questioning techniques (Altshuller, 1984, pp. 151ff.). Inductive creativity techniques, such as brainstorming or brainwriting methods (Geschka & Reibnitz, 1993, pp. 7 ff.; Stein, 1974, p. 209), aim at finding new associations, analogies and alienations. The goal of these techniques consists in the generation of manifold ideas, based on the assumption that a growing number of ideas positively correlate with the probability of finding a good one. While each instrument will potentially make a contribution to high-quality initiatives, a systematic management of initiatives will consider all three initiative competences. Therefore, personnel development methods have to be adequately combined having regard to the specific corporate requirements, that is, the status quo of individual compe-

Volume 15

Number 4

2006

tences. Hence, a systematic management of initiatives has to set off with identifying potential initiator’s competence deficits to reveal the gaps to bridge. These thoughts lead to our final research propositions: P3a: A systematic management of initiatives demands a simultaneous support of all three initiative competences. P3b: The positive relation between competence support measures and initiative effectiveness depends on initiative process requirements. P3c: The positive relation between competence support measures and initiative effectiveness depends on corporate requirements in terms of status quo of individual initiative competences. To provide an illustration of a systematic management of initiatives, we will present the activities of a large company of the chemical industry, Hüls AG (Mensel, 2004, pp. 276 ff.). In order to enhance their initiatives, Hüls AG installed a taskforce. The taskforce’s mission was to develop business ideas to reduce the company’s dependence on the chemical industry’s economic cycle. Because of the long-term character of this task, a project group composed of internal and external experts was assigned to generate initiatives on a strategic level. In the following section we describe how the team members’ competences were developed during initiative generation and selection.

Initiative Generation As project members for the initiative generation phase, both executive employees that had previously shown a high willingness to perform and external experts with a reputation for their dedication were chosen. The members’ intrinsic motivation was encouraged by means of team-building seminars, targets without predefined activities and regular topmanagement involvement. The task-related competence for initiatives was already considered when forming the project group. Staff members possess a (natural-)scientific background and knowledge of business objectives, plans and potentials. External experts enrich the competence profile with managerial experience, methods expertise and analytic skills for structuring tasks. To further develop task-related competences, regular, intensive meetings with the managing board were established, helping to bridge competence gaps regarding strategic knowledge. In addition, the elaboration of initial ideas was delegated to specialized working groups. In weekly meetings these working © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


groups present their findings to the project group, where ideas are discussed, modified and combined to new ones. Specially created communication clusters assist the formation of informal knowledge networks between departments and working groups. By exchanging information between experts, strategic knowledge of business objectives and plans as well as operational knowledge is accrued. The cognitive competence for initiatives is also supported systematically. Creativity workshops stimulate the development of inventive ideas. In the context of systematic analyses of market, competition and technology trends these ideas are enriched with detailed knowledge.

Initiative Selection In the second phase the existing initiatives are systematically evaluated. Within the scope of a three-step ‘initiative-screening’ the pool of initiatives is condensed together with external experts. On each step the remaining initiatives are further elaborated and assessed for technical feasibility and strategic fit with business objectives. The managing board attends the last evaluation step. Based on defined criteria the elaborated initiatives are assessed to see whether they comply with initial targets. The whole decision process is characterized by an analytic procedure, where strategic aspects dominate. The granted incentives are mostly extrinsic in nature and focus on participation in profits realized by the innovation.

Discussion and Conclusion Our review of the literature dealing with the fuzzy front end reveals that insight is provided mostly on aspects of the management of existing initiatives but not on the question of how initiatives emerge (Kim & Wilemon, 2002; Khurana & Rosenthal, 1997; Moenaert et al., 1995; Reid & de Brentani, 2004; Reinertsen, 1999; Rice et al., 2001). As existing research largely concerns stages of the fuzzy front end following the official announcement of the initial idea, an important gap is also left concerning the process of initiative emergence, its conditions and its outcome. These shortcomings in the literature guided our research efforts, with the following intention: we aimed to shed light on the starting point of the formation of initiative by examining the initiative emergence process and competences facilitating initiative formation. Our research started with a suggestion of a structured and comprehensive concept of the © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

381

initiative emergence process as a critical part of the fuzzy front end of the innovation process. The initiative emergence process is closely related to the innovating individuals. Hence, we develop a competence-based model for explaining the occurrence of initiatives by drawing on literature from individual competence models, creativity and motivation theory. This model relates initiatives to individual competence arenas including task-related, action-related and cognitive competences. Following this model, we developed propositions as to how these competences affect initiatives along the initiative process and how different competency constellations may determine specific initiative output. This research offers academic value by providing a more comprehensive understanding of the emergence of initiatives and by developing a series of research propositions. However, an empirical examination of the proposed effects remains to be done, which is the central limitation of our research. This examination may be addressed in future research. For an empirical analysis of the competences facilitating the emergence of initiatives our research propositions could serve as a structured basis. However, we recognize that empirical testing will not be simple. In order to test propositions P1a, P2b and P2c, future research may consider employing an experimental design. Within an arranged set-up, the creative output of individuals with different competence profiles concerning a given task can be monitored. Dependent on the respective competence profile, one might also assess differences in the individual ability to generate or to select initiatives. Research propositions P3b and P3c could be examined within a longitudinal study. Following an assessment of the initial status, such a procedure would allow the effects of organizational measures designed to develop the three initiative competences in accordance to both initiative process requirements and individual requirements to be evaluated. This research can also offer practical value as we suggest possible avenues for the support of initiative competences. For companies aiming at improving their innovative abilities in terms of number and quality of initiatives, our model could facilitate a structured approach. As a start, the framework can be used for assessing the current status of initiative competences. It can also provide guidance for developing all competences required for the emergence of initiatives in a targeted manner. As we develop propositions regarding the effects of different competency constellations at different stages of the initiative process our concept may further allow adjusting corporate

Volume 15

Number 4

2006

382


activities within the fuzzy front end of the innovation process towards a more effective innovation portfolio.

References Altshuller, G. (1984) Creativity as an Exact Science. Gordon and Breach Science Publishers, London, UK. Amabile, T.M. (1983) The Social Psychology of Creativity, Springer, New York. Amabile, T.M. (1988) A Model of Creativity and Innovation in Organizations. Research and Organizational Behavior, 10, 123–67. Amabile, T.M. (1990) Within You, Without You: The Social Psychology of Creativity, and Beyond. In Runco, M.A. and Albert, R.S. (eds.), Theories of Creativity (pp. 61–91), Sage Publications, Newbury Park, California, USA. Amabile, T.M. (1993) Motivational Synergy: Toward New Conceptualizations of Intrinsic and Extrinsic Motivation in the Workplace. Human Resource Management Review, 3, 185–201. Amabile, T.M. (1997) Motivating Creativity in Organizations: On Doing What You Love And Loving What You Do. California Management Review, 40(1), 39–58. Amabile, T.M. (1998) How to Kill Creativity. Harvard Business Review, 78(5), 77–87. Armbrecht, F.M., Chapas, R.B., Chappelow, C.C., Farris, G.F., Friga, P.N., Hartz, C., McIlvaine, M.E. and Whitwell, G.E. (2001) Knowledge Management in Research and Development. Research Technology Management, 44(4), 28–48. Augsdorfer, P. (1996) Forbidden Fruit: an analysis of bootlegging, uncertainty, and learning in corporate R&D, Avebury, Aldershot, UK. Baker, N.R., Winkofsky, E.P., Langmeyer, L. and Sweeney, D.J. (1980) Idea Generation: A Procrustean Bed of Variables, Hypotheses, and Implications. Studies in the Management Sciences, 15, 33– 51. Basadur, M., Graen, G.B. and Green, S.G. (1982) Training in Creative Problem Solving: Effects on Ideation, and Problem Finding and Solving in an Industrial Research Organization. Organizational Behavior and Human Performance, 30, 41– 60. Basadur, M. (1995) The Power of Innovation, Pitman, London. Bharadwaj, S. and Menon, A. (2000) Making Innovation Happen in Organizations: Individual Creativity Mechanisms, Organizational Creativity Mechanisms or Both? Journal of Product Innovation Management, 17, 424–34. Birkinshaw, J. (2000) Entrepreneurship in the Global Firm, Sage Publications, Newbury Park, California, USA. Cheetham, G. and Chivers, G. (1996) Towards a Holistic Model of Professional Competence. Journal of European Industrial Training, 20(5), 20– 30. Chen, C.C., Ford, C.M. and Farris, G.F. (1999) Do Rewards Benefit the Organization? The Effects of Reward Types and the Perceptions of Diverse

Volume 15

Number 4

2006

R&D Professionals. IEEE Transactions on Engineering Management, 46(1), 47–55. Cooper, R.G. and Kleinschmidt, E.J. (1986) Benchmarking the Firm’s Critical Success Factors in New Product Development. Journal of Product Innovation Management, 3, 71–85. Cooper, R.G. and Kleinschmidt, E.J. (1995) New Product Performance: Keys to Success, Profitability and Cycle Time Reduction. Journal of Marketing Management, 24(5), 315–37. Cooper, R.G., Edgett, S.J. and Kleinschmidt, E.J. (1999) New Product Portfolio Management: Practices and Performance. Journal of Product Innovation Management, 16, 333–51. Danneels, E. and Kleinschmidt, E. (2001) Product Innovativeness From the Firm’s Perspective: Its Dimensions and Their Relation with Project Selection and Performance. Journal of Product Innovation Management, 18, 357–73. De Charms, R. (1968) Personal Causation. The Internal Affective Determinants of Behavior, Academic Press, New York. Fien, J. and Skoien, P. (2002) “I’m Learning How You Go About Stirring Things Up – in a Consultative Manner”: Social Capital and Action Competence in the Community Catchment Groups. Local Environment, 7(3), 269–82. Foster, R.N. and Kaplan, S. (2001) Creative Destruction. From ‘Built to Last’ to ‘Built to perform’, Doubleday, London, UK. Frese, M., Kring, W., Soose, A. and Zempel, J. (1996) Personal Initiative at Work: Differences between East and West Germany. Academy of Management Journal, 39(1), 37–63. Gemünden, H.G. (2001) Die Entstehung von Innovationen: eine Diskussion theoretischer Ansätze. In Hamel, W. and Gemünden, H.G. (eds.), Außergewöhnliche Entscheidungen: Festschrift für Jürgen Hauschildt (pp. 408–39), Vahlen, Munich, Germany. Geschka, H. and Reibnitz, U.v. (1993) Vademecum der Ideenfindung – Eine Anleitung zum Arbeiten mit Methoden der Ideenfindung. Battelle, Frankfurt, Germany. Geschka, H. and Schwarz-Geschka, M. (2000) Ideenmanagement und dessen Unterstützung durch eine Ideen- und Projektdatenbank. In Dold, E. and Gentsch, P. (eds.), Innovationsmanagement (pp. 99– 114), Luchterhand, Neuwied, Germany. Gupta, A.K., Brockhoff, K. and Weisefeld, U. (1992) Making Trade-Offs in the New Product Development Process: A German, US Comparison. Journal of Product Innovation Management, 9, 11–118. Gupta, A.K. and Wilemon, D. (1990) Accelerating the Development of Technology-based New Products. California Management Review, 32(2), 24–44. Hauschildt, J. (1969) ‘Initiative.’ In Grochla, E. (ed.), Handwörterbuch der Organisation (pp. 734–41), Schäffer-Poeschel, Stuttgart, Germany. Hauschildt, J. (2004) Innovationsmanagement, 3. Auflage, München. Henle, M. (1974) The Cognitive Approach: The Snail beneath the Shell. In: Rosner, S. and Abt, L.E. (eds.), Essays in Creativity (pp. 23–44), North River Press, Croton-on-Hudson, NY, USA. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


Hennessey, B.A. and Amabile, T.M. (1988) The Conditions of Creativity. In Sternberg, R.J. (ed.), The Nature of Creativity (pp. 11–38), Cambridge University Press, Cambridge. Jensen, B.B. and Schnack, K. (1997) The Action Competence Approach in Environmental Education. Environmental Education Research, 3(2), 163–78. Jones, L. (1987) Barriers to Creativity. Conference Paper, British Psychological Society Seminar on Stimulating Creativity, Manchester Business School (June 1987). Katz, R.L. (1974) Skills of the Effective Administrator. Harvard Business Review (9, 10), 90–102. Khurana, A. and Rosenthal, S.R. (1997) Integrating the Fuzzy Front End of New Product Development. Sloan Management Review, 38, 103–20. Khurana, A. and Rosenthal, S.R. (1998) Towards Holistic ‘Front Ends’ in New Product Development. Journal of Product Innovation Management, 15, 57–74. Kim, J. and Wilemon, D. (2002) Focusing on the Fuzzy Front End in New Product Development. R&D Management, 32(4), 269–79. Kirton, M.J. (1976) Adaptors and Innovators: A Description and Measure. Journal of Applied Psychology, 61(5), 622–9. Knight, K.E. (1967) A Descriptive Model of the Intra-Firm Innovation Process. Journal of Business, 40(4), 478–96. Koen, P., Ajamian, G., Burkart, R., Clamen, A., Davidson, J., D’Amore, R., Elkins, C., Herald, K., Incorvia, M., Johnson, A., Karol, R., Seibert, R., Slavejkov, A. and Wagner, K. (2001) Providing Clarity and Common Language to the Fuzzy Front End. Research Technology Management, 44(2), 46–56. Kohn, A. (1994) Warum Incentive-Systeme oft versagen; Incentive-Systeme psychologisch falsch fundiert. Harvard Business Manager, 16(2), 15–23. Kruglanski, A.W. (1975) The EndogenousExogenous Partition in Attribution Theory. Psychological Review, 82(6), 387–406. Lovelace, R.F. (1986) Stimulating Creativity through Managerial Intervention. R&D Management, 16(4), 161–74. Mansfield, R. and Mathews, D. (1985) Job Competence: A Description for Use in Vocational Education and Training. Further Education College, Blagdon. Markham, S.K. and Aiman-Smith, L. (2001) Product Champions: Truths, Myths and Management. Research Technology Management, 44(3), 44– 50. Massey, A.P., Montoya-Weiss, M.M. and O’Driscoll, T.M. (2002) Performance-Centred Design of Knowledge-Intensive Processes. Journal of Management Information Systems, 18(4), 37–59. McFadzean, E. (1998) Enhancing Creative Thinking within Organizations. Management Decision, 35(5), 309–15. Mensel, N. (2004) Organisierte Initiativen für Innovationen. Dentscher Universitätsveslag, Wiesbaden, Germany. Messick, S. (1984) The Nature of Cognitive Styles: Problems and Promise in Educational Practice. Educational Psychologist, 19(2), 59–74. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

383

Moenaert, R.K., De Meyer, A., Souder, W.E. and Deschoolmeester, D. (1995) R&D-Marketing Communications during the Fuzzy Front End. IEEE Transactions on Engineering Management, 42(3), 243–58. Neubeiser, A. (1998) Die Effizienz des Betrieblichen Vorschlagswesens. Hampp, Munich, Germany. Nijhof, A., Krabbendam, K. and Looise, J.C. (2002) Innovation Through Exemptions: Building upon the Existing Creativity of Employees. Technovation, 22, 675–83. O’Connor, G.C. and Rice, M.P. (2001) Opportunity Recognition and Breakthrough Innovation in Large Established Firms. California Management Review, 43(2), 95–116. Osborn, A.E. (1966) Applied Imagination – Principles and Procedures of Creative Problem-Solving. 3rd edition, Charles Scribner’s Sons, New York, USA. Reid, S.E. and de Brentani, U. (2004) The Fuzzy Front End of New Product Development for Discontinuous Innovations: A Theoretical Model. Journal of Product Innovation Management, 21, 170– 84. Reinertsen, D.G. (1999) Taking the Fuzziness Out of the Fuzzy Front End. Research Technology Management, 42, (November–December), 25–31. Rice, M.P., Kelley, D., Peters, L. and O’Connor, G. (2001) Radical Innovation: Triggering Initiation of Opportunity Recognition and Evaluation. R&D Management, 31(4), 409–20. Rickards, T., Aldridge, S. and Gaston, K. (1988) Factors Affecting Brainstorming towards the Development of Diagnostic Tools for Assessment of Creative Performance. R&D Management, 18, 309–20. Rubenstein, A.H. (1994) At the Front End of the R&D/Innovation Process: Idea Development and Entrepreneurship. International Journal of Technology Management, 9(5, 6/7), 652–77. Sadler-Smith, E. (1998) Cognitive Style: Some Human Resource Implications for Managers. International Journal of Human Resource Management, 9(1), 185–202. Salomo, S. (2000) Wechsel der Spitzenführungskraft und Unternehmenserfolg. Duncker & Humblot, Berlin, Germany. Salomo, S. and Mensel, N. (2001) Front End Idea Generation for Innovation: Empirical Evidence from German Industrial Corporations. In Kocaoglu, D.F. and Anderson, T.R. (eds.), Technology Management in the Knowledge Era (pp. 117–23), PICMET, Portland, USA. Salomo, S., Brinckmann, J. and Gemünden, H.G. (2004) Management Competence and Development Speed of Young High-Technology Companies, Proceedings to RENT 2004 Conference, Copenhagen, 25–26 November. Schmiedel-Blumenthal, P. (2001) Entwicklung eines ganzheitlichen Wissensmanagements zur erfolgreichen Umsetzung von industriellen Innovationen, Eul Verlag, Köln, Germany. Schneider, U. (2004) Communities of Practice. In Schreyögg, G. and v. Werder, A. (eds.), Handwörterbuch Unternehmensführung und Organisation (pp. 144–152), Schäffer-Poeschel, Stuttgart, Germany.

Volume 15

Number 4

2006

384


Schon, D. (1983) The Reflective Practitioner: How Professionals Think in Action. Basic Books, New York, USA. Schroder, H.M. (1989) Managerial Competence: The Key to Excellence. Kendall/Hunt Publishing, Iowa, USA. Singh, R.P. (2000) Entrepreneurial Opportunity Recognition through Social Networks. Garland Publishing, New York, USA. Smith, P.G. and Reinertsen, D.G. (1991) Developing Products in Half the Time. John Wiley & Sons, New York, USA. Staudt, E. (1989) Die innerbetriebliche Förderung von Innovation. Personal, 9, 364–69. Stein, M.I. (1974) Stimulating Creativity. Academic Press, New York, USA. Sternberg, R.J. (1988) The Nature of Creativity: Contemporary Psychological Perspectives. Cambridge University Press, Cambridge. Van De Ven, A.H. (1986) Central Problems in the Management of Innovation. Management Science, 32(5), 590–607. Wallas, G. (1926) The Art of Thought. Harcourt Brace Jovanovich, New York, USA. Wenger, E.C. (1998) Communities of Practice – Learning as a Social System. The Systems Thinker, 9(5), 1–5. Zaltman, G., Duncan, R. and Holbeck, J. (1984) Innovations & Organizations. Robert E. Krieger Publishing, Malabar, Florida, USA. Zwicky, F. (1971) Jeder ein Genie. Peter Lang, Bern, Germany.

Volume 15

Number 4

2006

Katrin Talke ([email protected]) is Assistant Professor of Technology and Innovation Management at Karl-Franzens University Graz, Austria. Her research interests focus on innovation marketing. She is also addressing questions surrounding the fuzzy front end of the innovation process and the strategic mindset of innovating firms. Sören Salomo (soeren.salomo@uni-graz. at) is a professor of technology and innovation management at Karl-Franzens University Graz, Austria. His research interests cover corporate innovation management from a resource based perspective with a special focus on process and organizational system mechanisms for supporting radical innovation. He is also active in the field of innovation marketing. Dr. Nils Mensel ([email protected]) is a business consultant in the pharmaceutical industry. He is based in Hamburg, Germany and has a special interest in the organized initiation of innovations and company communication strategies.


DECISION-MAKING FACTORS IN INNOVATION PROCESSES

385

Assessing the Importance of Factors Determining Decision-Making by Actors Involved in Innovation Processes Hans Heerkens Innovations can be seen as chains of non-routine decisions. With each decision, the innovator has to assess how important the various decision attributes are. Because the decisions are nonroutine, innovators cannot fall back on judgements of past importance. Most decision support methods elicit importance judgements but do not help innovators or other decision-makers with the mental processes leading to the judgment. The ‘importance assessment process’ can be divided into seven phases (such as (sub-)attribute processing and various forms of weighting). The phase ‘(sub)-attribute processing’ is the most important phase in terms of effort devoted to it, and the most obvious pitfalls that prevent valid importance assessments appear in this phase. This article describes some of these pitfalls. A few simple instruments may provide better-founded importance judgements that can be better communicated to other actors involved in innovation processes.

Introduction

T

his article is about a specific aspect of decision-making: the assessing of the importance of attributes (characteristics that describe the alternatives from which the choice is to be made). If you want to buy a new car, the cars available on the market (the alternatives) can be described in terms of their top speed, their price, their roominess and other attributes. Which car you buy not only depends on the score on these attributes (for example: how fast can car X go?), but also on their importance; sometimes referred to as their weights. If top speed is important to you, you may buy that expensive and cramped Ferrari. But if you have to take the kids to school each day, roominess may be more important than top speed, and you will buy a Volkswagen. In this article we address the question ‘how do people think when assessing importance of attributes?’. This thinking process is called ‘importance assessment’, and it results in an ‘importance judgement’: a weight assigned to each attribute relevant for the decision at hand. The decision process is visualized in Figure 1. In this article, we are only concerned with step two: the importance assessment. © 2006 The Author Journal compilation © 2006 Blackwell Publishing

We focus on the decision context that innovators often find themselves in, and show which pitfalls innovators face and how these can be negotiated. The most important phase in the importance assessment process, (sub)attribute processing, is the focus of this article because it offers the most possibilities for improvement. We start with characterizing the situation in which an innovator often finds himself, and explore the problems he faces concerning the weighting of attributes. Then we review earlier research, culminating in a model of the importance assessment process. Our own research identified some major pitfalls in the importance assessment process and as a conclusion of this article we provide suggestions for some simple instruments that may be of help to those involved in innovative decision processes. Note: when ‘he’ or ‘his’ is used, ‘he/she’ or ‘his/her’ is meant.

The Problem An innovator is often confronted with an array of decisions to take; choices to make. For example: which technical solution is the best

Volume 15

Number 4

2006

doi:10.1111/j.1467-8691.2006.00401.x

386


Setting the goal of the decision

I need a new car

Thinking about how important the various attributes are (importance assessment)

I need to drive the kids to school… but I always wanted to be a racing car driver

Setting the importance of the attributes (importance judgement)

The number of seats is more important than top speed

Finding out the attribute scores of the available options

A Ferrari can takes two persons, a Volkswagen is slower but takes five

Choosing the best option

I’ll buy the Volkswagen

Figure 1. The Place of Importance Assessment in the Decision Process for a given problem? Which of his colleagues in the organization should he let in on his ideas in order to gain support? Is it worthwhile to apply for a patent? Every choice has its pros and cons. Sometimes these are clear, like the advantages and disadvantages of technical solutions or the cost of patents and the legal protection they can give. In this article, we assume that the pros and cons pertaining to decisions are known. But even so, the relative importance of these pros and cons may not be known, either because they are a matter of personal preferences or because the situation in which the innovator finds himself does not have a precedent, so that past experience is of only limited value. Imagine, for example, that someone has invented a new model skateboard, and that type-A wheels give much less friction but are more prone to breaking off than wheels of type B. Which wheel should the inventor choose? This depends on how important performance (dependent on friction) is, relative to safety (if a wheel breaks off the skateboarder may be injured). In addition, replacing a wheel will incur extra cost. How important are these costs? With a new product, market research (finding out how important customers consider the various attributes to be) is difficult because the customers may not be able to give valid opinions. The innovator has to rely on his own judgement. There is one simple reason for the difficulty in judging the importance of attributes; the attributes cannot be directly compared to each other. So the innovator has to rely on his expert knowledge and experience to assess the importance of the various pros and cons of the available options (in our example the types of skateboard wheels). The chance that there is a standard procedure that guarantees an optimal decision is negligible.

Volume 15

Number 4

2006

All in all, innovators frequently face the task of assessing the importance of attributes that influence the choices that they have to make. This article is about the way people do this, and how they can do it better. We focus on one phase of the importance assessment process, as stated earlier, and on the challenges that innovators face in that phase. The problem is non-routine, no clear criteria exist for judging which solution is the best, and evading a choice or leaving it to someone else is not an option. Innovators, faced with choices for which no precedents exist, are inevitably confronted with the importance judgement question. Many will rely on intuition, experience or ‘gut feeling’. They may make the right choices, or they may not. In short, innovators: • are faced with non-routine decision, or ‘wicked problems’ (Rittel & Webber, 1973); • often have to take crucial decisions by themselves, unable to rely on importance assessments made by others. Even if they can ask others to evaluate their choices, they first have to make them themselves. Their importance assessment is, at least initially, theirs alone; • have a strong need to justify their decisions in order to overcome opposition to their ideas; • are sailing uncharted waters in which even their own expertise is limited. So, innovators often operate in a different environment from decision-makers confronted with more routine tasks. The latter can fall back on their own experience with similar decisions, can call in help from others and probably do not even have to make explicit importance judgements. They can use importance judgements made in earlier decisions. The innovator does not have this luxury, and weighting is a truly challenging task. His con© 2006 The Author Journal compilation © 2006 Blackwell Publishing


fidence could receive a major boost if he could understand the mental processes involved in importance assessment, and if simple but effective instruments could be developed to help with weighting. Two limitations of this article should be stressed. First, it is focused on individual actors and not on group decisions. Many business decisions are, at least, prepared by groups and not individuals. But innovators are often individualists. Once a decision is placed on the agenda of a designated group the innovator apparently managed to get it there, so in a sense his work is already done. But even in group decisions the participants have to decide individually on their position as input for the group process. And, even though a group may propose a choice, it may be up to an individual actor to make it. Our skateboard inventor may have obtained the advice of engineers and marketers on the configuration of his board, but as chief designer he has to make the final decision. The second limitation is that the creative element of importance assessment is not addressed directly. We will see that creativity certainly plays a role in importance assessment process, especially in the phase that we will look at, but space constraints prevent us from going into this area.

Theoretical Background Previous Research An important area of research in decision theory linked to importance assessment is the study of factors that influence the assessment process. Notable examples are: perception of risk and attitudes towards risk (Kahneman & Tversky’s (2000) Prospect Theory), the perspective of the decision-maker (Kray, 2000; Kray & Gonzalez, 1999), information presentation and usage (Guo, 2001; Russo, Medvec & Meloy, 1996), the concept of attribute weights (Keeney & Raiffa, 1976), factors that cause biases in weighting (Borcherding, Schmeer & Weber, 1995; Fischer, 1995; Póyhónen & Hámáláinen, 1998), the influence of unimportant or irrelevant attributes on choice (Barlas, 2003; Goldstein & Busemeyer, 1992; Hsee, 1995), group decision-making, particularly the relationship between individual and collective preferences (for example Hollingshead, 1996; Wei et al., 2000), the internal and predictive validity of various methods for measuring importance judgements, such as conjoint measurement (Harte & Koele, 1995; Jaccard, Brinberg & Ackerman, 1986) and the influence of regret aversion (Zeelenberg et al., 1996). Marketing research has focused on socio-economic © 2006 The Author Journal compilation © 2006 Blackwell Publishing

387

and demographic factors influencing the perceived importance of attributes, and Meehl (1954) studied the degree to which limited numbers of relevant illness symptoms occurring with patients influence doctors’ diagnosis. Further contributions regarding the explanatory power of regression models have been realized by Dawes (1979) and others. Keeney’s (1992, 1994) value-focused thinking approach shows how various instruments assessing value preferences of decisionmakers can be used to optimize non-routine decisions within an organizational context. The above-mentioned research provides us with some building blocks for modelling importance assessment processes, such as the concept of ‘importance’ and the relationship between attribute scores, weights and attractiveness of alternatives. But it either takes the weights actors assign for granted and concentrates on the decisions made on the basis of these weights, is concerned with eliciting the weights with sufficient validity (necessary for linking them with choices) or looks at the factors influencing the weights, such as perceptions of risk. The importance assessment process, that is, the way in which actors think while weighting, is addressed merely incidentally. So, our research covers the process of generating weights, while the above-mentioned research concerns the weights when they are, or have been, assigned and measured with methods such as conjoint measurement. There are other areas of research that are of relevance to importance assessment, such as problem-solving, human choice strategies and bounded rationality (Heerkens, 2003). We will devote no attention to them here, but instead go straight to the model that we developed of the importance assessment that actors go through in non-routine decisions. This model is explained in detail in Heerkens (2003) and Heerkens and Van der Heijden (2003). From the model we can derive the aspects that are of particular interest to innovators. This model was made for exactly the situation in which innovators find themselves, and which was described earlier. We will select specific issues from the model that are relevant for innovators. We do not aim to discuss the model or the research that generated it in depth; that is done elsewhere (Heerkens & Van der Heijden, 2003, 2005).

The Weight Assessment Model (WAM) The weight assessment model (WAM) consists of seven main phases and six auxiliary activities. The seven main phases are presented in a

Volume 15

Number 4

2006

388


Table 1. The phases of the WAM Phase

1. 2. 3. 4. 5. 6. 7.

Structuring cluster Structuring cluster Weighting cluster Weighting cluster Weighting cluster Weighting cluster Evaluation cluster

Phase name

Percentage of segments devoted to the phase

Problem identification (Sub-) attribute processing Absolute sub-attribute weighting Homogeneous sub-attribute weighting Heterogeneous sub-attribute weighting Attribute weighting Evaluation

6.74 30.33 27.22 4.53 1.50 12.54 17.14

sequential way in Table 1. In reality, actors may go back and forth between phases and are likely to address phases more than once. The auxiliary activities pertain to areas such as information search and planning and are not linked to particular phases. We do not deal with them in this article, but a brief description can be found in Appendix 1. Based on Simon’s (1960) distinction of the problem-solving process into a structuring and a solving phase, we divide the WAM in a structuring cluster (phases 1 and 2) and a weighting (solving) cluster (phases 3 to 6). Phase 7, the evaluation phase, will not be covered in this article. We confine ourselves to the weighting itself, not to its evaluation. In the structuring cluster, the problem (in this case the task to weigh attributes) is formulated (phase 1) and the attributes are processed in phase 2 so that they can be readily weighted in subsequent phases. Then, the weighting takes place in the weighting cluster. Since the model was conceived on the basis of think-aloud protocols, we are able to indicate which percentage of all the thoughts that the subjects expressed was devoted to which phase. We arrived at these percentages by dividing each protocol in segments (the smallest possible meaningful statements made by the subjects) and coding these according to a formal coding scheme. The total numbers of statements pertaining to each of the phases was divided by the gross total of statements for all subjects taken together. This is further explained in Heerkens and Van der Heijden (2003, 2005). We use these percentages as an indicator of the relative importance of each phase.

Phases of the WAM Phase 1: Problem Identification This phase consists of activities such as elaborating on (understanding, concretizing) the

Volume 15

Number 4

2006

task at hand and, if desired, re-formulating it in one’s own words. Essentially, this phase concerns defining, so to speak, the task lying ahead. This may mean, for example, stating the attributes to be weighted, or making boundary conditions explicit. In our research, the subjects had to weigh safety and comfort of a minibus (explained later in this article). This means that other potentially relevant attributes, such as fuel consumption, were to be ignored. The generation of attributes to be weighted may take place in this phase, but it is also possible that they were identified before and merely have to be made explicit.

Phase 2: (Sub-)Attribute Processing If one wants to weigh attributes, one should first know what one is weighting. Attributeprocessing concerns giving the attributes a more precise meaning. This can be seen as a case of framing (Akin, 1994). An example of attribute processing is: concretizing ‘safety of a minibus’ into ‘number of deaths per 10 million km’. Attribute properties such as measuring level, measuring unit, level of abstractness and precision can change as a result of processing. The following forms of processing were identified: • Decomposing. An attribute (‘safety’) can be split up in several sub-attributes (such as ‘quality of the brakes’ and seatbelts present yes/no); • Re-formulating. When an actor gives an attribute or sub-attribute a different name while meaning the same attribute with a similar, not necessarily identical, measurement unit, the attribute is re-formulated. For example, ‘safety’ can be re-formulated as ‘ensuring a safe journey’; • Concretizing a (sub-)attribute. An example was given above; © 2006 The Author Journal compilation © 2006 Blackwell Publishing


• Integrating (sub-)attributes into a new (sub)-attribute. For example: taking ‘adjustability of seats’ and ‘amount of legroom’ together as ‘quality of seating’; • Making an attribute more abstract. This is the complement of concretizing. Definition is not included in the types of processing. The result of processing may be a description of an attribute that is so exact and formal that it can be called a definition. This is the phase that we will study in some depth in this article. The next phases concern the actual weighting process.

Phase 3: Absolute (Sub)-Attribute Weighting With ‘absolute’ weighting (based on Timmermans, 1993), we mean that a statement about the importance of a (sub)-attribute is made without reference to the importance of other (sub)-attributes. For example: ‘safety is important’. This statement does not say how much more or less important it is than ‘comfort’.

Phase 4: Homogeneous Sub-Attribute Weighting This phase is the first in which ‘true’ weighting takes place: the balancing of the weight of one sub-attribute against that of another. We call this ‘relative weighting’ (based on Timmermans, 1993). In this phase, two or more subattributes of the same main attribute are weighted against each other, and arguments for the weighting are given. For example: ‘good seatbelts are more important than good brakes’ (both sub-attributes of ‘safety’).

Phase 5: Heterogeneous Sub-Attribute Weighting This phase differs in only one respect from the previous one: the sub-attributes that are weighted belong to different main attributes. For example: good seatbelts (sub-attribute of ‘safety’) are more important than comfortable seats (sub-attribute of ‘comfort’).

Phase 6: Attribute Weighting This phase concerned the integral weighting of the (in our case, two) main attributes. It is the essence of any weighting task. For example: safety is more important than comfort’.

Phase 7: Evaluation This phase comprises the reflections by subjects on their activities and the results. Several types of evaluation can be identified, such as the extent to which the assignment has been © 2006 The Author Journal compilation © 2006 Blackwell Publishing

389

fulfilled, evaluations of weight judgements (is the actor, on hindsight, satisfied with assigned weights) and evaluations of arguments (how good are the reasons for particular weight judgements). From this model, several areas can be identified that are of particular significance to innovators, because they present special challenges or potential pitfalls. In this article we focus on phase 2; the main structuring phase. If attributes are not correctly formulated, the resulting weighting is bound to be flawed. In addition, it is clear from the table that phase 2 is the most important phase in terms of effort devoted to it (30 percent of the segments in the think-aloud protocols pertain to this phase), which seems logical in the case of non-routine problems. Furthermore, when an innovator has to weigh attributes, it is almost inconceivable that the attributes are so clear that no processing is needed. After all, an innovator has to deal with ill-defined, non-routine problems. So, if we want to help innovators to improve their weighting of attributes in the difficult decisions they have to take, phase 2 seems to be a good place to start. (Sub-)attribute processing has, to our knowledge, not been addressed in the literature (the WAM was, after all, only designed a few years ago), so it is worthwhile to explore it. In the remainder of this article, we will address the following questions: 1. How do actors involved in weighting attributes in non-routine decisions conduct the attribute processing phase? 2. Which pitfalls can be identified that actors should be aware of? 3. In which way can their performance be improved? Obviously, we have to answer question 1 and 2 before we can hope to answer question 3.

The Research Method In order to gain insight in cognitive processes of our subjects, we used a think-aloud method. This is a good method for analysing cognitive processes (Ericsson & Simon, 1993). Methods such as choice experiments and process tracing show the results of cognitive processes, but not the processes themselves, while retrospective reporting methods, such as interviews and diaries, leave too much room for interpretation of the cognitive processes by the subjects themselves and are vulnerable to lapses of memory (Ericsson & Simon, 1993). As there is little research on importance assessment processes (see above), we do not

Volume 15

Number 4

2006

390


formulate elaborate hypothesis, but we describe the attribute processing phase on a number of dimensions and link these to existing theory whenever possible. As stated before, we will not address creativity. It is clear that creativity probably plays an important role in generating (sub-)attributes, especially in non-routine decisions where there are no established sets of (sub-)attributes available. In addition, finding integrative attributes covering sets of sub-attributes can be a creative activity. We will, however, look only at the types of processing actors employ, not at their source of inspiration. We will look at pitfalls in the process from a rational perspective, not at enhancing creative activities. We do this because of space constraints and in order to keep our research focused.

Sample and Assignment Eighteen undergraduate students of the University of Twente in The Netherlands were given an individual assignment based on a fictional case. University students might be assumed to have enough analytical abilities to perform the assignment satisfactorily, without having the knowledge and skills that would enable them to rely on previous experience of importance assessments. Hence, the danger that they give weights based on previously obtained knowledge is minimized. The assignment consisted of supporting the acquisition process of new minibuses by a local company. This decision may not be innovative, but it certainly is non-routine for a local (small) company. Such a decision occurs only every so many years, and choosing the wrong bus can have dire consequences (for example: low load factor if the bus is too big, loss of clients if it is uncomfortable and so on). The management and the drivers may have practical experience with the minibuses the company operates, but they may well not have intimate knowledge of the buses available on the market, and of the demands of (potential) clients. A non-routine decision like this is a good case for studying the importance assessment process. The subjects were asked to establish the importance of two characteristics of the to-beacquired minibuses vis-à-vis each other while thinking aloud, and were told that they would be advising the management team during the acquisition process. The management team would use the generated weights as a basis for evaluating the available minibuses, and then choose the type to buy. The management team did not have to agree on the weights generated, but the students should be able to

Volume 15

Number 4

2006

explain their reasons behind their importance judgements if so required by the management. Therefore, while the students were free in the way to reach their importance judgment, it was not a purely personal exercise but an activity within an organizational context. The attributes, ‘safety’ and ‘passenger comfort’, were chosen to prevent comparability by some readily available algorithm or heuristic or easy expression in a common denominator such as money. The information that was supplied included a brochure of the company, a leaflet explaining the decision context and two brochures on minibuses; one on a Volkswagen and one on an Opel. The latter enabled the subjects to get familiar with the specific capital good to be acquired. It was made clear that these examples of minibuses did not mean that the subjects had to make a choice between them. The students were given an example of weighting before starting the assignment, but were allowed to use their own concept of ‘weight’ or ‘importance’, just as would be the case in real life. It should be stressed that the assignment was geared to provide the optimal context for an importance assessment process. This means that there was no direct relationship with innovation. The context was kept as simple as possible, so as not to distract the subjects from their task. A purchasing context is much easier to grasp than an innovative context, where our subjects could not have been expected to generate the innovation themselves anyway. The relevant circumstances, however, were there: a non-routine assignment where the subjects could not rely on previous knowledge or experience and had to use their own creativity and cunning to complete the task. It does not matter whether the importance assessment had to be done for an acquisition process (like that in our assignment) or for, for example, the judgement of innovative alternative solutions to a problem (generated by someone else).

Procedure The respondents were asked to think aloud during the assessment process. The general guidelines for think-aloud studies given by Ericsson and Simon (1993) were followed, including a practice session to familiarize the subjects with the think-aloud strategy. All verbal information given by the respondent was recorded and typed out literally. After completion of the assignment, a short interview was conducted. In total, each session lasted for a maximum of two hours, for which the subjects were paid €20. Two pilot sessions were con© 2006 The Author Journal compilation © 2006 Blackwell Publishing


ducted, which led to some minor adjustments of the assignment. Two kinds of analyses have been performed using the literally typed out protocols: 1. A largely qualitative analysis according to the general rules of the ‘grounded theory’ approach (Strauss & Corbin, 1998). 2. A more quantitative analysis, based on a formal coding scheme that was designed on the basis of the qualitative analysis. Two coders performed the coding activities. Although they worked independently of each other, during the coding of the first six protocols weekly meetings were held to discuss general coding issues in order to enhance the reliability of it. The coders retrospectively applied the refinements to the coding scheme independently. The overall Cohen’s Kappa (Baarda & de Goede, 2001) for inter-rater consistence was 0.97 over a total number of verbal segments of 7,253.

Limitations It should be stressed that this is not quantitative research. It was not clear beforehand which variables would turn out to be relevant and how they should be defined. Hence we used an extreme case and no control variables. We wanted to know how actors having to make an importance assessment for a non-routine problem would behave and which pitfalls they could encounter: whether, for example, a more routine problem would induce a different behaviour was not of interest. If we can identify pitfalls that innovators have to be aware of because they have a reasonable chance of occurring (even if the exact chance is not known) or because innovators can easily recognize them in their individual activities, regardless of the chance of occurring, our aim is fulfilled. If we could have linked the occurrence of pitfalls, or the behaviour of actors making importance assessments, to the quality of the resulting weighting, then it would have been appropriate to control that behaviour or the occurrence of the pitfalls. But we were unable to define a valid measure of ‘quality of the weighting’ that is usable in a laboratory context. An ‘extreme case’ in which we gave the subjects maximum freedom as to how to fulfill the assignment, so that we could expect a maximum of variation in behaviour, could be expected to yield more useful, although less precise, results than an experiment with a limited set of well-defined (control) variables. This research was conducted with a small group of subjects because the in-depth analysis of the think-aloud protocols was very time© 2006 The Author Journal compilation © 2006 Blackwell Publishing

391

consuming. As far as the qualitative analysis was concerned, it turned out that, after 12 or so protocols had been analysed, little new insights came from each extra protocol. So we accepted the lack of statistical validity as the price to be paid for in-depth qualitative analysis. The subjects were students with no prior experience with either the acquisition of minibuses or formalized organizational importance assessment processes. The research was conducted in a laboratory context. This means that the results have limited statistical and external validity. The first is, in our view, not a great problem. The trends in the results seem to be quite clear and multiple indicators were used for many variables, thereby increasing internal validity. It also, however, means that no definite conclusions can be drawn for other groups from the group that we studied. We can, and do, make propositions about how actors in real-life situations may behave, based on our results and on the literature. The basic regularities in importance processes that we describe will, we expect, be present in some form in real-life situations. After all, it is not uncommon for individuals (albeit often with a certain degree of expertise) to make importance assessment processes under circumstances similar to those in our research. So our research provides a basis from which to look at real-life situations: inventors, professionals and policymakers. Our research pertains only to decision contexts where there is explicit weighting and where the importance assessment process is separated from the evaluation of alternatives.

Results: How Actors Conduct the (Sub-)Attribute Phase Earlier, we mentioned five forms of attribute processing: decomposing, concretizing, integration, making abstract and re-formulating. We will use these to describe our results.

Decomposing Attributes Decomposing attributes can have the important function of making clearer what is actually meant with an attribute, and also whether it can be used to make a complex attribute measurable. ‘Safety’ for example, is a very broad concept. It can be seen as a potential outcome of action (the chance of dying in a traffic accident), as an attribute that increases the chances of a desired outcome (good brakes, so that accidents can be avoided), as an emotion (feeling safe) and so on. Of course, if one wants to attach importance to safety, one has to know what one means by it. It also may

Volume 15

Number 4

2006

392


turn out that ‘safety’ is actually a collection of sub-attributes, each with their own weight. The weights of the sub-attributes together should add up to the weight of ‘safety’ (Keeney & Raiffa, 1976), but chances are that they do not always do so in practice, although experimental findings are as yet inconclusive (Borcherding, Schmeer & Weber, 1995; Fischer, 1995; Póyhónen & Hámáláinen, 1998) Even if the actual weighting of attributes is only rudimentary, as it often will be when an innovator weights attributes (since his ideas are only in development and not yet crystallized), decomposing can give some anchors. It enables (sub-) attributes to be compared as to importance at an ordinal level, it shows the consequences of giving certain (sub-)attributes higher or lower weights and generally helps to ‘paint a picture’ of attributes in plotting a course of thought or action. So, decomposing can be functional during importance assessments. Our research shows, however, that it is used often in such a way as to be dysfunctional: • so many sub-attributes are generated that oversight is sure to be lost; • decomposing is not done systematically; • nothing is done with many of the sub-attributes that are the result of decomposition. Table 2 shows the number of sub-attributes for ‘safety’ at various levels of decomposition. A level is defined as the number of splits that resulted in a certain sub-attribute. So, if ‘comfort’ is split in a number of sub-attributes, amongst which ‘quality of the seats’, which in turn is split in ‘width of the seats’ and ‘height of the armrests’, then there are two levels of decomposition. For ‘comfort’, the numbers were roughly similar to those of ‘safety’. It can be seen that the attribute is decomposed in a large number of sub-attributes. The average

number of sub-attributes per subject for ‘safety’ was 19.6. For comfort the number was 24.4. Fifteen subjects generated ten or more sub-attributes for safety. All subjects generated ten or more sub-attributes for comfort. It is not difficult to imagine that one quickly loses oversight with such numbers of attributes. Short-term memory is limited to 7– 10 items (Miller, 1956). As most subjects did not write down the sub-attributes, they would find it very difficult to work with them. The problem was compounded by the fact that decomposition was purely associative; in fact almost completely unsystematic. For example: ‘safety’ can very well be split into ‘active safety’ (aimed at preventing accidents by, say, good brakes) and ‘passive safety’ (mitigating the consequences of accidents, as seatbelts are designed to do). From this point, one could go on generating sub-attributes for both types of ‘safety’. But not one subject worked like this. If divisions like ‘active’ and ‘passive safety’ were used, then they were put next to, instead of above, other sub-attributes. An example of one subject’s attribute processing is given in Appendix 2. An excellent use of decomposing would be to establish causal, or at least statistical, relationships between (sub-)attributes. An example of such a scheme is given in Figure 2. A scheme of causal relationships serves to eliminate unnecessary attributes and avoids double–counting (see, for example, Vincke, 1992). For example: some subjects thought that high weight makes a minibus unsafe, as does a long stopping distance. But braking distance is a function of, amongst others, weight (according to the formula acceleration is mass times force). So, ‘weight’ is superfluous and can be left out. That is, unless it has other effects on safety than through stopping distance, for example because high weight means a strong

Table 2. The Decomposition of ‘Safety’ Number of attributes as a result of decomposition 0 1–5 6–10 11–15 16–20 21–25 26–30 30–35

Volume 15

Number 4

2006

Number (%) of subjects, first level

Number (%) of subjects, second level

Number (%) of subjects, third level

2 (11) 5 (28) 5 (28) 5 (28)

4 (22) 7 (39) 4 (22) 2 (11) 1 (6)

15 (83) 1 (6) 1 (6) 1 (6)

1 (6)



Quality of brakes

stopping distance

quality of tires

393

ability to avoid collision

visibility conditions

chances of injury or death

Figure 2. A Scheme of Causal Relationships Between Sub-Attributes Table 3. Integration of Safety and Comfort Number of sub-attributes being integrated

Number (%) of subjects integrating sub-attributes of safety

0 1–5 6–10

chassis. But then we should leave either ‘weight’ out, or ‘stopping distance’ and ‘strength of the chassis’. If we do not, we count ‘weight’ twice, once directly and once through its effects. However, no subject even tried, much less succeeded, in establishing causal relationships. They sometimes mentioned causal relationships in passing, but they did not make any systematic use of them. Going one step further, a possible use of establishing causal relationships between (sub-)attributes is: find a common denominator for expressing some or all attributes to be weighted. With this, the weighting problem is effectively eliminated. For example, if it were possible to link both ‘safety’ and ‘comfort’ to ‘profit’ and it were turn out that ‘profit’ would be maximized with 10 units of ‘safety’ and 15 units of ‘comfort’, then the optimal decision is clear. No further weighting is required. This approach is difficult, but it is used in, for example, the acquisition of civil and military aircraft. Attributes are causally linked to, respectively, life-cycle cost and revenues and combat effectiveness. Yet none of the subjects even tried to find a common denominator. Only one of the subjects mentioned this as an explicit aim during certain stages of the assignment, but he did not pursue this aim consistently. The others did not address the quest for a common denominator in any systematic way. We will return to the causal scheme when we suggest ways of improving the importance assessment process. Knowing the causal relationships between attributes makes integrating attributes possi© 2006 The Author Journal compilation © 2006 Blackwell Publishing

9 (50) 6 (33) 3 (17)

Number (%) of subjects integrating sub-attributes of comfort 13 (72) 5 (28)

ble; the opposite of decomposition. With luck, all attributes can be integrated in two or three main attributes that can then be causally related to a common denominator. At the very least, integration reduces the number of subattributes to be weighted and makes the weighting less complex. How did the subjects go about integrating?

Integration of Sub-Attributes There was hardly any integration of subattributes, contrary to what one expects when weights are to be given not to sub-attributes but to main attributes. Integration was rare, as can be seen from Table 3. This table shows the number of (sub-)attributes integrated. We classify a subject as a non-incidental user of integration if integration occurs in at least four cases over safety and comfort together. We see that at least half of the subjects do not integrate at all, and the other half integrates only a fraction of the sub-attributes generated. Another way to look at the significance of integration is to observe how many (sub-) attributes are the result of integration. The maximum number of attributes that were the result of integration was four (one subject); two integrated attributes were found with only three subjects. Only in two instances was a sub-attribute resulting from integration given a weight during the final weight assignment. Integration always resulted in a new sub-attribute, not in the main attributes to be weighted according to the assignment (safety and comfort). The logic of the integration was often implicit and nearly always purely qual-

Volume 15

Number 4

2006

394


itative. No indexing or other quantitative methods were used. In sum, integration was by and large irrelevant. That leaves only re-formulation, abstraction and concretizing to be addressed. We can be short about these. Their roles in attribute processing were minor, both quantitatively (number of abstractions and so on) and qualitatively (contribution to the weighting process). Only in five cases were between five and ten sub-attributes of either safety or comfort concretized, and never more than ten. Nine subjects in all made either safety or comfort concrete, but then they either proceeded with decomposition or eventually weighted only safety and comfort and not the sub-attributes. Only half of the subjects made abstractions for safety and for comfort, and never for more than two sub-attributes. We will not devote any attention to abstraction. Sometimes (sub-)attributes were reformulated, but this was almost exclusively limited to inconsequential changes such as ‘comfort’ becoming travelling comfort’. Whenever reformulations were more significant they could be classified as concretizations or abstractions. Is it a good or a bad thing that there is so little concretization? At first sight, a logical function of concretization is operationalization. One subject used concretization for this purpose, and made it explicitly known that he wanted to operationalize sub-attributes and did so by taking the judgement of outside experts (for example, the Dutch Consumer Association) as an indicator of safety, driving quality, and some other attributes. However, this subject was the only one who used concretization in this way explicitly and with any pretence of being systematic. So, although subjects used concretization to make (sub)-attributes more concrete, they had no use for the end-product of specification other then getting a better idea of the meaning of the (sub)-attributes. A way of concretization that was used by a number of subjects was explicitly or implicitly specifying the extremes of an attribute. This amounts to the beginning of scale construction. Three subjects (17 percent) used this directly in order to define weights. One subject used two types of cars to define the extremes of comfort (a Limousine and a Ford Fiesta). He used them only once, as examples, so this did not have any further measurable influence on the execution of the assignment. Subjects did not evaluate the result of specification. It seemed to be a largely unintentional process, even more so than decomposition, where subjects sometimes used some system, however unsophisticated.

Volume 15

Number 4

2006

Discussion: Pitfalls in (Sub-)Attribute Processing The attribute-processing phase is meant to convert attributes in such a way that they can be weighted. This can be done by defining them so clearly that their importance can be fixed, or by finding a common denominator. Our subjects, for the following identified reasons, achieved neither aim: • there was no system in the decomposition; • no causal relations were established between sub-attributes so; • integration, concretization, abstraction and re-formulation hardly took place and; • the unworkably huge number of subattributes was thus not reduced. All this makes it unlikely that assessing the weights of the sub-attributes would in any way be easier than assessing the weights of the main attributes. It could be assumed that subjects used the processing of attributes mainly for framing purposes, that is, to find out what ‘safety’ and ‘comfort’ actually mean. But since all subjects devoted considerable effort to phase 3 (absolute weighting, mainly of subattributes, to be covered in a forthcoming article) the processing of sub-attributes appeared to mean more than just a framing function. The subjects obviously were not merely interested in obtaining concepts of ‘safety’ and ‘comfort’. The sub-attributes were important in their own right. But it is unclear what their relevance is. What does all this mean for an innovator in real life, wrestling with a choice to make between several alternative courses of action, each with their own pros and cons? It means a high degree of uncertainty in deciding what is important and what is not. The sub-attributes in the innovator’s mind, and hence the assigned weights, may vary from one moment to the next. When there are a lot of (sub-) attributes, such variations can alter choices significantly. It also becomes more difficult to give good and consistent arguments for a decision or weight judgement, and to communicate those arguments to others. This is especially likely with ‘wicked problems’, the type of problems that an innovator is confronted with almost by definition. It is clear that ambiguity does not help in convincing others, or in securing alliances. There is ample research to show that people’s weighting is often inconsistent with logic, in that choices made are not in line with the weights assigned (see, for example, Kahneman, 1994; Kahneman, Knetch & Thaler, 1990, 1991). They will sometimes choose from two options the one © 2006 The Author Journal compilation © 2006 Blackwell Publishing


that is least attractive, given the weights they chose and the scores on the attributes. For example, they may choose the attribute with the highest score on the most important attribute, ignoring the alternative with a much higher score on a marginally less important attribute that makes it, all-in, the most attractive. And it gets worse. Levine, Halberstadt and Goldtsone (1996) showed that people’s weighting gets less consistent when they have to deliberate about the weights. That may not be surprising, in the light of our research. It is conceivable that if people do not have to reason, they do not consider many sub-attributes of the attributes to be weighted, so their image of the attributes stays reasonably constant over time. Once they get to reason about the attributes, the images start changing as various sub-attributes are activated sequentially. Unfortunately, for an innovator having to choose a path of action, not reasoning about weights is not an option. So there is good reason to look for ways to improve the consistency and hence the quality of the weighting process.

Let Us Make Things Better: Improving Performance of (Sub-)Attribute Processing Our advice for anyone who has to weigh attributes in an non-routine decision is simply to make the (sub-)attribute processing phase of the importance assessment process more transparent. This can be done in the following ways: 1. Instead of splitting attributes, try defining (concretizing) them. For example: ‘safety’ can be described as ‘the number of deaths and wounded per 1,000,000 km where attributes of the minibus are the root cause’. With ‘safety’ thus specified, it may no longer be necessary to split it up in sub-attributes, because the decision-maker does not have to know which attributes play a role in accidents. The manufacturer of the bus knows, and gives a score for ‘safety’. 2. If you feel you have to decompose, do it systematically. Some decision-makers may not be satisfied with (1) and they may want to specify an attribute further by splitting it up in sub-attributes. But, then do not just start throwing sub-attributes around. For ‘safety’, several possible decomposition systems are possible: active versus passive safety, attributes concerning the engine, brakes, chassis, lighting or interior and so on. Such a system makes it easier to assess © 2006 The Author Journal compilation © 2006 Blackwell Publishing

395

the completeness of the list of sub-attributes (to be covered in a future article) and makes the next two points easier to perform. 3. Make a ‘cognitive map’ (de Boer, 1998; Warren, 1995) of the causal relationships between (sub-)attributes. A cognitive map is simply a drawing of the attributes connected by arrows. The arrows represent cause-effect relationships. Figure 2 is essentially an example of a cognitive map. In this way, the really important attributes stand out, as do the superfluous ones (remember the example of weight, stopping distance and strength of the chassis). If you are lucky, you end up with only four or five relevant sub-attributes per attribute. Those can all be taken into account if the importance of the attribute is to be set, much more so than the 19 to 25 sub-attributes our subjects generated per attribute. 4. Try to integrate sub-attributes to a level that is as high as possible but at which you feel you can still assign weights. For example: it may not be more difficult to set a weight for ‘active safety’ than it is to weigh ‘quality of the brakes’ or ‘having power steering’. Here again; the fewer (sub-)attributes to weigh, the better. Fewer (sub-)attributes means more consistency (hopefully), a better oversight of what you are doing (certainly) and a better argumentation to give if your decision is challenged (very likely). 5. After having done one or more of the previous points, decide at which level the (sub-)attributes should be weighted. As we showed in Table 1, our subjects frequently generated three levels of sub-attributes. Even after having left out and/or integrated some sub-attributes, more than one level may be left. It is not a foregone conclusion that weighting has to be done at the lowest level. Sure, it promises more precision, but there are more sub-attributes to weigh and, ideally, all sub-attributes should be weighted against all other sub-attributes to get the optimal result (for a method to do this, see de Boer, 1998; Saaty, 1980). This can be a daunting task.

Closing Remarks We have shown that the most important phase of the importance assessment process, phase 2 of the WAM, contains pitfalls that actors can not only identify but can also negotiate. Hopefully, although we cannot prove it, improving the quality of the (sub-attribute processing phase will improve the quality of the subsequent weighting phases. And it may increase the confidence of the actors concerned in their

Volume 15

Number 4

2006

396


weighting, which they may experience as a positive effect in itself. The instruments discussed above are conceptually simple and can be utilized with only a sheet of paper at hand and a few hours’ time to spare. They are ideal for individual actors with few resources at their disposal and the need to clarify for themselves, if not for others, the pros and cons of their decisions. Thus, they can be considered ideal for innovators (who often, although not always, work alone in conceptualizing their new ideas) to carry them in their mental arsenal. But the use of these instruments has to be learned. In the experience of the author with under- and postgraduate university students, making a cognitive map, for example, takes at least an afternoon to practise. A global understanding of the elements of a decision (alternatives, attributes and weights) is also needed. In addition, it is not clear how much the entire importance assessment process, and the decisions based on it, will improve by focusing on the (sub-)attribute phase But it seems the results are well worth the limited training effort required.

References Akin, Ó. (1994) Creativity in Design. Performance Improvement Quarterly, 7, 9–23. Baarda, D.B. and de Goede, M.P.M. (2001) Basisboek methoden en technieken. Stenfert Kroese, Groningen. Barlas, S. (2003) When Choices Give in to Temptations: Explaining the Disagreement Among Importance Measures. Organizational Behavior and Human Decision Processes, 91, 310–21. Boer, L. de (1998) Operations Research in Support of Purchasing; Design of a Toolbox for Supplier Selection. Universiteitsdrukkerij, Enschede. Borcherding, K., Schmeer, S. and Weber, M. (1995) Biases in Multi-Attribute Weight Elicitation. In Caverni, J.P., Bar-Hillel, M. and Jungemann, H. (eds), Contributions to Decision Making-1. Elsevier Science BV, Amsterdam, pp. 3–28. Dawes, R.M. (1979) The Robust Beauty of Improper Linear Models in Decision Making. American Psychologist, 34, 571–82. Ericsson, K.A. and Simon, H.A. (1993) Protocol Analysis: Verbal Reports as Data (2nd edn). Bradford Books/MIT Press, Cambridge MA. Fischer, G.W. (1995) Range Sensitivity of Attribute Weights in Multiattribute Value Models. Organizational Behavior and Human Decision Processes, 62, 252–66. Goldstein, W.M. and Busemeyer, J.R. (1992) The Effect of Irrelevant Variables on DecisionMaking: Criterion Shifts in Preferential Choice. Organizational Behavior and Human Decision Processes, 52, 425–54. Guo, C. (2001) A Review on Consumer External Search: Amount and Determinants. Journal of Business and Psychology, 15, 505–19.

Volume 15

Number 4

2006

Harte, J.M. and Koele, P. (1995) A Comparison of Different Methods for the Elicitation of Attribute Weights: Structural Modeling, Process Tracing and Self-Reports. Organizational Behavior and Human Decision Processes, 64, 49–64. Heerkens, J.M.G. (2003) Modeling Importance Assessment Processes in Non-Routine Decision Problems. Enschede (dissertation). Heerkens, J.M.G. and Van der Heijden, B.I.J.M. (2003) On Importance Assessment and Expertise in Non-Routine Decisions; an Exploratory Study on the Cognition of Weighting Processes of Capital Goods’ Attributes. International Journal of Management and Decision Making, 3, 370–98. Heerkens, H. and Van der Heijden, B.I.J.M. (2005) On a Tool Analyzing Cognitive Processes using Exploratory Think-Aloud Experiments. International Journal of Human Resources Development and Management, 5(3), 240–83. Hollingshead, A.B. (1996) The Rank-Order Effect in Group Decision-Making, Organization Behaviour and Human Decision Processes, 68, 181–93. Hsee, C.K. (1995) Elastic Justification: How Tempting but Task Irrelevant Factors Influence Decisions. Organizational Behavior and Human Decision Processes, 62, 330–7. Jaccard, J., Brinberg, D. and Ackerman, L.J. (1986) Assessing Attribute Importance. Journal of Consumer Research, 12, 463–7. Kahneman, D. (1994) New Challenges to Rationality Assumption. Journal of Institutional and Theoretical Economics, 150, 18–36. Kahneman, D., Knetch, J.L. and Thaler, R.H. (1990) Experimental Tests of the Endowment Effect and the Coase Theorem. Journal of Political Economy, 96, 1325–48. Kahneman, D., Knetch, J.L. and Thaler, R.H. (1991) Anomalies: the Endowment Effect Loss Aversion and Status Quo Bias. Journal of Economic Perspectives, 5, 193–206. Kahneman, D. and Tversky, A. (ed.) (2000) Choices, Values and Frames. Cambridge University Press, Cambridge. Keeney, R.L. (1992) Value-Focused Thinking. Harvard University Press, Cambridge MA. Keeney, R.L. (1994) Creativity in Decision Making with Value Focused Thinking. Sloan Management Review, 35, 33–41. Keeney, R.L. and Raiffa, H. (1976) Decisions with Multiple Objectives, Preferences and Value Tradeoffs. John Wiley & Sons, New York. Kray, L.J. (2000) Contingent Weighting in SelfOther Decision Making. Organizational Behavior and Human Decision Processes, 83, 82–106. Kray, L. and Gonzalez, R. (1999) Differential Weighting in Choice versus Advice: I’ll Do This, You’ll Do That. Journal of Behavioral Decision Making, 12, 207–17. Levine, G.M., Halberstadt, J.B. and Goldtsone, R. (1996) Reasoning and the Weighting of Attributes in Attribute Judgments. Journal of Personality and Social Psychology, 70, 230–40. Meehl, P.E. (1954) Clinical Versus Statistical Prediction. University of Minnesota Press, Minneapolis. Miller, G.A. (1956) The Magical Number Seven, Plus or Minus Two. Some Limits on Capacity of © 2006 The Author Journal compilation © 2006 Blackwell Publishing


Processing Information. Psychological Review, 63, 81–7. Póyhónen, M. and Hámáláinen, R. (1998) Notes on the Weighting Biases in Value Trees. Journal of Behavioral Decision Making, 11, 139–50. Rittel, H.W.J. and Webber, M.M. (1973) Dilemmas in a General Theory of Planning. Policy Sciences, 4, 155–69. Russo, J.E., Medvec, V.H. and Meloy, M.G. (1996) The Distortion of Information During Decisions. Organizational Behavior and Human Decision Processes, 66, 102–10. Saaty, T.L. (1980) The Analytic Hierarchy Process. McGraw-Hill, New York. Simon, H.A. (1960) The New Science of Management Decision. Prentice-Hall, New Jersey. Strauss, A. and Corbin, J. (1998) Basics of Qualitative Research; Grounded Theory, Procedures and Techniques. Sage Publications, Thousand Oaks CA. Timmermans, D. (1993) The Impact of Task Complexity on Information Use in Multi-Attribute Decision Making. Journal of Behavioral Decision Making, 6, 95–111. Vincke, P. (1992) Multiple Decision Aid. John Wiley & Sons, Chicester.


397

Warren, K. (1995) Exploring Competitive Factors using Cognitive Mapping. Long Range Planning, 28(5), 10–21. Wei, Q., Yan, H., Ma, J. and Fan, Z. (2000) A Compromise Weight for Multi-Criteria Group Decision Making with Individual Preference. Journal of the Operational Research, 51, 625–34. Zeelenberg, M., Beattie, J., van der Pligt, J. and de Vries, N.K. (1996) Consequences of Regret Aversion: Effects of Expected Feedback on Risky Decision Making. Organizational Behavior and Human Decision Processes, 65, 148–58.

Hans Heerkens ([email protected]) is assistant professor at the School of Business, Public Administration and Technology, department of Operational Methods for Production & Logistics, University of Twente, PO Box 217, 7500 AE Enschede, Netherlands, phone (–31)-53-4893492

Volume 15

Number 4

2006

398


Appendix 1: The Auxiliary Activities During the formal coding of the protocols, some activities were found that could not be placed in particular phases, because they occurred in several phases or because they logically are not part of the weighting process or to the assignment. The first category comprises activity planning and information assessment. Examples of the second category are alternative judging (choosing a minibus) and attribute scoring (assessing whether a minibus is, for example, comfortable). These activities certainly are phases of the acquisition process but they should follow the weighting process, not be part of it. Also in the second category is weighting procedure design for a real-world situation. The assignment did not call for this and the designed procedures could not be used in the experimental situation. Expressing emotions is part of moth categories. The table shows the auxiliary activities of the WAM. Table A1. Auxiliary Activities of the WAM Activity number 1 2 3 4 5 6

Activity name

Alternative judging Attribute scoring Activity planning Information assessment Weighting procedure design Expressing emotions

We will now briefly discuss the auxiliary activities.

1 Alternative Judging Subjects can make judgements about the attractiveness of alternative minibuses (mainly the two minibuses mentioned in the information package) or about the attractiveness of minibuses in general vis-à-vis other modes of transport, such as trains or private cars.

2 Attribute Scoring Subjects might make statements about how they think a particular minibus, minibuses in general or alternative modes of transport might score on certain (sub)-attributes. Examples of such statements are: ‘Both the Opel and the Volkswagen have a stereo-set on board’.

Volume 15

Number 4

2006

3 Activity Planning Prior to starting one of the phases of the WAM, a subject might plan how to execute the phase. An example is the statement: ‘I think I should start by defining what I think “safety” and “comfort” actually mean’.

4 Information Assessment With information assessment, we mean activities concerned with searching for information, and assessing the value of the available information for the task at hand. All subjects, at one point or another, concerned themselves with issues like the sort of information that they felt was needed, the information that could or could not be found in the information package and the quality of the information provided.

5 Weighting Procedure Design Since the subjects were instructed to imagine that they had to perform the weighting in support of an acquisition process of an organization, they might indicate how they would propose to conduct the weighting process (or the acquisition process as a whole) if they were really working for a company and not just imagining it. Some subjects, for example, stated that they would in reality propose to conduct a market survey in order to assess how important safety and comfort are to present or potential customers.

6 Expressing Emotions Statements such as: ‘Oh, how difficult this is! I don’t think that I can do it’, and: ‘I am distracted by all the birds I see flying outside’ perhaps do not say very much about the actual weighting process, but they form a clearly identifiable category and they might affect the weighting. For example, if a subject feels he or she cannot cope with the assignment and still has this feeling after weights have been given, the confidence level probably is very low, which could lead to equal weights for all attributes.

Appendix 2: Example of An Attribute-Processing Scheme by One of the Subjects (Students) in Our Research The schemes should be read as follows. Safety always gets the number 1 and comfort the number 2. Decomposed attributes at the first level get the numbers 1.1, 1.2, 2.1, 2.2 etc. At © 2006 The Author Journal compilation © 2006 Blackwell Publishing


the second level, the numbers consist of three digits and can be, for example, 1.1.1, 1.1.2 etc. A letter placed after a certain attribute number means that the attribute is a re-formulation. If an attribute is the abstraction of another attribute, this is noted between brackets. An integration is always the result of two or more attributes being processed, and is also indicated between brackets. The sub-attributes of the first level are listed as much as possible in the order in which the subjects mentioned them. This is the (sub-)attribute processing scheme of just one of the subjects in our research, of only one of the two attributes. The scheme is representative of the schemes made by the other subjects, in terms of size and structure.

1.2b: 1.2c: 1.2d: 1.2.1:

1.3: 1.4: 1.5: 1.6: 1.7:

The Processing of ‘safety’ 1.

Safety 1a:

1b:

1c: 1.1: 1.1a: 1.1b:

1.1c: 1.1d: 1.1e:

1.1f: 1.2:

If an accident happens, you want to get out in one piece, preferably unhurt (specification) I want to get out in one piece or with very minor injuries, but not so that I can sit in a wheeled chair for the rest of my life Accidents Number of deaths per year with a certain brand Number of accidents with which it has occurred Number of deaths per year with accidents (from the context it is clear that it is meant per type) Accident numbers (from the context it is clear that it concerns deaths per year) Maximum so many deaths per year How many deaths per year with accidents and with how many accidents does this happen? Number of deaths per year

Number of serious injuries 1.2a: Number of serious injuries per year per accident


1.8: 1.9: 1.10: 1.11: 1.12:

1.13:

399

How often does it occur (serious) injuries? Figures about serious injuries Number of serious injuries per year Paralyzed (downwards) from a certain body part or really loose a body part Paralyzed Body part coming off

1.2.1.1: 1.2.1.2: Seatbelts 1.3a: Are seatbelts in the car? Seat broke loose Anti-skid system Are there headrests? 1.6a: Headrests Can headrests be adapted? 1.7a: Are headrests adaptable? 1.7b: Are they adjustable in height (no specification because this is what he meant with 1.7 and 1.7a) 1.7c: Are headrests adjustable? 1.7d: Adjustable headrests Safety for driver Safety for assistant-driver Safety for passengers Airbag 1.11.1: Airbags on the side How does a bus fare if you smash into it from the front, the rear, the side and from above? 1.12a: With crash tests what was the result (abstraction) 1.12b: Result with type of accident 1.12c: Result with crash tests 1.12.1: If an airplane crashes on your car 1.12.2: If such traffic pole like you have in Enschede comes crashing into your car from underneath 1.12.3: From the side they come 1.12.3a: If someone comes from the side To what extent does a baby sit safely in the car? 1.13.1: Has it got baby seats? 1.13.2: Does the possibility exist to install them (baby seats) 1/13.1/ Baby seats are they there, can they be installed (integration) 1.13.2(a) 1.13.3: Do baby seats have to be with the face forward or with the face rearward? 1.13.3a Which baby seats are dangerous, which are not dangerous?

Volume 15

Number 4

2006

400


A Multidimensional Product-Concept Model Enhancing Cross-Functional Knowledge Creation in the Product Innovation Process: The Case of the Suunto t6 Training Wrist Computer Seppo Hänninen and Ilkka Kauranen This article presents a four-dimensional product-concept model enhancing cross-functional knowledge creation in product innovation: the dimensions presented in the new model are: technology, end-user, brand and business logic. The application of the model is described in the case study of the Suunto t6 training wrist computer. For the Suunto t6 development it was revealed that cross-functional knowledge creation had happened on an even larger scale than expected. Analysis of the results suggests that certain dimensions of the product concept can be especially indicative of cross-functional knowledge creation, such as concern with the enduser’s experience. Furthermore, the new product concept may provide an early warning of innovation-based diversification. Nicknames for the product concept under development, which requires the verbalisation of tacit subjective associations, can be used as indicators of cross-functional integration within the organization and as suitable indicators of tacit knowledge. Measurement of all relevant organizational capabilities is discussed.

Introduction Product development is knowledge creation in which tacit images are translated into sharable concepts (cf. Madhavan & Grover, 1998; Nonaka, Toyama & Nagata, 2000). The underlying mental models may differ, however, and cause difficulties in cross-functional knowledge creation (cf. Cohen & Levinthal, 1990). In particular, non-articulated images and mental models are liable of causing misunderstandings. Generaly, product concepts have emphasized only one or a few knowledge bases. For example, implementing product development from a technical solution to commercialization leads to low-level cross-functional knowledge creation, since in building the product concept in such a way the development stages follow each other consecutively instead of interacting concurrently. The product concept is the first explicit phase in product-innovation related knowledge creation (Nonaka, Toyama & Nagata, 2000). Thus, the product concept

Volume 15

Number 4

doi:10.1111/j.1467-8691.2006.00409.x

2006

demands the conceptualization of tacit knowledge in a well-rounded way, taking, as exhaustively as possible, all the relevant knowledge bases into consideration. In contrast to this, patents, which are typically perceived to be a comprehensive means of conveying the essence of a new product concept, in fact describe only the technical solution. The co-presence of both individual and organizational creativity mechanisms will lead to the highest level of innovation performance (Bharadwaj & Menon, 2000). Well-functioning, multidimensional product-concept models support such creativity, and can be used as a medium for practical cross-functional knowledge creation. For the purposes of the present study, a definition for product concept can be derived by taking the essence of the definitions of ‘concept’ and ‘product definition’ as presented in the Product Development and Management Association (PDMA) Glossary for New Product Development. There, the [product] concept is defined as ‘a clearly written and possibly © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

A MULTIDIMENSIONAL PRODUCT-CONCEPT MODEL

visual description of the new product idea that includes its primary features and consumer benefits, combined with a broad understanding of the technology needed’, and the product definition ‘defines the product, including the target market, product concept, benefits to be delivered, positioning strategy, price point, and even product requirements and design specifications’ (PDMA, 2005). In the product innovation literature, one knowledge base is typically discussed at a time (Kogut & Zander, 1992; Tushman & Anderson, 1986; Walsh, 1995); very rarely are more than two knowledge bases discussed together (cf. Leonard-Barton, 1992). Gomes et al. (2001) recommend using various knowledge tools to support product innovation development. The various stakeholders have the best knowledge of their own knowledge bases, but the more they also know of other knowledge bases, the more cross-functional knowledge creation is enhanced. Without a shared product concept, arguing about the pros and cons can continue interminably, leading to battles between departments and individuals within the product innovation development organization. The need to enhance cross-functional knowledge creation in product innovation development is clear, and this is even more the case with modern knowledge-based products. The objective of the present study is to present and analyse a novel multidimensional product-concept model enhancing crossfunctional knowledge creation in product innovation development.

Existing Knowledge and New Theory Development Product Concepts Built on a Single Knowledge Base In the emergence phase of innovation literature, product innovation was typically addressed with a sole emphasis or an overemphasis on the technological knowledge base. Extending the focus of attention to other knowledge bases and their relationships is a more recent phenomenon and remains an ongoing process. The advantages of technology-knowledgedriven product concepts are thorough technological know-how and the option for innovative products. The risk, however, is that other knowledge bases may be partly neglected or totally forgotten, and the product may therefore fail (cf. Cooper, 1975; Hänninen, 2006; Hänninen et al., 2006). © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

401

Cross-functional knowledge creation can be cumbersome and ineffective if the product concept utilizes the technological knowledge base alone, since in such a case the other phases, such as developing end-user solutions and developing business-logic solutions, follow linearly and not simultaneously.

Product Concepts Built on the Technology Knowledge Base and the Market Knowledge Base The product innovation process, including the development of product concepts, is in the literature traditionally understood as either a technology-driven process or a market-driven process (Orihata & Watanabe, 2000). These two approaches seem to have developed initially as parallel lines of thinking. However, if market information can be integrated with the technological possibilities, more powerful product concepts can be built. The recommended approach for creating a product concept based jointly on the technology knowledge base and the market knowledge base is to agree initially on unified objectives. The option to be avoided is a continuous battle between the two functions for dominant position in the product-concept definition process. Solutions that concurrently utilize the technology knowledge base plus the market knowledge base solution in product development are typical in consumer product companies. One tool for furthering this outcome is that market information is collated with technological possibilities by regular candid and open meetings between the marketing function and the product development function within the company. In such a case, the probable outcome is that the product development is enhanced but also concentrated inside the company. In reality, such technology-market polarisation is an oversimplified view. For example, the brand is also another strong stakeholder in the product concept. Second, a viable product concept should acknowledge that competitors are only one group of market intermediaries, and competition is only one of the market processes (Porter, 1980).

Product Concepts Built on the Technology Knowledge Base, the End-User Knowledge Base and the Brand Knowledge Base The importance of understanding the customer needs is one of the most widely researched topics in innovation literature. Thus, the end-user knowledge base is a necessary part of a multidimensional productconcept model. In practice, usability research as

Volume 15

Number 4

2006

402


part of the product innovation process offers companies a functional approach to studying the end-user’s relationship with the product (Scholtz & Salvador, 1998; Wood, 1998). What has been less emphasized is the experience created for the end-user by the product (Prahalad & Ramaswamy, 2003). Some firms actively trigger, monitor and guide the consumer’s knowledge base in order to develop and promote the adoption of successive product generations (Hänninen & Sandberg, 2006). As ‘market’ has been too fuzzy a definition to cover all the market-related stakeholders exhaustively, ‘brand’ has also been exploited in the product concept, for example, in product design. An enterprise’s brand does not exist independently of its products; at the heart of the concept of a brand is an innovatory product with the added value that this innovation creates for the end-users (Kapferer, 1994). The product’s values, quality, form and functional attributes all affect the brand value. The design needs to be consistent and coherent throughout, from the innovation itself to the firm’s entire mode of operation. The importance of the brand aspects of the product concept is growing. A strong brand is used more and more to facilitate the launching of innovative high-technology products. Cross-functional knowledge creation is raised to a new level with brand knowledge, because there are many stakeholder groups in the product innovation process, and the sharing of brand knowledge offers an option for higher integration.

Product Concepts Built on the Technology Knowledge Base, End-User Knowledge Base, Brand Knowledge Base and Business Logic Knowledge Base Prahalad and Bettis (1986) have discussed the concept of the business logic of an industry and have emphasized the importance of conceptualizing and challenging it. Kim and Mauborgne (1997) were among the first to stress the value of business logic as a product innovation-related attribute. Walsh (1995) sees business logic as a managerial cognition, which is dependent on the position of the company in the market. In order to avert a situation of direct competition, enterprises need to conceptualize the kinds of innovation that will be capable of changing the business logic of their industry (Hänninen & Kauranen, 2006; Kim & Mauborgne, 1997). Firms that cannot alter the prevailing rules of the market, on the other hand, must make their products comply with the dominant business logic. Business logic and business model are overlapping terms. They both define value-creation priori-

Volume 15

Number 4

2006

ties with respect to the utilization of both internal and external resources, for the purpose of creating value for and with the customers. Business logic looks at value creation from the viewpoint of the market, and at the business model from the viewpoint of the firm (Wallin, 2000). What is essential is that although business logic can be influenced by the company, it is not under the company’s full control. A radically novel product concept can change power relationships within the organization profoundly (Normann, 1971). Resistance against a product innovation, and the organizational changes that it causes, can be minimized if the innovative product concept collates knowledge from many related functions. The four-dimensional product concept introduced in the present study enhances cross-functional knowledge creation, but at the same time the conceptualization process demands more integration time, good relationships between different functions and high negotiation skills. All in all, the product development literature suggests that early and widely accepted innovative multidimensional product concepts shorten the time to market (e.g. Crawford, 1996). Additional knowledge bases related to a product concept include the marketing knowledge base, manufacturing knowledge base, corporate strategy knowledge base and combinative knowledge base, among others. The combinative knowledge base is analogous to the combinative capabilities discussed by Kogut and Zander (1992). In the future development of multidimensional product concepts, these and other additional knowledge bases need to be included.

Methodology Cross-functional knowledge creation is a challenging phenomenon to study in research. The explicit component of knowledge can typically be measured with precise quantities, but the contents of tacit knowledge include components such as skills and values, which are difficult to quantify. In the present study ‘knowledge’ is defined as cross-functional knowledge if the same innovation-specific knowledge is expressed by many interviewees representing various different functions in the organization.

Questionnaire Figure 1 sets out how the four-dimensional product-concept model introduced in the present study was translated into a research questionnaire. Technology, end-user, brand and © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


PRODUCT

Name, type number, nickname

TECHNOLOGY -Technical solution -Features-main functions -Unique selling proposal

– Underlying technologies – What solutions do the technologies offer? – What critical features do the technologies offer? – What are the added value functions? – Is there a single unique benefit?

END-USER -Ethnology -Usage context -Needed value -Customer’s experience about the product, in their own words

– Personality, profession, hobbies, consumerism – Real-world behavioural patterns in brief – How is the product really used? – Which customer needs does the product satisfy? – “What is my experience of using the product?”

BRAND -Promise of value -Values -Brand characters -Design hardware -Design software -Surprise BUSINESS LOGIC

403

-Nature of expectations, e.g. from the brand slogan -Brand’s main values -Concretely verifiable characters -Brand indicators of form of the product -Software’s brand characters -Something unexpected -Key drivers of profit flow -Solutions to the barriers of profit flow

Figure 1. The Four-Dimensional Product-Concept Model business logic refer to the corresponding knowledge bases. A person who has appropriated only one knowledge base will be unable to respond properly to all the questions included in the questionnaire. First, the interviewees were asked to name a unique selling proposal that is the output of the technical solution. Second, the interviewees were asked to predict end-user experiences in their own words. Responding to this question demands empathy. Third, the aim was to obtain a thorough understanding of the brand values and brand characteristics. Fourth, the business logic of the product innovation was also investigated, by asking about the commercialization capabilities of the organization. All the questions presented are listed in Appendix 1.

Case Selection The case product innovation was selected on the basis of recommendations by technology experts, who were given the criteria that the case should fulfil. The framework developed in the present study was assessed by comparing it with one instrumental case (cf. Stake, 1995). The case was selected to represent successful cross-functional knowledge creation.

Data Collection The material was collected in October 2004. The managing director, marketing director, © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

technological director, technical manager and project manager of the case company were interviewed in person. Similarities and differences between different interviewees’ responses were investigated, as every interviewee had to respond to the same questions about the case product innovation. The interviews were recorded and the responses were then content-analysed and compared. The interview data was supplemented with information obtained, for example, through the Internet and from written sources.

Case Study Product Innovation Suunto t6 The case product innovation in the present study is the training wrist computer Suunto t6, which was developed by the company Suunto Oy. Suunto Oy is part of the Amer Sports Corporation, which is an internationally operating public company, with annual sales of about €1,000 million in 2004. The background of Suunto Oy is in the compass industry, from which the company has moved on to making use of the Global Positioning Service (GPS) – the satellite-supported navigation system. The Amer Sports Corporation has a strong global position in many sports, such as golf and tennis. The impulse to develop the Suunto t6 came from research results regarding top sports

Volume 15

Number 4

2006

404


obtained at the University of Jyväskylä. The innovation development process in the company influenced many stakeholders internally and externally. The biggest challenges posed by the Suunto t6 were in radio signalling and user-interface development, in order to achieve the quality level set by the Suunto brand values. The Suunto t6 provides laboratory-accurate information about seven body parameters for later analysis with Suunto Training Software. The wrist unit has training control functions and memory functions. The training control functions include simultaneous display of heart rate, stop-watch and time/lap time/ countdown, adjustable heart-rate zone with alarm, and interval timer with warm-up. Memory functions include average heart rate, altitude data for total training and laps, and training history information. The wrist unit is upgradeable with wireless accessories. The Suunto Training Software provides a deeper analysis of training sessions on seven body parameters: excess post-exercise oxygen consumption (EPOC), heart rate, respiration rate, ventilation volume, oxygen intake, energy consumption and training effect. The training effect is an indicator of how much the training session improved the person’s aerobic fitness, especially the maximum performance of the person’s cardiovascular system and the ability to resist fatigue during endurance training.

Results Some results were expected, but there were also some surprises. As expected, the divisions knew their respective specific knowledge bases well. This result is almost too trivial to be presented. Most of the results, however, were unexpected. First, in this case the different divisions also knew each other’s fields very well. For example, the interviewees with marketing background demonstrated a profound understanding of technical issues. On the other hand, interviewees with a technical background were also able to give an in-depth response concerning the brand identity of the Suunto t6. This result indicates that cross-functional knowledge creation had happened on an even larger scale than expected. Cross-functional knowledge creation was one of the reasons for the success of the product innovation, and a shared multidimensional product concept model recognizably enhanced knowledge creation. A second unexpected finding was that the nicknames suggested spontaneously in the interviews differed considerably from each other. The following nicknames were proposed: ‘Personal Trainer’, ‘Training Partner’,

Volume 15

Number 4

2006

‘Training Effect’, ‘Virtual Trainer’, ‘Personal Coach’ and a playful nickname in Finnish that does not translate into English. The wide range of different nicknames indicates that within the group, which had been developing the same product for a longer time, many different perceptions of the tacit components of the product concept still existed. It seems that eliciting nicknames can serve as a method for unveiling hidden differences in the productconcept that different sub-groupings within the development group might have, even when the explicit product concept is shared throughout the group and the product development has already been completed. Clearly, a prerequisite for analysing the nickname in this fashion is that no conscious decision has yet been made on a common nickname for the product.

Technology One of the main key features of the technical solution in the Suunto t6 is data transmission by radio waves from the body band to the wrist unit, which facilitates the economic and efficient transmission of data. This technical solution, however, is closely related to design features. For example, when radio transmission is utilized, it is not possible to use a metal case in the wrist computer. The Suunto t6 is therefore housed in a plastic case, although other Suunto products are typically cased in metal in order to promote a uniform, chic Suunto product image. This is an illustrative example of an instance where the multidimensional product concept model has an opportunity to enhance cross-functional knowledge creation in the product innovation process. The challenge of integrating the technology perspective into the development requirements of the other functions could also be seen in the interview material, as the technical solution of data transmission by radio waves was highlighted by interviewees representing the technology function but not by the other interviewees.

End-User Top athletes were typically identified as endusers. Consumer experience was defined in much the same way by different respondent groups, as indicated in the following sample responses. It was not expected that the responses would be so similar to each other. The Suunto t6 is an incredible tool when you consider all the information that it gives to the user. The way it analyses training is in a class of its own. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


I don’t need to guess. I can train safely. I can avoid overtraining and train adequately. I don’t just do it, I do it smarter. As a trainer, I can achieve good training results in a rational way. It is unbelievable what quantities of important information this kind of tool can show. These responses show how clear the consumer benefit was for all interviewees concerning the successful product innovation. This emphasizes the importance of integrating the end-user knowledge base as a separate knowledge base into the traditional models incorporating the technology knowledge base and the market knowledge base.

405

interviewees. The managing director and the product manager mentioned the longer predicted payback time for this product. The innovation-base diversification was obvious in the case of the Suunto t6. Thus, it would have been useful if the business logic knowledge base could have been more widely shared and recognized within the organization. This emphasizes the usefulness of using a multidimensional product concept model and the importance of including the business logic knowledge base in this. Figure 2 summarizes the product concept of the wrist computer Suunto t6 as seen by the interviewees.

Discussion and Conclusions Brand Hardware-related brand indicators of the Suunto t6 include its round shape and the ‘five buttons’ logic. Software-related brand indicators, for example, are the user logic and information display. Responses conveying a surprise component varied considerably. Using the Suunto t6 has in some cases resulted in very positive experiences and in some cases caused quite strong disappointments, as indicated by the following sample points raised by the interviewees. This part of the interview is aimed at measuring consistency of the cross-functional product concept concerning brand values. Before starting to use the device, the user is required to specifically turn on the radio signal. This was perceived as a surprising inconvenience to the users. By means of using the Suunto t6 it is possible to make laboratory-level measures in every-day training. For the first time in my life I can really see how intensive my training is. It’s unbelievable that it’s possible to display this magnitude of information. It seems that including the surprise component into the product-concept model is justified. The consistency of the responses would have been misleadingly uniform without analysing the surprise component, too.

Business Logic Suunto Oy launched the Suunto t6 to a new target group, with which its sales organization was previously unfamiliar. This diversification effect caused by the product innovation Suunto t6 was identified only by two of the © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

Several preliminary conclusions can be suggested based on the new theory development and the case study. First, certain dimensions of the product concept can be particularly indicative of cross-functional knowledge creation, such as a concern with the end-user’s experience. This result is consistent with the finding by Nonaka, Toyama and Nagata (2000), that the product concept is a first explicit phase in the process of innovative knowledge creation. Second, the real levels of revolutionary knowledge and of awareness about the revolutionary innovation within the organization are related to the absorptive capacity of the organization. This result is consistent with research results presented by Cohen and Levinthal (1990). In the present study, the surprise component of the innovative product concept is a key indicator of the radical dimensions of the innovation. Investigating the surprise component can also be used to measure the strength of uniqueness in the selling proposal. Third, the present study suggests that the product concept may provide an early warning of innovation-based diversification. The literature on product innovation development and on diversification has not paid much attention to the value of innovative product concepts in creating diversification effects. This finding also has implications for diversification management practices. Fourth, a point that has been less discussed in the existing innovation literature is the possibility of using nicknames as indicators of cross-functional integration within the organization. The present study suggests that a nickname is a revealing indicator of tacit knowledge, since coming up with a nickname requires the verbalization of tacit subjective associations. Nicknames can be utilized as an

Volume 15

Number 4

2006

406


PRODUCT

TECHNOLOGY -Technical solution

-Features -Main functions -Unique selling proposal END-USER -Ethnology -Usage context -Needed value -Customer’s experience of the product in their own words BRAND -Promise of value -Values -Brand characters -Design hardware -Design software -Surprise

BUSINESS LOGIC

– Wrist computer Suunto t6 – “Personal trainer”

– Radio data transmission, algorithms to calculate training effects based on data between heart beats – Oxygen consumption, PC software: EPOC – Clock, pulse, trip – Training effect – Extrovert, professional athletic, high income – Turn on and train, after work if as hobby – Mostly for training, less promotional value – Advance planning of the training, explicit specification of the target training effects – “I don’t just train. I train smarter.”

– Replacing luck by controlled training impact – Control, end-user added value – Round shape, Suunto design – ‘Five buttons’ logic – Suunto user interface, tailored information – Laboratory-level training effect measurements in every-day training – New target group, slower profit flow growth – Training results for promotion, new channels

Figure 2. Summary of the Product Concept of the Wrist Computer Suunto t6 as Seen by the Interviewees

indicator of cross-functional integration in future research studies. Some additional remarks can also be made regarding the capacity of the product-concept model to measure knowledge: First, does the new product-concept model describe the future product adequately? The Suunto t6 was in the product concept phase. It had only recently been launched, and had not yet achieved market success as defined by Crawford (1996). The product-concept model described both the product innovation Suunto t6 itself, and its relationship to the four knowledge bases. The ‘product concept’ based on the four knowledge bases was something deeper than a mere ‘product idea’: the product concept realistically described the product innovation in its market context, and prepared for the demands imposed by the commercialization of the product innovation. Second, does the product-concept model introduced in the present study measure knowledge in a novel way? Some features of the product can be found in the technical descriptions, but these represent merely ‘information’. Other features, such as the design parameters, predicate company-specific explicit knowledge. Drafting a definition of the end-user experiences in one’s own

Volume 15

Number 4

2006

words, as required in the new product concept model, demands market insight, and the responses are therefore clearly based on tacit knowledge. Thus, the product-concept model now introduced does indeed measure a broad range of knowledge. The product-concept model introduced in the present study was able to measure cross-functional knowledge creation, and to identify corresponding meaningful differences between individual interviewees’ perceptions of the product concept, both explicit and tacit. The model describes both the radical innovative features in consumer experiences and the potential surprise elements concerning the product innovation. Such features are difficult to translate into any product conceptualization software, such as the Nortel software package (Montoya-Weiss & O’Driscoll, 2000). The elicited descriptions were effective in evaluating cross-functional knowledge creation, and the qualitative responses indicated the spectrum of cognitive models being deployed. The present study also has some limitations: First, people typically know more than they can express in verbal form (Polanyi, 1966). Obviously, many essential forms of tacit knowledge, such as product development pro© 2006 The Authors Journal compilation © 2006 Blackwell Publishing


cess skills, were not measured in the study. However, expression tasks such as those required in the conceptualizing process of the new product-concept model pose demands on conceptualizing tacit knowledge into a communicative form and sharing it. Second, the product-concept model measures only some of the relevant organizational skills. It measures organizational capabilities concerning commercializing innovations, but does not measure other resources of the organization, such as financial or marketing resources. Third, the product-concept model is a descriptive tool, and it is therefore advisable to use, in addition, quantitative measures. For example, in the case of the present study, the sales potential of the Suunto t6 should also be evaluated, to supplement the product-concept data. Similarly, measures based on estimations such as predicted market size can increase the organization’s understanding of the expected commercial success of the innovation. Fourth, the product-concept model is appropriate for providing preliminary responses in the initial stages of product development about the opportunities and challenges of the product innovation, but in the later development stages more detailed insight will be needed. This can be done by consistently using the same product-concept model and by also addressing the question by means of other, more detailed models. In this present study, the a priori framework has been discussed on the basis of one case study only. Since at this stage the results are exploratory, there is clearly a strong need to test the framework further with other case studies. However, there are doubts about the relevance of quantitative measures concerning research in cross-functional knowledgecreation, because of the decisive role of the tacit knowledge component.

Preliminary Managerial Implications On the basis of the present research study, the following preliminary managerial implications can be presented. The product development group members should define their personal product concepts in advance of group discussions. In Suunto Oy, this allowed a fuller utilization of the creative power of the organization and secured a greater variety of input to the cross-functional knowledge creation. From the very start of the development process, the project manager should repeatedly challenge the people involved to define in their own words the end-users’ predicted © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

407

experiences of using the product. This is a powerful tool in creating a better market fit. Planning positive surprises when developing the new product turned out to be a useful tool for cross-functional knowledge creation in Suunto Oy. These positive surprises might include, for example, an unexpected competitive advantage to be used in marketing. This will give inspiration to the whole development team and will spur creativity in further knowledge creation. This will also help to utilize the full potential of the opportunities offered by the new technology. The product-development team should be prepared for additional demands created by the product-innovation-based diversification by analysing the barriers of commercialization in order to identify the potential increase in resources needed in various functions of the company. Suunto Oy was no different from other innovative high-technology companies in which product-innovation-based diversification was not fully acknowledged in advance. This kind of diversification cannot be avoided; but what is crucial is how prepared the organization is for the consequences of this diversification. The project manager should collect product concept views from all stakeholders in the value net of the product innovation. The widest possible range of ideas is needed for effective cross-functional knowledge creation. From an early stage of development, the product-development team should develop and agree on a well-chosen shared nickname for the product innovation idea, to be used throughout the entire product-development organization. This will enhance the integration of the shared product concept and thus the cross-functional knowledge development. This can be one method for achieving the difficult task of integrating the possibly differing tacit aspects of the product concept. To achieve an even more comprehensive measurement of the relevant organizational capabilities, the future development of the model should include additional knowledge bases such as the marketing and manufacturing knowledge base, and the combinative knowledge base.

References Bharadwaj, S. and Menon, A. (2000) Making Innovation Happen in Organizations: Individual Creativity Mechanisms, Organizational Creativity Mechanisms or Both? Journal of Product Innovation Management, 17(6), 424–34. Cohen, W.M. and Levinthal, D.A. (1990) Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly, 35(1)

Volume 15

Number 4

2006

408


(Special Issue: Technology, Organizations, and Innovation, March), 128–52. Cooper, R.G. (1975) Why New Industrial Products Fail. Industrial Marketing Management, 4(4), 315– 26. Crawford, M.C. (1996) New Products Management, 5th edn. Irwin Professional Publishing, Burr Ridge, IL. Gomes, J., de Weerd–Nederhof, P.C., Pearson, A. and Fisscher, O.A.M. (2001) Senior Management Support in the New Product Development Process. Creativity and Innovation Management, 10(4), 234–42. Hänninen, S. (2006) The ‘Perfect Technology Syndrome’: Sources, Consequences, and Solutions. International Journal of Technology Management. (forthcoming) Hänninen, S. and Kauranen, I. (2006) Product Innovation as Micro-Strategy. International Journal of Innovation and Learning. (forthcoming) Hänninen, S., Kauranen, I., Serkkola, A. and Ikävalko, J. (2006) Barrier to Commercialisation from the ‘Four Knowledge Bases’ Perspective: A Study of Innovation in the Software Development Sector. International Journal of Management Practice, 2(3). (forthcoming) Hänninen, S. and Sandberg, B. (2006) Consumer Learning Roadmap: a Necessary Tool for New Products. International Journal of Knowledge and Learning, 2(3/4), 298–307. Kapferer, J-N. (1994) Strategic Brand Management: New Approaches to Creating and Evaluating Brand Equity. Free Press, New York. Kim, W.C. and Mauborgne, R. (1997) Value Innovation: The Strategic Logic of High Growth. Harvard Business Review, 75(1), 103–12. Kogut, B. and Zander, U. (1992) Knowledge of the Firm, Combinative Capabilities, and the Replication of Technology. Organization Science, 3(3), 383–97. Leonard-Barton, D. (1992) Core Capabilities and Core Rigidities: A Paradox in Managing New Product Development. Strategic Management Journal, 13(3) (Special Issue: Strategy Process), 111–25. Madhavan, R. and Grover, R. (1998) From Embedded Knowledge to Embodied Knowledge: New Product Development as Knowledge Management. Journal of Marketing, 62(10), 1–12. Montoya-Weiss, M.M. and O’Driscoll, T.M. (2000) From Experience: Applying Performance Support Technology in the Fuzzy Front End. Journal of Product Innovation Management, 17(2), 143–61. Nonaka, I., Toyama, R. and Nagata, A. (2000) A Firm as a Knowledge-Creating Entity: A New Perspective on the Theory of the Firm. Industrial and Corporate Change, 9(1), 1–20. Normann, R. (1971) Organizational Innovativeness: Product Variation and Reorientation. Administrative Science Quarterly, 16(2), 203–15. Orihata, M. and Watanabe, C. (2000) The Interaction between Product Concept and Institutional Inducement: a New Driver of Product Innovation. Technovation, 20(1), 11–23.

Volume 15

Number 4

2006

PDMA (2005) The PDMA Glossary for New Product Development. Available at: http://www. pdma.org/library/-glossary.html?PHPSESSID= 6086d8456711f44e0f4122bc5fa859d4. Polanyi, M. (1966) The Tacit Dimension. Anchor Day Books, New York. Porter, M. (1980) Competitive Strategy. Free Press, New York. Prahalad, C.K. and Bettis, R.A. (1986) The Dominant Logic: A New Linkage between Diversity and Performance. Strategic Management Journal, 7(6), 485–501. Prahalad, C.K. and Ramaswamy, V. (2003) The New Frontier of Experience Innovation. MIT Sloan Management Review, 44(4), 12–18. Scholtz, J. and Salvador, T. (1998) Systematic Creativity: A Bridge for the Gaps in the Software Development Process. In Wood, Larry (ed.), User Interface Design: Bridging the Gap from User Requirements to Design. CRC Press, Boston, IL, pp. 215–44. Stake, R.E. (1995) The Art of Case Study Research. Sage Publications, Thousand Oaks, CA. Tushman, M.L. and Anderson, P. (1986) Technological Discontinuities and Organizational Environments. Administrative Science Quarterly, 31(3), 439–65. Wallin, J. (2000) Customer Orientation and Competence Building. A doctoral dissertation, Acta Polytechnica Scandinavica, The Finnish Academy of Technology, Industrial Management Series No:6, Espoo, Finland. Walsh, J.P. (1995) Managerial and Organizational Cognition: Notes from a Trip Down Memory Lane. Organization Science, 6(3), 280–321. Wood, L. (ed.) (1998) User Interface Design: Bridging the Gap from User Requirements to Design. CRC Press, Boston, IL.

Seppo Hänninen (seppo.j.hanninen@tkk.fi) is researcher and doctoral candidate at the Department of Industrial Engineering and Management at Helsinki University of Technology. He also has eight years of job experience as an advertising planner. His research interests include commercializing technological inventions, technology marketing and technology-based companies. Ilkka Kauranen ([email protected]) is Visiting Professor at the School of Management at the Asia Institute of Technology in Thailand. He holds a professorship in Development and Management in Industry at the Department of Industrial Engineering and Management at Helsinki University of Technology. He has research interests in technology-based companies, entrepreneurship and commercializing technological inventions.



Appendix 1. Research Questionaire General questions – What is the name of the product? – What is the type number of the product? – What is the nickname of the product? Technological questions – Which are the underlying technologies? – Which solutions do the technologies offer? – Which critical features do the technologies offer? – What are the added value functions? – Is there a single unique benefit? End-user questions – – – – –

What is the personality of the end-user? What is the profession of the end-user? Which are the hobbies of the end-user? What is the consumerism of the end-user? How do you describe the real-world behavioral patterns in brief?


409

– How is the product really used? – Which customer needs does the product satisfy? – Imagine yourshelf as the end-user: ‘What is my experience of using the product?’ Brand questions – How do you describe the nature of expectations, e.g. from the brand slogan? – Which are the main values of the brand? – Which are the concretely verifiable characters of the brand? – Which are the brand indicators related to the physical form of the product? – Which are the software’s brand characters? – Does the product have something unexpected? Business-logic questions – What are the key drivers of profit flow? – What are the solutions to the barriers of profit flow?

Volume 15

Number 4

2006

410


Do Localized Clusters Influence Creativity of Inventors? Sherwat E. Ibrahim, M. Hosein Fallah and Richard R. Reilly Do localized clusters affect the creativity of inventors? What contributes to the creative environment of localized clusters that increases innovation output? This paper presents the results of a study investigating the environment of localized clusters and how it affects the creativity of individual inventors working in those clusters. The subjects for the study are inventors in the telecommunication industry who filed for patents individually or collectively. Localized clusters for telecoms are identified, and inventors are asked to rate the different influences existing in their local environment and how they affected the way the inventors came up with the ideas behind their inventions. Local influences are categorized under direct sources of ideas, stimuli in the local environment, and situations that take place locally. The ratings given to the importance of each category are evaluated and compared to each other, focusing on what to look for in clusters in order to further enhance the creativity of the innovators in those areas.

Introduction

A

localized cluster is a geographical area with a high concentration of interrelated firms. A body of knowledge has evolved around localized clusters and the economic growth of certain regions (Enright, 1991; Krugman, 1994; Porter, 1990, 1998, 2003; Pouder and St. John, 1996). For example, in Europe, watchmakers clustered in Switzerland and fashion designers in Paris. In the United States, well known clusters include Detroit for automotives, Hollywood for motion pictures, New York City for financial services and advertising, and Silicon Valley for electronics. There has been increasing interest by many researchers to understand the causes of such growth. In the view of some researchers, the clustering phenomenon creates externalities that lead to increased innovation fuelling the economic growth (Marshall, 1920; Krugman, 1991, 1994). The literature dealing with regional agglomeration and regional innovation has been studied mainly under the research umbrella of economic geography1 where these hot spots have been given desig1 Also called New Economic Geography (NEG) by Paul Krugman (1991, 1995, 1998, 1999).

Volume 15

Number 4

doi:10.1111/j.1467-8691.2006.00408.x

2006

nations such as ‘geographical clusters’, ‘industrial districts’, ‘technological clusters’, ‘learning regions’ and ‘innovation milieux’.

Creativity and Patents In this study we refer to creativity as the ability to come up with ideas that could lead to new inventions. New inventions are generally protected by patents. Inventions that are developed and commercialized successfully are considered as innovations. The creative process usually starts with the identification of a problem that needs to be solved followed by a search for possible alternatives that could provide a solution. This search is highly dependent on the individual’s view of the problem, where a person explores his/her past experiences and knowledge for a solution. In cases where a person cannot find a solution on his/her own, he/she reaches out to other people and organizational sources of knowledge (Farr and Ford, 1990), making it important for organizations to understand which kinds of managerial practices foster knowledge sharing and creativity, and which practices inhibit them (Amabile, 1998). Each individual has a favoured cognitive style for problem solving. The ideas and approaches to solutions would © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

DO LOCALIZED CLUSTERS ENHANCE CREATIVITY?

differ as a result of variation in individual cognitive styles. Regardless of how a solution is created, the output is often captured as an invention in a patent. The United States Patent and Trademark Office (USPTO) maintains a database of patents and patent applications belonging to individuals or groups of individuals as owners of particular inventions. Research has shown that localized clusters have a higher level of creative output as measured by the number of patents issued to individuals and corporations in those clusters (Jaffe, 1993, 2000).

Idea Generation Smith et al. (1999) link ‘idea generation’ to the concept of ‘knowledge creation’ and explains that idea generation comes as a result of grouping and integrating the sources of established knowledge. Roberts and Fusfeld (1981) identify idea generation as one of the most important and critical parts of innovation activity. Koen and Kohli (1998) recognize idea generation as a vital part in the front-end of the innovation process. McAdam and McClelland (2002) point out that there are two aspects to the ‘front-end’ of the innovation process in organizations, namely ‘knowledge creation’ and ‘idea generation’ and that they are both key elements for creativity and the process of innovation. They suggest three constructs that lead to the generation of new ideas. The first construct, ‘segregation’, stresses the importance of separating ‘idea generation’ from ‘idea evaluation’. Idea screening and filtering should be done at a later stage. Unexploited ideas should be logged for further application or for when an opportunity arises. The second construct, ‘structure’, refers to organizational practices that enhance the generation of ideas, most popularly ‘brainstorming’ (Osborn, 1963). The third construct, ‘strategic intent’, calls for a matching between the newly generated ideas and the organizational goals. McAdam (2004) examines the role of individuals and teams in idea generation and how they affect the creativity and innovation process in organizations. He reviews the literature on idea generation and identifies the need for more research at the front-end of the innovation process.

Sources of Ideas While research on sources of ideas for new products has been continuing for two decades (Cooper and Kleinschmidt, 1986; von Hippel, © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

411

1986, 1988; Yoon and Lilien, 1988; Coates et al., 1996), more empirical studies linking sources of ideas to creativity and the front-end of the innovation process are needed (Koen and Kohli, 1998). McAdam (2004) points to the limited number of studies on idea sources and the lack of sector-specific analysis across industries. He points to a lack of research examining why companies choose and exploit particular sources of ideas. The focus of past research has been on a source-based approach to idea generation with emphasis on internal knowledge sources like R&D and sales, and external knowledge sources such as customers, markets and competitors. Von Hippel (1986) has suggested ‘lead user’ perceptions and preferences as sources of ideas for new products. He surveyed the sources of ideas coming from users, manufacturers and suppliers and found that the results varied by industry (von Hippel, 1988). Sowrey (1987, 1989) studied the organizational sources and techniques for generating ideas for new consumer market products and concluded that marketing people followed by R&D were mostly responsible for delivering new ideas. Yoon and Lilien (1988) found that ‘customer requests’ followed by ‘market research’ were considered the most important sources of ideas by manufacturers. Guimaraes and Langley (1994) surveyed 108 US companies and found that internal teams and employees were the most effective source of new ideas. They also found that the most commonly used ‘methods’ to generate ideas were internal meetings and brainstorming sessions. Ibrahim and Fallah (2005) identified the internal organization environment as the most important source of new ideas. Koen and Kohli (1998) surveyed companies for the different sources that led to the initial ideas for their product innovations. The best sources of ideas varied depending upon whether the innovation was radical, platform or incremental. R&D engineers and scientists, customers, senior sales managers and operations engineers were selected as the most valuable sources overall. Interactions between customers and an organization’s engineers and scientists were considered the most important source of radical and profitable ideas.

Sources of Ideas in Localized Clusters As shown, the current literature on sources of ideas focuses mainly on internal and organization-related sources. However, economic geographers have recognized the effect of regional clustering on the generation of new

Volume 15

Number 4

2006

412


WWW

Direct Factors found in Clusters

TRAVELING

Company factors Company arranged Situations

Stimulus from Company

Stimuli that exist in Clusters

Situations that take place in clusters

Figure 1. Sources of Ideas Affecting Individual Creativity

ideas in terms of patents and increased innovation (Feldman, 1994; Lawson and Lorenz, 1999; Acs et al., 2002). This study attempts to bridge the relationship between these two research areas, namely sources of ideas and localization. We classify the sources of ideas affecting an innovator’s creativity into internal and external influences as depicted in Figure 1, and focus only on the ‘localized external influences’ affecting an inventor in a particular region (the white realm in Figure 1). Internal influences related to the inventor’s own organization, as well as the global external influences related to travelling and the use of the Internet and remote communications mechanisms are outside the scope of this paper. To better understand the dynamics of this relationship, all possible sources of ideas an inventor could use in coming up with his/her invention were identified and tested on a sample of inventors. In doing so, we relied on previous literature, brainstorming with other experts and interviews with local inventors. We found the sources of ideas to generally fall under three categories: interactions that an inventor could point to as directly having affected his/her invention, events that he/she could recall participating in that affected his/ her thinking about the invention and the circumstances of the local environment that he/

Volume 15

Number 4

2006

she believed were an influence in coming up with the invention. We group these ‘localized external influences’ that affect the way inventors come up with their ideas into the following three categories and test for them separately: • Direct localized factors. In this study, localized factors refer to the local people and local material (external to the inventors’ organizations) found in the geographic region that could have directly influenced the inventors’ creativity. Questions tested under this category include one-on-one interactions with other professionals in the cluster, access to working papers and documents available locally, and access to local products and prototypes. • Localized situations. Rather than identifying a particular local person or local product that affected the inventors, this category of localized situations measures the effect of a particular local setting on them. A list of possible events and situations that take place locally (outside and without the involvement of the inventors’ organizations) that could influence the inventors’ creativity were identified and tested. Examples include brainstorming sessions, meetings, social gatherings, conferences, seminars and fairs. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


413

• Localized stimuli. This category includes general influences of the localized cluster environment (regardless of the inventors’ organization) that could influence the creativity of the inventors. Questions of the ability to monitor state-of-the-art innovations and emerging techniques related to the problem being solved by the inventor, being in the area, are tested. The long-term effect of personal relationships and ongoing interactions with locals can also indirectly influence inventors’ thinking and motivation and the way they come up with their ideas. This category captures any influences that could not be explicitly expressed but could have affected the creation of the invention.

Table 1. US Telecommunications Clusters

The Study

Survey Design

Identifying the Clusters

We created a survey considering all sources of knowledge that could potentially influence inventors. In particular, the survey included a series of specific questions on the direct local factors, local situations and local stimuli. Specific questions regarding the localized sources of knowledge for this paper are listed in Table 2. The survey was customized for each respondent with the inventor’s name, coinventors (if any) and the specific patent number and title as the focus of the survey. This was followed by an introduction to the study and a description of what the study intended to measure. The introduction provided the definition of ‘technological clusters’ and asked any inventor who did not consider him/herself to be in a cluster to refer to the local city he/she resided in, and the neighbouring cities within a 50-mile radius. Also, the need to differentiate between geographical-related factors and organizational- or company-related factors that affected the inventor’s creativity in the course of the relevant invention is explained and elaborated. At the start of the survey, general information was collected about the inventor and his/ her surroundings. This included confirmation of the city of residence during the invention, the name and size (in terms of the number of employees) of the company the inventor worked for during the invention, and the motivation for the invention. During the survey design we interviewed three inventors on the structure of the survey and kinds of questions we intended to ask. The initial version of the survey was then tested with 15 inventors in New Jersey. These respondents were asked to complete the survey and comment on the clarity of the questions, their ability to answer the questions and

As defined earlier, technological clusters are geographical regions with a high concentration of innovative firms and people forming around a particular industry. In this study, we focused on one industry, telecommunications. We identified the telecom clusters in the United States by viewing the distribution of a sample of telecommunication patents, under USPTO code 455. The sample consisted of the patents granted for the period from January 2000 to October 2003. International patents were eliminated leaving a total sample of 5353 to sort. To extract the cities with the highest patent output, the 5353 telecommunication patents were sorted by first inventor’s city and state, and a count per city was carried out. The top 49 cities (each with 19 or more patents which was the average of patents per city) were selected and plotted on the United States map. Then a circle with a 50-mile radius from the centre of each city was drawn to determine the cluster boundaries.2 Due to the closeness of some cities to each other, this process led to the identification of 15 clusters. The cluster boundaries for these overlapping circles would be the external border around the overlapping circles. From this set of clusters, we selected the top 7 clusters with the highest patent counts for our study (those with patent counts above the median of the clusters). These clusters are shown in Figure 2 and listed in Table 1.

2 According to Anselin et al. (2000), 50 to 75 miles is considered a reasonable commuting distance to define a cluster.


Cluster

Dallas, TX Research Triangle, NC Silicon Valley, CA San Diego, CA NY/NJ Chicago, IL Los Angeles, CA

Number of Telecom Patents from 1 January 2000 to 1 October 2003 569 348 328 322 309 266 169

Volume 15

Number 4

2006

414


Figure 2. Top Seven Telecommunication Clusters in the United States offer any suggestions on how to improve the survey. The survey was then refined and finalized before administering it to the study subjects. In the survey the respondents were asked to consider the particular invention and rate the influence of each of the specific survey items on a Likert-type scale (1 = no influence to 5 = strong influence) on his/her coming up with the problem or the solution that led to the particular invention.

Data Collection From the USPTO database, we identified a total of 1954 recently published telecommunications patent applications (from January 2003 to December 2004). A sample of 810 was randomly selected from this population (eliminating international patent applications). We compiled a list of the names and addresses of the first authors of the selected sample. The cluster map would allow us to locate each inventor in our sample and determine if he/ she belonged to one of the previously identified clusters or was outside of the cluster boundaries. Our target was to contact each inventor in our sample by telephone. We were able to reach 390 inventors to whom we

Volume 15

Number 4

2006

explained the survey and asked if they were willing to participate in the study. Each inventor who agreed to participate in the study received the survey via email. We were able to collect survey responses from 165 of the inventors in our sample.

Hypotheses The following hypotheses test the significance of the influence of localized clusters on the creativity of inventors. H1: Inventors in localized clusters will rate the influence of direct localized factors from their geographical area (FACTORS) higher than inventors in non-clusters. H2: Inventors in localized clusters will rate the influence of localized situations from their geographical area (SITUATIONS) higher than inventors in non-clusters. H3: Inventors in localized clusters will rate the influence of localized stimuli from their geographical area (STIMULI) higher than inventors in non-clusters. Each of these influences is measured through several specific questions in the survey. Table 2 lists the dimensions of each variable. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


415

Table 2. Dimensions of the Independent Variables Variable

Scale Dimensions

FACTORS

• Accessing a particular publication or paper from a local author • Interacting with a subject matter expert (SME) in the same profession in the local area • Interacting with a person working in a different profession in the local area • Identified informal discussions with colleagues who work in the local area • Seeing a product or prototype developed in local area • A brainstorming session with people in the local area • A formal meeting with people in the local area • An informal meeting with people in the local area • A knowledge sharing session with people in the local area • A social gathering in the local area • A conference, seminar and/or workshop in the local area • A science, technology fair or symposium in the local area • Knowledge gained from monitoring the emerging technologies in the local area • Knowledge gained from reviewing state-of-the-art innovations in the local area • Being presented with a problem or need in the local area • Having ongoing interactions with customers, suppliers, competitors in the local area • Personal relationships developed over time with other researchers in the local area • The overall working environment of the local area

SITUATIONS

STIMULI

Table 3. Logistic Regression Analyses Hypothesis H1: H2: H3:

Variable

B

S.E.

Wald

d.f.

Sig.

Exp (B)

FACTORS SITUATIONS STIMULI

0.473 0.361 0.405

0.242 0.227 0.189

3.820 2.522 4.581

1 1 1

0.025 0.056 0.016

1.604 1.434 1.500

Scale Reliability All variables showed a high degree of internal consistency with SITUATIONS showing the highest at (0.895) followed by STIMULI at (0.868) and FACTORS at (0.805).

Study Findings We used logistic regression to test the stated hypotheses. Logistic regression coefficients were used to measure the strength of association between the stated influences affecting creativity of the inventor and the probability © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

of the respondent being in a localized cluster. The results of the regression analysis are summarized in Table 3. The results show support for all three hypotheses. Hypotheses 1 and 3 are strongly supported (p < 0.05) and H2 shows a trend toward significance (p = 0.056). Table 4 lists comparative statistics for all components of the influences and the significance levels between the means for respondents from clusters and respondents from non-cluster. Not all the items considered within each category of influences, however, are significant.

Volume 15

Number 4

2006

416


Table 4. Summary Statistics for Creativity Influences on Innovation within and outside Localized Clusters Variable

FACTORS: Summary Variable for Direct Local Factors that Influence Creativity • Accessing a particular publication or paper from a local author • Interacting with a subject matter expert in the same profession in the local area • Interacting with a person working in a different profession in the local area • Identified informal discussions with colleagues who work in the local area • Seeing a product or prototype developed in local area SITUATIONS: Summary Variable for Local Situations that Influence Creativity • A brainstorming session with people in the local area • A formal meeting with people in the local area • An informal meeting with people in the local area • A knowledge sharing session with people in the local area • A social gathering in the local area • A conference, seminar and/or workshop in the local area • A science, technology fair or symposium in the local area STIMULI: Summary Variable for Local Stimuli that Influence Creativity • Knowledge gained from monitoring the emerging technologies in the local area • Knowledge gained from reviewing state-of-the-art innovations in the local area • Being presented with a problem or need in the local area • Having interactions with customers, suppliers, competitors in the local area • Personal relationships developed with other researchers in the local area • The working environment of the local area

Localized Cluster

Non-Cluster

P value (1-tailed)

N

Mean

St.dev.

N

Mean

St.dev.

83

1.53

0.83

80

1.30

0.58

0.021*

80

1.44

0.99

80

1.28

0.78

0.125

82

1.76

1.25

80

1.36

0.93

0.012*

83

1.48

1.07

80

1.21

0.65

0.027*

80

1.80

1.25

79

1.61

1.25

0.167

81

1.44

1.01

80

1.19

0.64

0.028*

84

1.50

0.87

80

1.31

0.57

0.051

82

1.76

1.23

80

1.39

0.86

0.014*

82

1.55

1.11

80

1.43

1.04

0.233

82

1.56

1.07

80

1.31

0.77

0.046*

82

1.51

1.10

80

1.24

0.75

0.033*

81 82

1.30 1.40

0.83 0.97

80 80

1.25 1.33

0.77 0.91

0.357 0.300

81

1.37

0.93

80

1.25

0.70

0.178

84

1.84

0.97

80

1.54

0.76

0.014*

84

1.86

1.21

80

1.76

1.08

0.300

84

1.82

1.22

80

1.51

0.99

0.039*

84

1.77

1.25

80

1.48

0.98

0.044*

82

1.68

1.16

79

1.42

0.91

0.055

83

1.87

1.34

80

1.55

1.07

0.048*

84

2.04

1.27

79

1.56

0.96

0.003**

Note: *significant at 0.05; **significant at 0.01.

Volume 15

Number 4

2006



For the direct localized factors, interactions with subject matter experts and reviewing products and prototypes produced locally had a strong influence on creativity. Papers and publications and identified informal talks were not different within and outside localized clusters. In terms of specific situations, brainstorming sessions, informal technical meetings and knowledge sharing forums in localized cluster were significant. In terms of creativity stimuli, ongoing interactions with customers, suppliers and competitors had a similar effect whether or not these entities were localized. The environment of the cluster, however, showed the strongest statistical significance.

Discussion and Implications Corporations in related industries cluster in specific geographical areas to benefit from the resources available in those areas. It is well recognized from the literature that localized clusters have higher innovative output. This study provides insight into the ways localization can influence innovation by measuring the effect of the different local sources on inventors’ creativity. While all three categories of localized sources were rated higher by inventors located in clusters than by those not in clusters, the different dimensions that compose each category were rated differently, providing insight into the influence of specific attributes. Attending seminars, forums and conferences, for example, did not significantly differentiate inventors in clusters from other inventors. This does not imply that such events have no effect on creativity. It shows that individuals within and outside clusters have similar opportunities to participate in such forums and there is no significant advantage to those being in the clusters. A similar argument applies to the knowledge gained from monitoring emerging technologies, since perhaps such reviews and scouting can be done from anywhere. On the other hand, the working environment of the cluster was rated as most significant. This finding points to some of the intangible attributes and motivators of creativity that exist by simply being in a vibrant and active geographical area. Corporate executives can promote creativity and innovation in their organizations by paying attention to the effect of localization. Companies operating in clusters, have the opportunity to take advantage of the local sources and the intangibles that exist in the cluster environment. Those outside clusters need to identify the different ways inventors in non-clusters compensate for the lack of external resources and provide for it. Further research is needed in this area. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

417

Conclusions This paper shows that localized clusters influence inventors’ creativity. The authors present the results of a study identifying three categories of influences namely direct local factors, local situations presenting themselves to inventors locally, and local stimuli found in the cluster environment. Analysing a survey of inventors in the telecommunication industry, the authors find that respondents in localized clusters rate these influences significantly higher than respondents in noncluster areas, indicating a local advantage. This study supports earlier research on geographical clusters and provides further insights into the attributes that contribute to innovation output of clusters. The findings have implications for management of R&D in localized clusters and for policy makers promoting development of clusters and economic growth of their regions.

References Acs, Z.J., Anselin, L. and Varga, A. (2002) Patents and innovation counts as measures of regional production of new knowledge, Research Policy, 31, 1069–85. Amabile, T.M. (1998) How to kill creativity, Harvard Business Review, 76, 76–87. Anselin, L., Varga, A. and Acs, Z.J. (2000) ‘Geographic spillovers and university research: a spatial econometric perspective’, in Nijkamp, P. and Stough, R. (eds), Special Issue on Endogenous Growth: Models and Regional Policy, Growth and Change, 31, 501–16. Coates, N.F., Cook, I. and Robinson, H. (1996) Idea generation techniques in an industrial market, Journal of Marketing Practice, 4(3), 107– 18. Cooper, R.G. and Kleinschmidt, E.J. (1986) An investigation into the new product process: steps, deficiencies, and impact, Journal of Product Innovation Management, 3(2), June, 71–85. Enright, M.J. (1991) ‘Geographic concentration and industrial organization’, Unpublished doctoral dissertation, Harvard University, Cambridge, MA. Farr, J.L. and Ford, C.M. (1990) In West, M.A. and Farr, J.L. (eds), Individual Innovation: Innovation and Creativity at Work, John Wiley & Sons, Chichester. Feldman, M.P. (1994) The Geography of Innovation, Kluwer Academic Publishers, Boston, MA. Guimaraes, T. and Langley, K. (1994) Developing innovation benchmarks: an empirical study, Benchmarking for Quality Management & Technology, 1(3), 3–20. Ibrahim, S. and Hosein, F.M. (2005) Where do inventors get their ideas? PICMET 05, Portland, August. Jaffe, A.B., Trajtenberg M. and Henderson, R. (1993) Geographic localization of knowledge spillovers

Volume 15

Number 4

2006

418


as evidenced by patent citations, Quarterly Journal of Economics, 108, 577–598. Jaffe, A.B., Trajtenberg, M. and Fogarty, M.R. (2000) Knowledge spillovers and patent citations: Evidence from a survey of inventors, American Economic Review, 9(2), 215–19. Koen, P.A. and Kohli, P. (1998) Idea generation: Who has the most profitable ideas, Engineering Management Journal, 10(4), 35–41. Krugman, P. (1991) Geography and Trade, MIT Press, Cambridge, MA. Krugman, P. (1994) ‘Location and competition: Notes on economic geography’, in Rumelt, R.P. Schendel, D.E. and Teece, D.J. (eds), Fundamental Issues in Strategy, Harvard Business School Press, Boston, MA, pp. 463–93. Krugman, P. (1995) Development, Geography and Economic Theory, MIT Press, Cambridge, MA. Krugman, P. (1998) What’s new about economic geography, Oxford Review of Economic Policy, 14(2), 7–17. Krugman, P. (1999) The role of geography in development, International Regional Science Review, 22, 142–61. Lawson, C. and Lorenz, E. (1999) Collective learning, tacit knowledge and regional innovative capacity, Regional Studies, 33, 305–318. McAdam, R. and McClelland, J. (2002) Individual and team-based idea generation within innovation management: Organisational and research agendas, European Journal of Innovation Management, 5(2), 86–98. McAdam, R. (2004) Knowledge creation and idea generation: a critical quality perspective, Technovation, 24(9), 697. Marshall, A. (1920) Principles of Economics, Macmillan, London. Osborn, A.F. (1963) Applied Imagination, Scribners, New York. Porter, M.E. (1990) The Competitive Advantage of Nations, Free Press, New York. Porter, M.E. (1998) Clusters and the new economics of competition, Harvard Business Review, 76(6), 77–90. Porter, M.E. (2003) The economic performance of regions, Regional Studies, 37(6–7), 549. Pouder, R. and St. John, C.H. (1996) Hot spots and blind spots: Geographical clusters of firms and innovation. Academy of Management Review, 21(4): 1192–1225. Roberts, E.B. and Fusfeld, A.R. (1981) Staffing the innovative technology based organizations, Sloan Management Review, 22(3), 19–34. Smith, G.R., Herbein, W.C. and Morris, R.C. (1999) Front-end innovation at AlliedSignal and Alcoa, Research Technology Management, 42(6), 15–24. Sowrey, T. (1987) The Generation of Ideas for New Products, Kogan Page, London. Sowrey, T. (1989) Idea generation: identifying the most useful techniques, European Journal of Marketing, 24(5), 20–9. von Hippel, E. (1986) Lead users: a source of novel product concepts, Management Science, 32(7), July, 791–805. von Hippel, E. (1988) The Sources of Innovation, Cambridge University Press, Cambridge.

Volume 15

Number 4

2006

Yoon, E. and Lilien, G.L. (1988) Characteristics of the industrial distributor’s innovation activities: an exploratory study, Journal of Product Innovation Management, 5(3), 227–40.

Sherwat E. Ibrahim holds a PhD in Technology Management from Stevens Institute of Technology. Dr Ibrahim also has a degree in Finance from Helwan University in Cairo, Egypt, and has worked as a cost analyst for Procter and Gamble Inc. At Stevens, Dr Ibrahim has taught courses in financial and managerial accounting. Her previous research focused on technology transfer and developing countries. Her current research is in the area of knowledge management and transfer and knowledge creation in technological clusters. M. Hosein Fallah ([email protected]) holds a PhD in Applied Science from the University of Delaware and is an Associate Professor of Technology Management at Stevens Institute of Technology. Before joining Stevens, Dr Fallah was Director of Network Planning and Systems Engineering at Bell Laboratories. He has over 25 years of experience in the areas of product/service realization, software engineering, project management, business process reengineering, innovation and R&D effectiveness. His research at Stevens is centered on technological clusters and management of innovation. Dr Fallah has over 30 publications related to technology clusters, innovation and process management. Richard R. Reilly holds a PhD in Organizational Psychology from the University of Tennessee and is a Research Professor in the Howe School of Technology Management. Before joining Stevens, Dr Reilly was a research psychologist for Bell Laboratories, the Educational Testing Service and AT&T and has been a consultant to Fortune 500 and governmental organizations. He is on the Editorial Board of Personnel Psychology, and is a Fellow of the American Psychological Association and the American Psychological Society. He has over 60 publications related to organizational behaviour and team performance. Recent publications include Blockbusters: The Five Keys to Developing Great New Products (HarperCollins, 2002), and How to Build a Blockbuster, Harvard Business Review, 80(10), 10–19. Dr Reilly is the Director of the PhD programme at the Howe School of Technology Management, Stevens Institute of Technology.


PHRONESIS AND CREATIVITY

419

Phronesis and Creativity: Knowledge Work in Video Game Development Peter Zackariasson, Alexander Styhre and Timothy L. Wilson This article presents a study of the knowledge work involved in the development of video games. The success of these games is based on the ability to create a sense of immersion for the gamers. In the case presented here, dedicated gamers were also preferred when hiring personnel to develop the games. Speaking about the know-how of this specific group in terms of phronesis, the detailed and practical understanding of a particular field, enables an understanding of the idiosyncratic competence of this group and its importance for the development process. The video game development process is also structured to enable an open-ended process under the continuous influence of the gamers. The article concludes that innovative and creative work needs to be able to exploit a variety of competencies and that the notion of phronesis has to date been relatively under-theorized and therefore deserves more detailed attention.

Introduction ne of the most dynamic and creative industries today is the video game industry. The processes that are involved when producing a video game range from technical competences to artistic skills. This industry has also in a very short time evolved to a size that can be compared with the Hollywood film industry, both economically and in terms of consumer involvement. Consequently, whether a gamer or not, the way in which the creative and economic potential of this industry has been actualized has to be admired. Games have progressed from the moving dot of the video game Pong, to the icons in Pac-man, to the three-dimensional ‘real-life’ figures and objects that are now movie-like with manipulative endings. Needless to say, the creative and innovative process of producing video games has posed new challenges in terms of the management of new product development processes. It is at the same time claimed that we are living in a ‘knowledge society’ wherein knowledge is assigned a more central role than in previous economic eras. This practice has subsequently led to an increase in the interest

O


regarding knowledge as an organizational asset. This trend is visible both in the contemporary press, covering success stories of ‘knowledge organizations’, and in academia where the spread of ‘knowledge whatever’ education has made an imprint in curricula. Developing video games is a specific form of knowledge work. However, the video game development industry as such has been stigmatized, probably because of the very product it provides, namely games. Management research has failed to acknowledge the creativity and innovativeness of industries producing fun and frivolity (Rehn, 2004) and has instead tended to emphasize their immaturity and playfulness in a negative sense. Such accounts fail to appreciate the unique features of the video game development process, the combination of the adherence to strict rules or codes and the artistic creativity aimed at producing something fun and what gamers call an immersive experience. The aim of this article is to highlight the knowledge-based and creative work in one Swedish company producing video games. In the study, the concept of phronesis, defined here as having ‘street smarts’, will be employed. Specifically, this concept reflects the ability to operate within a

Volume 15

Number 4

2006

doi:10.1111/j.1467-8691.2006.00400.x

420


specific domain, and is a term much neglected in the mainstream business literature addressing knowledge. The article concludes that creative and knowledge-intensive firms would be well served to exploit and draw on the phronesis of its personnel.

Video Game Development One can argue that video games have almost been around as long as computers have. In a history of video games Kent (2001) ascribes the first game to, at that time, the MIT student Steve Russell. The game was Space Wars, constructed for the mini-computer, and in 1962, when it was developed, it became very popular. After the impact of this first game many other games were spread among users, or gamers. Most, if not all, of these games were developed by private individuals and distributed through friends, colleagues, and over networks of computers. The ‘professional’ market for video games started to take off in the 1970s with bulky arcade-machines. At that time the most popular game on this platform was the by now legendary, Pong; a tennis game for two. In the early 1980s consoles were introduced on the home-entertainment market, and initially these became highly popular, only to later be replaced by the PC. At this point in time, consoles have again reclaimed the market as the platform with the highest market shares. Video games are constantly driven to become more technically advanced and the barrier to entry into this market as a developer is getting increasingly costly. From the perspective of the video game developer, there is a fine balance between pushing the game technology (mainly the graphical engine) to what is wanted and what the customers will equip their computers with (graphic card, processor, and RAM-memory) during the production cycle – a miscalculation in technology adoption could be the difference between a top ranking and a failure. Considering that the development costs usually start at $1 and $1.5 million, depending on genre and innovativeness and the like, there is little room for mistakes. This increase in stakes has also meant a professionalism of the business as a whole, that is, it is no longer possible to produce quality video games without major funding, expertise and experience.

Knowledge and Creativity Labour markets today strongly emphasize the personal knowledge of individual co-workers,

Volume 15

Number 4

2006

that is, knowledge and know-how making individuals indispensable for the company. Stewart argues ‘Knowledge has become the most important factor in economic life. It is the chief ingredient of what we buy and sell, the raw material with which we work. Intellectual capital – not natural resources, machinery, or even financial capital – has become the one indispensable asset of corporations’ (1997). However, similar claims have been made about the critical importance of knowledge in the past; Marshall had already suggested 116 years ago that ‘Capital consist in a great part of knowledge and organisation . . . Knowledge is our most powerful engine of production’ (1972). The concepts of creativity and knowledge are closely associated, and adequate knowledge is in most cases regarded a conditio sine qua non for all creative undertakings. In the knowledge management literature, a number of taxonomies and typologies of knowledge are suggested (see for example, Tell, 2004; Yanow, 2004). Knowledge is consequently conceptualized into a number of categories, for instance, ‘tacit knowledge’ (Polanyi, 1958; Tsoukas, 2003), ‘local knowledge’ (Jørgensen Mjølberg, 2002; Sole & Edmondson, 2002), or ‘embodied knowledge’ (see for example, Fine, 1996). One may also distinguish between advanced, highly codified and propositional science-based knowledge (Cardinal, Alessandri & Turner, 2001) and more mundane forms of knowledge, often formulated in terms of ‘practice’ (Gherardi, 2001). In many cases, the knowledge of the practitioner, for example, a user of a specific technology, are brought into the organization in user-driven new product development activities (von Hippel, 1998). In such cases, the layman’s knowledge is influencing the knowledge base of the focal organization. In this study, rather than using the taxonomies of the knowledge management literature, the Greek knowledge taxonomy will be invoked in the analysis; based on descriptions and observations by Styhre (2003). Broadly speaking, there are three categories of knowledge in the Greek tradition of thinking that can be distinguished: episteme, techne and phronesis. Episteme is the term that sometimes is translated directly as knowledge, or even scientific knowledge, a translation that sometime does not do justice to the term itself. Epistemic knowledge is the idea of a general scientific knowledge, spatially and temporally independent and always valid. In a sense, it is this kind of knowledge that would be scientific knowledge, the capability of abstract thinking and analyses. In modern language episteme is found in the words ‘epis© 2006 The Authors Journal compilation © 2006 Blackwell Publishing


temology’ and ‘epistemic’. Techne, on the other hand, is a practical kind of knowledge to be compared to a craft or an art. A good example of this would be carpenters’ knowledge of working with wood, a knowledge that includes the bodily senses in a way that epistemic knowledge does not. In this aspect techne is a more embodied knowledge then episteme. In modern language techne can be found in the words ‘technology’ and ‘technical’. The last one of these three, phronesis is a kind of knowledge that can be compared to practical reasoning, or being ‘street smart’ (for a discussion, see Townley, 1999). This knowledge consists of acting from what one knows, to ‘make things happen’. In one sense it could be argued that this is a kind of entrepreneurial knowledge. But, compared to the first two kinds of knowledge, in modern language phronesis lacks its equivalent. Flyvbjerg argues ‘[i]t is indicative of the degree to which scientific and instrumental rationality dominate modern thinking and language that there is no modern word that similarly incorporates the classical word for the one intellectual virtue, phronesis, which Aristotle and other founders of the Western tradition saw as necessary condition of successful social organisation, and its most important prerequisite’ (2004, p. 285). Thus, while episteme and techne have acquired much attention and exploration, phronesis has not; the knowledge of being street smart has been marginalized and excluded from the mainstream research on knowledge. As Flyvbjerg (2004) has pointed out in his construction of a methodology based on this concept, phronesis might actually be more important category of knowledge than what we tend to think. In the following, a study of video game development work will be presented wherein the phronesis of the co-workers plays a central and decisive role. Therefore, rather than regarding phronesis as additional to other forms of knowledge, it is suggested that it plays a very central role in creativity and to date hase been neglected in the innovation and creativity literature.

Methodology Observations made in this study come from an ethnographic study of a game developer with approximately seven years’ experience in producing video games (Zackariasson, 2003). Empirical information was collected in line with the classic ethnographic studies of Mintzberg (1973) and Carlson (1964). These types of studies have gained a great deal of acceptance because of their assertions that © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

421

the construction of knowledge about a phenomenon should based on actual practice in everyday situations. Mintzberg based his observation study on what managers actually did in their daily practice. Carlson, whose study preceded the work of Mintzberg, based his studies on interviews and diaries of the managers. Practically, a combination of these previous approaches was utilized. That is, two separate weeks were spent on site in which the primary researcher ‘lived’ with the development group. During this time, field notes were made on what happened and what was said within the different working groups and group meetings (i.e. what, when, who). These notes were complemented by semi-formal interviews with both the CEO and personnel. The interviews with the CEO covered the topics of leadership, group culture and values, technology, strategy, power, ethics and morale, the product output and participant observations (eight interviews). Interviews with the personnel were focused on the four different leads (four interviews). Information from the rest of the personnel is to be found in the field notes (Zackariasson, 2003, p. 26) The observations made in this article are drawn from these transcribed observations and interviews.

Managing Creativity in Video Game Development The company studied is one of the major Swedish video game developers, with around 30 employees at the time of this study (today there are around 65 employees). The firm’s development followed a fairly classic model of success in the industry. Two entrepreneurs with different backgrounds, but common interests, got together after graduation from university and succeeded in raising the funds needed to start out in the game development business. The business orientation, as well as the quality of offering, led to a degree of success from its very beginning. Initially the company expanded, but as it was associated with a general set of IT business, its business receded with the crash of the ‘IT bubble’. Today the company has a more stable financial situation in that it has been acquired as a part of their former publisher, one of the major actors worldwide. Related to the topic of knowledge work in this article, the observations made in this company can broadly be categorized in four significant areas: (1) organization, (2) leadership, (3) personnel and (4) technology.

Volume 15

Number 4

2006

422


Art Team

Art Lead Programming Team

Programming Lead

Design Team Design Lead

Audio Lead

Internal Producer Product Manager

External Producer

Gamer – Buyer

Audio Team

Figure 1. Organization and Workflow

These areas will subsequently be presented below. 1

Organization The organization chart used by the company captured not only some sense of organization, but also the flow of work (see Figure 1). The firm is loosely built on a ‘project organization’, but the term ‘production team’ used within the company (insofar as they tend to work on a single project at a time) is equally applicable. As is common in most professional game production, four teams comprise the working organization, and these teams coincide with the tasks that must be accomplished in the production of any game – programming, art, audio and design. The arrows indicate the general flow of work from programming through design, but of course there is significant feedback among the teams during any given development cycle. The organization is focused on the product manager and centres on the leads of each of the four development areas. These leads are the individuals responsible for the output in their respective areas. 1

Because some confusion appears to have arisen at times in the reading of this article, it is important to note the process that is being described is the initial game development and the role that we believe phronesis plays in this process. It is not a development after a game has been released to which a user-developed paradigm might relate (cf. Franke & Shah, 2002; Franke & von Hippel, 2002; Shah, 2000; von Hippel, 2005). That type of development does occur. It is not uncommon for game producers to enable expert players, called ‘mods’ to be able to modify the game within limits. These individuals cannot open the source code or game engine, but they do have access to a programming interface that permits game modification. Naturally, game developers tend to monitor the activities of these individuals, but that aspect of development is not covered here.

Volume 15

Number 4

2006

The product manager has a role equivalent to that of a project manager, co-ordinating resources and keeping track of the project. The arrows leading to the development leads, which tend to make Figure 1 a little messy, represent this co-ordination and control. In this particular situation, the CEO played the role of project manager. There are two producers in the organization – one internal and one external. The internal producer, also the CEO, is responsible for the biweekly ‘builds’ that capture up-to-date progress, whereas the external producer, the publisher, produces the commercial version and interacts (markets the product) with consumers. In this case, the external producer owns the company, which of course has implications for the financing and selection of game projects. At the time of the field visits, there were 20 to 25 members in this production team, and the game that they were working on has recently been released with a reasonable share of acclaim in the press and praise from the gamers. The CEO and leads were of the opinion that traditional project models, such as a milestone model, are still widely applied when developing video games. Many times this use is a result of the contractual relationship between producer and publisher with respect to deliverables and payments. Although these models are used as generic models, early on they were found not to fit well to the actual development of video games. Despite broad applications, the main criterion for the success of video games lies in the very subjective concepts of ‘fun’ and immersion. The main argument that the CEO in particular had against milestone models was that it was not possible to define at project onset what a fun and immersive game was. In order to develop these kinds of video games, the CEO as a part of work organization developed a project model of its own. The driving philosophy behind this move was © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


423

that such a model would enable the company as a whole to drive projects in a game developer culture. There would, of course, be a design document, an initial idea what the game would be about. But as production proceeded what was and what was not fun and immersive could only be assessed through interim results. The internal model that was developed worked around two-week, recurring cycles where a ‘build’ was developed at the end of each period, as shown in Figure 2 (for an more extensive description see Walfisz, Zackariasson & Wilson, 2006). We have called this approach a Lindblomian approach of ‘muddling through’ because it is reminiscent of the classic papers written by Lindblom (1959, 1979) that taught an approach to optimal decision-making in the absence of initial, quantifiable objectives. For the main part of the production cycle, the functional teams focus on production according to a specification. Each cycle starts with an evaluation day, which is Day 1 in Figure 2. Throughout this day the different teams meet to evaluate the work done during the preceding cycle and draw up guidelines for the present cycle. The aspects that are to be addressed during this day are: (1) ‘gameability’ (i.e. ‘how much fun it is’), (2) technical stability and (3) teamwork. Naturally, the first day starts with the game design document. This document is the result of the pre-production phase and focuses on the most basic game design and premises of the game. This document is also complemented with a time plan where the project deadline and budget are specified. Subsequently, the cycle ends with the construction of a build, and this function occurs on Day 10. This build is a playable game with

Day 10 Progress

Day 1

Continuous opportunities of redefinition Redefinitions

Day 5

Figure 2. Evolutionary Game Development © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

all the completed features implemented. The game tends to be rather crude at the beginning of the project, but as more features are added, the characteristics of the finished game are more easily evaluated. In this approach, not only can the build be evaluated in-house, but also by distributing this software to the publisher, it is easier to have fruitful communications with regard to expectations of end results. This build is the foundation for discussion on Day 1, and the cycle is completed. The outputs of the project are first alpha and beta versions. These versions are test versions, for game ability and ‘bugs’ (errors as a result of coding). The final version is the ‘gold’, the version later sold to costumers. In these projects, experience suggests it is not possibility to predict what changes will occur, only that there will be changes. Thus, during the cycles there are ample possibilities for redefinitions of approaches, both of technical and organizational nature. At the end of the first week of a cycle, Day 5, there is a scheduled meeting between the project manager and leads from the different disciplines. At this time the current cycle has proceeded about 50 percent and there is a need to check how work has progressed. There are also possibilities, or rather obligations, for the individuals working on the project to continuously self-evaluate progress and take action when the need occurs.

Leadership Writers on management from Drucker (1993) through Mintzberg (1973) to Collins (2001) have suggested leadership can be important, indeed may be the most important factor, in creating an outstanding firm. The CEO’s opinion on leadership tended to follow classical thinking (these thoughts have been presented in length in Zackariasson, Walfisz & Wilson, 2006). Fundamentally, he suggested that leadership is basically about the ability to lead a group of individuals toward a goal as effectively as possible. But he also strongly emphasized motivating individuals to gain control over events, to distribute leadership, was one of his functions. But knowing how to lead has to be accompanied with knowing what is produced. Literature on creativity suggests that creative leaders do not necessarily make for creative organizations (Rickards, 1994). Bassett-Jones (2005) has suggested that highcommitment organizations will tend to prefer an outcome-driven approach to managing people. The required levels of quality and output are seen to result from employee skills and knowledge, rather than high levels of

Volume 15

Number 4

2006

424


supervision. Suitable conditions to promote creativity and innovation are associated with management of work routines and the creation of appropriate teams. Discussions and meetings are vital in that aspect. Because all individuals in the company were gamers (except for those few in the administrative department), the discussions were made easier because all individuals had a common knowledge base concerning the final output. In the words of von Hippel, the information needed to innovate was ‘widely distributed’ (2005, p. 14). In practice, it thus becomes relevant not so much what a leader says, but what he/she does. Specifically, if one is ‘street smart’, his or her ability to take advantages of opportunities that avail themselves on the one hand, but to stay out of trouble on the other, should avail themselves. Part of this ability in this case is captured in the early history of the firm as mentioned previously. The business orientation of the company led to a degree of success from its very beginning. Initially the company expanded, but went through a period that was marked by more downs than ups. The low point occurred when the IT bubble burst, but the company crawled out of this hole by ‘right sizing’ and focusing of its core business, which it still pursues. Four years ago financial stability was added, when it was acquired by its major customer. Subsequent to that, the CEO also got a position in the parent company and now holds two positions within the expanded organization. Company employment had at this time grown to about 30, which is fair-sized for a game developer. With regard to activities, if one were to say what has been done to make this company successful, five factors can be identified. First, of course, financial stability has been added, which protects against downside risk. Second, the company started to focus its effort, which Porter (1996) would say was essential to success – if a company is not going to be the low-cost producer, then it should be focusing on one specific niche. Those two items relate to any start-up that aspires to success. Consequently, then come three decisions that seem novel and can be associated with this particular company and its leader. It went to an evolutionary game development approach, briefly described above and shown in Figure 2. This approach essentially associated the company with the ‘outcome-driven’ and ‘work routine’ focus that Bassett-Jones (2005) indicates is required by creative, high-commitment organizations. Fourth, the CEO introduced the practice of hiring gamers into the organization, which supplied the company with the knowledge

Volume 15

Number 4

2006

and skill required to meet objectives (BassettJones, 2005, p. 172), or as Collins has written, he had the ‘right people on the bus’ (2001, pp. 41ff.). Part of the knowledge and skill these people possessed, we would assert, was a high degree of phronesis. Lastly, the CEO had the good sense to form teams around functions as shown in Figure 1 and turn these teams loose to complete their tasks – with encouragement for individual contributions (shown by the small dashed circles in Figure 2). The phrase ‘loose-tight’ has been used to describe such approaches and organizations (Lawrence & Lorsch, 1967, p. 161; Peters & Waterman, 1982, pp. 318– 325) – loose, very loose, with respect to daily activities, but tight with regard to complying with the ten-day cycles.

Human Resources The personnel in this production studio fit into the characteristics of the business, which is also in line with the hackers culture (Ullman, 1997) and even the coders at the Atari research lab (Stone, 1995), or the young men working in Data General constructing computers (Kidder, 1981). Generally, they consisted of men in their late 20s and early 30s. The only two women in the group are working with administrative and marketing/sales chores. The environment around these men gives the impression of a ‘teen dream for gamers’: state-of-the-art computers, odd posters, strange decorations, Coke vending machine and informal dressing. When one of the authors arrived at this company he was introduced by the CEO, with the words, ‘he’s not dangerous. He hasn’t got any tie’. Nearly everyone in this company is a gamer. That is, to work here one has to play video games. Talking to the CEO he explains that it is a requisite that the one producing the games also has to have a great deal of experience playing the same sort of games. This background will give an understanding to the person that no formal education can provide; choosing between a technically very skilled person and a devoted gamer, the CEO claims that he would in most cases choose the latter one. Because, as he explains, ‘it is easier to train a gamer to produce graphics for games, then it is to train a skilled graphical artist to do good game graphics’. That everyone in the company are gamers is highly visible, both through talk and through the constant sound of gunfire or crashes from games somewhere in the office. Playing games seem to be part of work, both playing games developed within the firm and those developed by other studios. It may seem as if we have focused on the knowledge and contributions the CEO has © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


made. That tendency is natural because of the stage of creative growth that the company happens to be in at this time. Nevertheless, each member of the staff had the opportunity to contribute to the firm’s success. We would like to cite two examples where members were expected to use their knowledge in the process of advancing game development and also development of the firm. First, we have already alluded to the fact that the individuals working on the project were encouraged to continuously self-evaluate progress and take action when the need occurs in the description of Figure 2. Those are the small dashed circles that are described as ‘continuous opportunities of redefinition’. In these redefinitions, individuals were expected to not only use their knowledge of games, but their general knowledge as well – as would be expected from a phronesis consideration. An exceptional gathering of group knowledge occurred in the ‘Day 1’ regime. On that day, the focus is on evaluating what has been done after playing the working build over the weekend. In a general meeting, individuals are encouraged to comment on the game’s features, both positive and negative. Anyone is allowed to present his or her opinion on anything. Two things, for instance, were changed during the periods of observation. In one case, the artificial intelligence was changed so that the gamer had a chance to win. In another, random access was provided so that all troops were not loading and shooting at the same time. Subsequent to this first-day preliminary review, the leads set up an agenda for the next cycle and then the teams meet again to decide how they will proceed. One of the more interesting aspects of the observation was how teams working in parallel, but depending on series progress, were able to bring work together. For example, game speed, coding, landing zones, weapons, armour, woods and avatar appearance (Zackariasson, 2003, pp. 46ff). In observing the process, there appeared to be no centre, but instead was more individually focused. The second example of group contribution was associated with hiring practice (and this example was also part of the observation portion of the study). It has already been said that the firm employed gamers. In practice, some basic capabilities such as programming and play were used as screens. They were ‘necessary’ conditions for employment. Subsequently, candidates would have some dinner and/or drinks with individuals who would be working with the candidate. If those individuals thought that the candidate would fit, then hiring was probably – that was the condition of ‘sufficiency’. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

425

Technology Being a video game developer means working very closely with information technology, both as a medium for production and as platform for the product. Computers are exclusively used for producing games, a product that in the end will consist of different voltages and a digital duality of ones and zeroes. In this aspect, video game developers were in the late 1990s placed in the same category as the then blossoming IT-business. At that time it was a good thing, but a few years later many of them suffered under the same crash as the IT businesses. In this company there was, in addition to all computers, an intranet that is of special interest in this article. This intranet, which can only be accessed internally, consisted of three different sections: standards, projects and issues/ tasks. The first section, standards, consists of a number of different documents that specify how things should be done when producing games. For instance, there are documents specifying how code should look. From a first glance, this is a topic that might not look that complicated, but as Piñeiro (2003) has stated, is an aesthetic area of great importance to coders. These documents are not fixed templates to be used, but documents constantly undergoing revisions. This is mainly done by the lead-person of the coders. This section does also contain much description of the project model, Evolutionary Game Development. The second section in the intranet is the project site. This site contains all documents related to ongoing and completed projects. The major difference from the previous section would be the locked status on most documents. These documents are not under construction, or a topic for refinement, but they are final specifications for the project. The last section is the most interactive of them all, the issue/task section. Here different things that must be accomplished in the project are posted as issues. When these issues are assigned to specific persons they become tasks and get a specific deadline for completion. Actually, this section works in somewhat the same way as a note board. They are categorized, both after the priority as after the importance, in terms of fun and immersion. The issues that take the least time to implement and yield the most fun and immersion are priorities.

Discussion On basis of the account of the work practices of this Swedish video game developer knowledge-based and creative work can be

Volume 15

Number 4

2006

426


described from the perspective of phronesis. Following Alvesson’s (2001) idea of knowledge as to some extent being ‘slippery’ and ambiguous, knowledge is made visible and manifested in actions taken. In the company, knowledge was employed in interactions between co-workers. This practice is pronounced in the case of the CEO, who uses the whole office as a ground for interaction, both with the personnel and with the product itself. New computer development work unfolds as a process of continuous negotiation and testing of what works and what does not. In fact the whole organization is designed to nurture interaction, from individuals working at the computer to the lead, and from leads to product manager and so on. After all, knowledge of what constitutes a fun and immersive game is constructed in the creative interaction between key individuals. Gherardi argues: ‘Knowledge resides in social relations, and knowing is part of becoming an insider in a community of practice’ (2001, p. 133). The project model Evolutionary Game Development (Figure 2 above) is a good example of how knowledge construction takes place in the interaction between the individuals in the organization. The constant assessments of the emerging product enable interactions and negotiations of what actually constitutes what is fun and immersive in the game. The interaction and negotiation between key actors are embedded in the phronesis of the actors, that is, their ability to evaluate what is constitutive of a fun game. Since there is no possibility of defining the outcome from the new video-game development process when starting the project, the development model needs to be based on flexible interaction between the actors. There are possibilities of defining visions and goals in general terms, but not a complete specification of the end product. What is in-between the rather crude first specifications and the finalized product is the outcome from the management of knowhow and phronesis – not only street knowledge of games, but general knowledge as well. It has been noted that a person cannot be creative in the abstract, however, but only within the rules of some practice of the idea system (Csikszentmihalyi, 2000). What the company accomplishes when only employing dedicated gamers is the construction of an environment, a practice and a community wherein the individuals constructing the game basically ‘know what works’. It should be noted that these individuals are not ‘users’ as von Hippel defines them. To him, users are ‘firms or individuals that expect to benefit from using a product or service’ (von Hippel, 2005, p. 3). Nevertheless, because of their

Volume 15

Number 4

2006

extensive experience from playing computer games these individuals have the know-how, that is, the phronesis, of what graphic designs are appealing or exciting, what sounds, visual effects, stories, which are constitutive of a fun and immersive video game. Put another way, video games do not need to have the best available graphics or the best sound to be fun and immersive; instead it is the combination of these parts into a unified structure that produces success in video games. It would be an oversight, however, to suggest that this knowledge reduces the importance of technical knowledge, that is, the techne, of how to actually produce the game. What the company has done has been to ‘level the playing field’ (von Hippel, 2005, p. 14) and brought the technology and capability together to produce the essentials of innovation. Concerns exist not only in what individuals know, but how the individuals are grouped. During discussions of this manuscript, attention was called to the recent article in this journal by Bassett-Jones (2005) that considered creativity and innovation as affected by diversity. Specifically, it was wondered if a very homogeneous design group (young, white males) might in effect limit creativity. We, of course, have limited information with which to generalize observations with respect to human resource management as Bassett-Jones (2005) had done – this study was conducted on a single firm during its stage of creative growth (cf. Greiner, 1972). Nevertheless, diversity goes beyond demographics, as BassettJones clearly indicates (2005, pp. 169–170). It includes function, lifestyle and intellectual capability, which clearly were observed among employees in the group. Second, Bassett-Jones (2005, p. 171) indicate involvement counts for a lot in building high-performance organizations. Involvement is something difficult to quantify, but there appeared to be little doubt in observation that group members were involved in this work – whooping and hollering were observed as individuals played their games and worked past 7:00 p.m. was common. Furthermore, there appears to be no clear indication that diversity does singularly contribute to high-commitment organizations; group-think tendencies of homogeneous groups are balanced against possibilities of conflict that damages cohesiveness (BassettJones, 2005, p. 172). Thus, we are left with the observation in this case that the hiring of gamers worked in the development of these games. Before leaving this topic, it may be worthwhile suggesting that gamers may be interesting studies with regard to creative © 2006 The Authors Journal compilation © 2006 Blackwell Publishing


capabilities. Correlation studies show that playful individuals tend to score higher on tests of creativity and tend to be judged more creative by others (Dansky, 1994). These individuals were nothing if not playful. The classic story concerning gamers is the one told of Steve Wozniak, co-developer of Apple Computers (Kent, 2001, pp. 71, 361), who was an avid and accomplished gamer. Perhaps a period of advanced creativity awaits us as these individuals saturate the workforce. A friendly suggestion was also made during review that the process described in this article could be described by Akrich’s (1992, 1995) methodology, specifically her I-methodology. In that approach it is stated that all designers of new products anticipate and represent the envisioned user. In effect, the constructers of these games are users. It is tempting to accept this suggestion because elements are present in any development process that meets anticipated wants or needs. Certainly one could not deny that developers had no idea of final design and envisioned users. We think, however, that a more appropriate model for describing the process comes from Simon. In this model he suggests, ‘Making complex designs that are implemented over a long period of time and continually modified in the course of implementation has much in common with painting in oil. In oil painting every new spot of pigment laid on the canvas creates some kind of pattern that provides a continuing source of new ideas to the painter’ (1996, pp. 162–163). Video game development as observed in this firm appeared to follow this type approach for two reasons. First, there was the fuzzy front end. The game was sketched only roughly as an initial step. Developments were centred upon features that the games would provide and not some final design. Second, both technology and expectations of consumers were likely to change over the twoyear period that it takes to develop a game. Consequently, progress was driven by the tenday evaluations shown in Figure 2, where the interim build became Simon’s pigment on the canvas to which the development team reacted. When looking at the key components of the model – organization, leadership, human resources and technology – it is clear that actors are involved in all of these parts. This involvement implies that the final video game is the outcome of the work of actors and the interactions among them. But the activity taking place, the interactions, are for most part about how to make things work, as a practical matter. These developers are pragmatics. They learn to code or to construct graphics in order © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

427

to make a game that is fun and immersive. So while techne builds the very base for the interaction, the interaction concerns primarily knowledge qua phronesis.

Conclusion This article has presented a study of how video game development is structured around technical skills and knowledge. We refer to this knowledge as phronesis, that is, the insight of what features of video games that members of the community of games appreciate and find intriguing. It is inferred that innovation and creativity in the video game industry is therefore determined by at least two different bodies of know-how: the technical skills of programming and designing the game and the phronesis skills providing the more subtle insights into the actual game-playing experience. In order to comprehend the full potential of a video game, one needs, industry representatives argue, to have to demonstrate a longterm commitment to the field. The study thus points at a number of implications for innovation and creativity work, for instance the emphasis on the type of know-how embodied by the notion of phronesis that is not easily represented and codified. In addition, the videogame development model suggested captures the open-ended nature of video game development, enabling a substantial degree of continuous adjustment the comments and suggestions provided by the gamers.

References Akrich, M. (1992) The De-scription of Technical Objects. In Bijker, W.E. and Law, J. (eds.), Shaping Technology/Building Society: Studies in Sociotechnical Change. MIT Press, Cambridge, MA, pp. 205– 24. Akrich, M. (1995) User Representations: Practices, Methods and Sociology. In Rip, A., Misa, T.J. and Schot, J. (eds.), Managing Technology in Society. The Approach of Constructive Technology Assessment. Pinter Publishers, London, pp. 167–84. Alvesson, M. (2001) Knowledge Work: Ambiguity, Image and Identity. Human Relations, 54, 863–86. Bassett-Jones, N. (2005) The Paradox of Diversity Management, Creativity and Innovation. Creativity and Innovation Management, 14, 169–75. Carlson, S. (1964) Företagsledare i arbete. Prisma, Stockholm. Cardinal, L.B., Alessandri, T.M. and Turner, S.F. (2001) Knowledge Codifability, Resources and Science-Based Innovation. Journal of Knowledge Management, 5, 195–204. Collins, J. (2001) Good to Great. Harper Business, New York.

Volume 15

Number 4

2006

428


Csikszentmihalyi, M. (2000) Creativity: An Overview. In Kazdin, A.E. (ed.), Encyclopedia of Psychology, vol 2. Oxford University Press, Oxford, pp. 337–42. Dansky, J.L. (1994) Play. In Runco, M.A. and Pritzker, S.R. (eds.) Encyclopedia of Creativity. Academic Press, San Diego, pp. 393–408. Drucker, P. (1993) The Practice of Management, Harper Business, New York. Fine, G.A. (1996) Kitchens: The Culture of Restaurant Work. University of California Press, Berkeley. Flyvbjerg, B. (2004) Phronetic Planning Research: Theoretical and Methodological Reflections. Planning Theory & Practice, 5, 283–306. Franke, N. and Shah, S. (2002) How Communities Support Innovative Studies: An Exploration of Assistance and Sharing Among End-Users. Sloan Working Paper 4i64, Massachusetts Institute of Technology, Sloan School of Management, Cambridge, MA. Franke, N. and von Hippel, E. (2002) Satisfying Heterogeneous Needs Via Innovation Toolkits: The Case of Apache Security Software. Sloan Working Paper 4341-02, Massachusetts Institute of Technology, Sloan School of Management, Cambridge, MA. Gherardi, S. (2001) From Organizational Learning to Practice-Based Knowing. Human Relations, 54, 131–9. Greiner, L. (1972) Evolutions and Revolutions as Organizations Grow. Harvard Business Review, 50, 37–46. Jørgensen Mjølberg, K. (2002) The Meaning of Local Knowledges: Genealogy and Organizational Analysis. Scandinavian Journal of Management, 18, 29–46. Kent, S.L. (2001) The Ultimate History of Video Games: the History Behind the Craze That Touched Our Lives and Changed the World. Three Rivers Press, New York. Kidder, T. (1981) The Soul of a New Machine. Little, Brown & Company, New York. Lawrence, P. and Lorsch, J. (1967) Organization and Environment. Harvard University Press, Boston, MA. Lindblom, C. (1959) The Science of ‘Muddling Through’. Public Administration Review, 19, 79–88. Lindblom, C. (1979) Still Muddling, Not Yet Through. Public Administration Review, November–December, 517–26. Marshall, A. (1972) Principles of Economics. Macmillan, London. Mintzberg, H. (1973) The Nature of Managerial Work. Prentice Hall, Englewood-Cliffs. Peters, T.J. and Waterman, R.H., Jr. (1982) In Search of Excellence: Lessons from America’s Best-run Companies. Harper & Row Publishers, New York. Piñeiro, E. (2003) The Aesthetics of Code: On Excellence in Instrumental Action. Royal Institute of Technology, Stockholm. Polanyi, M. (1958) Personal Knowledge: Toward a Post-Critical Philosophy. Chicago University Press, Chicago.

Volume 15

Number 4

2006

Porter, M. (1996) What is Strategy? Harvard Business Review, November–December, 61–78. Rehn, A. (2004) The Serious Unreal: Notes on Business and Frivolity. Dvalin Books. Rickards, T. (1994) Organizations Interested in Creativity. In Runco, M.A. and Pritzker, S.R. (eds.), Encyclopedia of Creativity. Academic Press, San Diego, pp. 319–435. Shah, S. (2000) Sources and Patterns of Innovation in a Consumer Products Field: Innovations in Sporting Equipment. Sloan Working Paper 4105, Massachusetts Institute of Technology, Sloan School of Management, Cambridge, MA. Simon, H. (1996) The Sciences of the Artificial, 3rd edn. MIT Press, Cambridge, MA. Sole, D. and Edmondson, A. (2002) Situated Knowledge and Learning in Disperse Teams. British Journal of Management, 13, 17–34. Stewart, T.A. (1997) Intellectual Capital: The New Wealth of Organizations. Nicholas Brealey, London. Stone, A.R. (1995) The War of Desire and Technology at the Close of the Mechanical Age. MIT Press, Cambridge, MA. Styhre, A. (2003) Understanding Knowledge Management: Critical and Postmodern Perspectives. Liber, Malmö. Tell, F. (2004) What do Organizations Know? Dynamics of Justification Context in R&D Activities. Organization, 11, 443–71. Townley, B. (1999) Practical Reason and Performance Appraisal. Journal of Management Studies, 36, 287–306. Tsoukas, H. (2003) Do We Really Understand Tacit Knowledge? In Easterby-Smith, M. and Lyles, M.A. (eds.), Handbook of Organization Learning and Knowledge Management. Blackwell, Oxford, pp. 410–27. Ullman, E. (1997) Close to the Machine. City Lights Books, San Francisco, CA. Von Hippel, E. (1998) Economics of Product Development by Users: the Impact of ‘Sticky’ Local Information. Management Science, 44, 629– 44. Von Hippel, E. (2005) Democratizing Innovation. MIT Press, Cambridge, MA. Walfisz, M., Zackariasson, P. and Wilson, T.L. (2006) Real-Time Strategy: Evolutionary Game Development. Business Horizons, in press. Yanow, D. (2004) Translating Local Knowledge at Organizational Peripheries. British Journal of Management, 15, 9–25. Zackariasson, P. (2003) Cyborg Leadership: Including Nonhuman Actors in Leadership. Licentiate Thesis, Umeå School of Business, Umeå University. Zackariasson, P., Walfisz, M. and Wilson, T.L. (2006) Management of Creativity in Video Game Development: A Case Study. Services Marketing Quarterly, 27, 73–97.



429

Peter Zackariasson (peter.zackariasson@ usbe.umu.se) is a Doctoral student at Umeå Business School, Umeå, Sweden. His research covers the production and consumption of video games. He also has a wider interest in organizing/ordering, technology and aesthetics. His doctoral thesis is expected to be published at the end of 2006. Alexander Styhre (alexander.styhre@ fenix.chalmers.se) is Professor at the Department of Technology Management, Chalmers University of Technology, Gothenburg, Sweden. He is interested in knowledge management, organization learning and organization creativity. His most recent books are Managing Creativity in Organizations (co-authored with Mats Sundgren, 2005, Palgrave) and Management Writing Out of Bounds (2005, Copenhagen Business School Press). Timothy L. Wilson (tim.wilson@usbe. umu.se) is Adjunct Professor of marketing and management at Umeå Business School, Umeå, Sweden. He has an ongoing interest in applied business topics and has co-operated with Zachariasson in video game developments in a number of venues.


Volume 15

Number 4

2006

430


Book Reviews Avlonitis, G. and Papastathopoulou, P. (2006) Product and services management, Sage Publications, London, UK. Cloth (1-4129-0864-7) £75.00 / Paper (1-4129-0865-5) 280 pages

In defining, describing or positioning a product or service or indeed a book, sometimes it is easier to first define, that which it is not. What this book is not is another review of the differences and similarities of tangible goods and intangible services in the academic literature. The clue is in the title; this book describes, to borrow a phrase from a wood preserve advertisement, ‘exactly what it says on the’ cover. The focus that Avlonitis and Papastathopoulou adopt is one of detailed review of the stages and processes of managing product; whether tangible products (goods) or intangible products (services). The result is a text that successfully bridges the gap between academic theorizing and practitioner applicability because it uses multiple real-world examples/ mini-cases of management techniques to illustrate the well-researched academic theoretical foundations of the book. The book could be used as a how-to guide for product managers, guiding them from an understanding of the nature of the product through development and evaluation of new products. At the same time it offers the researcher in product innovation a step-by-step guide through key stages of new product development as outlined in key references and through the illustration of theory with examples it offers the student in product management and innovation a clear learning tool. Where Avlonitis and Papastathopoulou provide added value is in chapters 9 and 10 where, having dealt with the standard topics of product development and innovation, they continue to review the problems of identifying weak goods and services and the management of product revitalization or elimination procedures, once again combining academic theory with case illustrations. This is a critically important area that is often insufficiently covered in new product development texts and Avlonitis and Papastathopoulou cover it in depth, including a useful flow diagram for the sequence of product elimination process decisions. At a time when it seems that the world and his dog are focused on managing the customer and developing new and improved services to

Volume 15

Number 4

doi:10.1111/j.1467-8691.2006.00403.x

2006

develop relationships with him or her, it is refreshing to review a text that attempts to discuss the importance of combining management and innovation of both tangible products and intangible services in equal measure. The book opens with a semi-introductory chapter dealing with the nature of the product as economic entity and briefly introducing key product concepts in the literature; differential competitive advantage, product levels, hierarchy, life-cycle, positioning, innovation and classification. This introductory chapter is followed by a review of the types of product decisions faced by managers including good and service type; the tangible/physical functional elements and the intangible product elements incorporating branding. Chapter 3 discusses marketing strategies at different stages of the product life-cycle, illustrated with a number of mini case-studies outlining business behaviour at different stages of life cycles – for example, the mobile telephony market in Greece. Chapter 4 is more theoretical in focus, outlining various models of product/service portfolio evaluation. Chapter 5 is also broadly theoretical, dealing with the development and performance management of new products and portfolios of products, combining the models of Kotler, and of Booz, Allen and Hamilton (p. 101). Chapters 6 to 8 then take a more detailed look at new product and service development, utilizing examples to illustrate the theory outlined; pre-development activities (Ch. 6), development, testing and launching (Ch. 7) and success in adoption and diffusion of new products (Ch. 8). In the latter chapter the example of Starbuck’s Frappuccino highlights how successful product development and diffusion can result from understanding and responding to consumer needs, that luck plays a part and that senior management buy-in is not always vital. Following chapters 9 and 10 on managing weaker goods and services, chapter 11 concludes the book with an organizational management perspective on managing product development or elimination functions. Budgeting issues are covered in a brief appendix. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

BOOK REVIEWS

In summary the book provides a holistic overview of the key stages of both good and service management, right through to product disposal, offering a synthesis of academic writing, practical advice and illustrative examples of experience in various industries. It offers value to the inquisitive manager, highlighting key procedures and explaining why various

431

practices are conducted, and at the same time would provide students or new researchers in new product development with a clear understanding of key concepts and literature. Jamie Burton

Manchester Business School

Blackwell Publishing Ltd.Oxford, UK and Malden, USACAIMCreativity and Innovation Management0963-1690Blackwell Publishing Ltd, 2006.2006154432433BOOK REVIEWSBOOK REVIEWSCREATIVITY AND INNOVATION MANAGEMENT

Karlsson, Charlie, Flensburg, Per and Hörte, Sven-Ake (2004) Knowledge Spillovers and Knowledge Management, Edward Elgar, Cheltenham, UK, Northampton, MA, USA. (1-84376785-6) $145/£141.50, 510 pages

The book is a 500+ page edited volume of papers taken from the workshop on Knowledge Spillovers and Knowledge Management held from 19 to 21 September 2002 at the Centre for Information Logistics in Ljungby, Sweden. It consists of 17 chapters by participants of the workshop, with an introduction by the editors. The editors have divided the book into three parts, entitled ‘Knowledge Spillovers’, ‘Regional Innovation Systems’ and ‘Knowledge Management’. Since it is a collection of individual articles, it is hard to summarize the contents of the book in a reasonable number of lines. Instead, I will try to give a flavour of the book by highlighting the variety of its contents, starting with the 17 chapters of the contributing authors, followed by the contributions of the editors. The chapters of the book represent research on a very diverse set of topics, using a variety of methods, applied to a wide range of industries. Some of the topics addressed in this volume are in geographical proximity (Caniëls & Romijn, Chapter 2), R&D co-operation (Hörte, Chapter 3), accessibility (Andersson & Karlsson, Chapter 10), the role of labour in regional innovation systems (Hommen & Doloreux, Chapter 11), the triple helix (Ylinenpää, Chapter 12), regional clusters (Steiner, Chapter 13; Asheim, Chapter 14), communities of practice (Haynes, Kulkarni, & Stough, Chapter 15), sense-making (Ifvarsson & Häckner, Chapter 16) and content management (Flensburg, Chapter 17). In terms of industries that are studied, the chapters range from the pharmaceutical industry (Cooke, Chapter 1) and manufacturing (Andersson & Ejermo, Chapter 5), to arts (Norton, Chapter 6) and fish farming and aquaculture (Orstavik, Chapter 7). In addition, as they try to develop general mod© 2006 The Authors Journal compilation © 2006 Blackwell Publishing

els, some chapters do not focus on a particular industry (for example, Olsson, Chapter 9). Concerning methods, the book contains chapters using formal modelling and production functions (Greunz, Chapter 4), patent analysis (Andersson & Ejermo, Chapter 5), historical research (Norton, Chapter 6), surveys (Koschatzky, Chapter 8) and simulation (Haynes, Kulkarni, & Stough, Chapter 15). From this sample of topics, industries and methods can be induced, so the scope of the book is wide. Whether this should be seen as an advantage or a disadvantage I leave up to the individual reader. The editors have tried to organize the eclectic set of articles into three parts: ‘Knowledge Spillovers’, ‘Regional Innovation Systems’ and ‘Knowledge Management’. The reasons for leaving the title of the second part out of the title of the book are unclear to me, particularly because I think it best covers the contents of the book as a whole. In addition, I was a bit confused as to how the 17 chapters have been divided over the three parts. For example, why is a chapter on ‘regional dynamism’ (Chapter 2) positioned in Part I and not in Part II? Furthermore, why is a chapter on ‘clusters’ (Chapter 14) positioned in Part III and not in Part II? Unfortunately, the editors do not explain this in their introduction. The introductory chapter of the book provides the reader with a discussion on knowledge, a description of its role in the economy, a summary of the three main topics of the book (knowledge spillovers, regional innovation systems and knowledge management) and a summary of all the chapters. The editors do a good job of summarizing the chapters. However, it would have been very beneficial for the book if they would have put more effort in integrating the contents of the book into a real

Volume 15

Number 4

2006

BOOK REVIEWS

In summary the book provides a holistic overview of the key stages of both good and service management, right through to product disposal, offering a synthesis of academic writing, practical advice and illustrative examples of experience in various industries. It offers value to the inquisitive manager, highlighting key procedures and explaining why various

431

practices are conducted, and at the same time would provide students or new researchers in new product development with a clear understanding of key concepts and literature. Jamie Burton

Manchester Business School

Blackwell Publishing Ltd.Oxford, UK and Malden, USACAIMCreativity and Innovation Management0963-1690Blackwell Publishing Ltd, 2006.2006154432433BOOK REVIEWSBOOK REVIEWSCREATIVITY AND INNOVATION MANAGEMENT

Karlsson, Charlie, Flensburg, Per and Hörte, Sven-Ake (2004) Knowledge Spillovers and Knowledge Management, Edward Elgar, Cheltenham, UK, Northampton, MA, USA. (1-84376785-6) $145/£141.50, 510 pages

The book is a 500+ page edited volume of papers taken from the workshop on Knowledge Spillovers and Knowledge Management held from 19 to 21 September 2002 at the Centre for Information Logistics in Ljungby, Sweden. It consists of 17 chapters by participants of the workshop, with an introduction by the editors. The editors have divided the book into three parts, entitled ‘Knowledge Spillovers’, ‘Regional Innovation Systems’ and ‘Knowledge Management’. Since it is a collection of individual articles, it is hard to summarize the contents of the book in a reasonable number of lines. Instead, I will try to give a flavour of the book by highlighting the variety of its contents, starting with the 17 chapters of the contributing authors, followed by the contributions of the editors. The chapters of the book represent research on a very diverse set of topics, using a variety of methods, applied to a wide range of industries. Some of the topics addressed in this volume are in geographical proximity (Caniëls & Romijn, Chapter 2), R&D co-operation (Hörte, Chapter 3), accessibility (Andersson & Karlsson, Chapter 10), the role of labour in regional innovation systems (Hommen & Doloreux, Chapter 11), the triple helix (Ylinenpää, Chapter 12), regional clusters (Steiner, Chapter 13; Asheim, Chapter 14), communities of practice (Haynes, Kulkarni, & Stough, Chapter 15), sense-making (Ifvarsson & Häckner, Chapter 16) and content management (Flensburg, Chapter 17). In terms of industries that are studied, the chapters range from the pharmaceutical industry (Cooke, Chapter 1) and manufacturing (Andersson & Ejermo, Chapter 5), to arts (Norton, Chapter 6) and fish farming and aquaculture (Orstavik, Chapter 7). In addition, as they try to develop general mod© 2006 The Authors Journal compilation © 2006 Blackwell Publishing

els, some chapters do not focus on a particular industry (for example, Olsson, Chapter 9). Concerning methods, the book contains chapters using formal modelling and production functions (Greunz, Chapter 4), patent analysis (Andersson & Ejermo, Chapter 5), historical research (Norton, Chapter 6), surveys (Koschatzky, Chapter 8) and simulation (Haynes, Kulkarni, & Stough, Chapter 15). From this sample of topics, industries and methods can be induced, so the scope of the book is wide. Whether this should be seen as an advantage or a disadvantage I leave up to the individual reader. The editors have tried to organize the eclectic set of articles into three parts: ‘Knowledge Spillovers’, ‘Regional Innovation Systems’ and ‘Knowledge Management’. The reasons for leaving the title of the second part out of the title of the book are unclear to me, particularly because I think it best covers the contents of the book as a whole. In addition, I was a bit confused as to how the 17 chapters have been divided over the three parts. For example, why is a chapter on ‘regional dynamism’ (Chapter 2) positioned in Part I and not in Part II? Furthermore, why is a chapter on ‘clusters’ (Chapter 14) positioned in Part III and not in Part II? Unfortunately, the editors do not explain this in their introduction. The introductory chapter of the book provides the reader with a discussion on knowledge, a description of its role in the economy, a summary of the three main topics of the book (knowledge spillovers, regional innovation systems and knowledge management) and a summary of all the chapters. The editors do a good job of summarizing the chapters. However, it would have been very beneficial for the book if they would have put more effort in integrating the contents of the book into a real

Volume 15

Number 4

2006

432


edited volume. They do not do this in the introduction, and there is no concluding chapter either. Given the wide variety of the articles, integration would probably have been a very difficult task. Nevertheless, I feel the editors have missed a great opportunity to improve the quality of the book. I would like to highlight three further issues. First, the book contains multiple attempts to distinguish knowledge from information and data (for examples, Introduction, p. 4; Chapter 6, p. 173, and Chapter 17, p. 480) and to explain the differences between tacit and explicit knowledge (for examples, Introduction, p. 5; Chapter 10, p. 286; Chapter 11, p. 319; and Chapter 16, p. 447). In this respect, the book is illustrative for many other publications in the area of ‘knowledge management’. What perhaps is most interesting about this is that all explanations of knowledge in the book are different. For the rest, however, I always wonder why scholars put so much emphasis on these particular differences. In most cases the distinction has no relevance for the particular study that is conducted. The chapters of this book form no exception here. In addition, there are numerous other ways to classify types of knowledge that are often much more relevant. Examples are classifications of internal/external, strategic/operational and atomic/systemic knowledge. The time has come to move on beyond exploiting the DIK paradigm and beyond quoting Nonaka and Polanyi. Second, the book contains a lot of writing and suggesting, but little theorizing and empirical research. As it says on the back cover, the book ‘highlights a number of issues at the leading edge of both research and policy making, such as knowledge generation/ production, knowledge distribution/transfer,

Volume 15

Number 4

2006

knowledge spillovers, learning, knowledge management, information logistics, industrial clusters, industrial networks and regional innovation systems’. That is indeed what it does. It highlights a variety of related topics. Given that the topics of this book have been studied for decades by now, I think both the authors and the editors could have gone beyond that. In this respect, Ifvarsson and Häckner (Chapter 16, ‘Strategic Knowledge Management: Sensemaking for Control of Spillover Effects in Interorganizational Networks’) make a step in the right direction by conceptualizing a sense-making framework. Lastly, there is the peculiar relationship between the book and the concept of information logistics. Throughout the book there are references to this concept: the workshop behind this book was sponsored by the Centre for Information Logistics in Ljungby, the topic is explained by the editors in the introduction, and there is one chapter on information logistics (Flensburg, Chapter 17). While the editors suggest that the book ‘can be seen as a macro-level introduction to some mechanisms enabling information logistics’ (p. 18), I think that the contents of the book have no significant relation with the concept. Who should read the book? Given the broad range of topics, industries and topics, and the lack of thorough theorizing and research, the book is best suited for researchers and practitioners that are new to the field – it provides quick overview of contemporary work on knowledge spillovers and regional innovation systems. Jeroen Kraaijenbrink

Dutch Institute for Knowledge Intensive Entrepreneurship (NIKOS), University of Twente


INDEX TO VOLUME 15

433

Index to Volume 15 (e.g. 1/74 = number 1, page 74)

ARTICLES ABETTI, Pier A. Case Study: Jack Welch’s Creative Revolutionary Transformation of General Electric and the Thermidorean Reaction (1981–2004) 1/74 ALLOCCA, Michael A., and KESSLER, Eric H. Innovation Speed in Small and Medium-Sized Enterprises 3/279 BAKKER, Hans, BOERSMA, Kees and OREEL, Styse. Creativity (Ideas) Management in Industrial R&D Organizations: A Crea-Political Process Model and an Empirical Illustration of Corus RD&T 3/296 BASADUR, Min and GELADE, Garry A. The Role of Knowledge Management in the Innovation Process 1/45 CHEN, Ming-Huei. Understanding the Benefits and Detriments of Conflict on Team Creativity Process 1/105 CHUN, Rosa. Innovation and Reputation: An Ethical Character Perspective 1/63 DAVENPORT, John. UK Film Companies: Project-Based Organizations Lacking Entrepreneurship and Innovativeness? 3/250 DEMPSTER, Anna M. Managing Uncertainty in Creative Industries: Lessons from Jerry Springer the Opera

3/224

EIKHOF, Doris Ruth and HAUNSCHILD, Axel. Lifestyle Meets Market: Bohemian Entrepreneurs in Creative Industries 3/234 HÄNNINEN, Seppo, and KAURANEN, Ilkka. A Multidimensional Product-Concept Model Enhancing Cross-Functional Knowledge Creation in the Product Innovation Process: The Case of the Suunto t6 Training Wrist Computer 4/400 HEERKENS, Hans. Assessing the Importance of Factors Determining Decision-Making by Actors Involved in Innovation Processes 4/385 HYLAND, Paul W., MARCEAU, Jane and SLOAN, Terry R. Sources of Innovation and Ideas in ICT Firms in Australia 2/182 © 2006 The Authors Journal compilation © 2006 Blackwell Publishing

IBRAHIM, Sherwat E., FALLAH, M. Hosein and REILLY, Richard R. Do Localized Clusters Influence Creativity of Inventors? 4/410 JEANES, Emma L. ‘Resisting Creativity, Creating the New’ A Deleuzian Perspective on Creativity 2/127 JØRGENSEN, Frances, BOER, Harry and LAUGEN, Bjørge Timenes. CI Implementation: An Empirical Test of the CI Maturity Model 4/328 KALTOFT, Rasmus, BOER, Harry, CHAPMAN, Ross, GERTSEN, Frank and NIELSEN, Jacob S. Collaborative Improvement – Interplay but not a Game

4/348

KRATZER, Jan, LEENDERS, Roger TH.A.J. and VAN ENGELEN, Jo M.L. Team Polarity and Creative Performance in Innovation Teams 1/96 LASSEN, Astrid Heidemann, GERTSEN, Frank and RIIS, Jens Ove. The Nexus of Corporate Entrepreneurship and Radical Innovation 4/359 LAUGEN, Bjørge Timenes, BOER, Harry and ACUR, Nuran. The New Product Development Improvement Motives and Practices of Miles and Snow’s Prospectors, Analysers and Defenders 1/85 LIGHTFOOT, Geoff, LILLEY, Simon and KAVANAGH, Donnacha. The End of the Shock of the New 2/157 MATHIEU, Chris. Transforming the Danish Film Field Via ‘Professionalization’, Penetration and Integration 3/242 MIDDEL, Rick, BOER, Harry and FISSCHER Olaf. Continuous Improvement and Collaborative Improvement: Similarities and Differences 4/338 MIETTINEN, Reijo. The Sources of Novelty: A Cultural and Systemic View of Distributed Creativity 2/173 MOEHRLE, Martin G., and WENZKE, Sven. Exploring Problems with Function Analysis. Experimental Insights for Team Management 2/195

Volume 15

Number 4

2006

doi:10.1111/j.1467-8691.2006.00412.x

434


NAPIER, Nancy K., and NILSSON, Mikael. The Development of Creative Capabilities in and out of Creative Organizations: Three Case Studies 3/268

VAN LUENEN, Henri G.A.M and VAN HARTEN, Willem H. Quality Management and Stimulation of Technology Transfer in a Research Institute 2/207

O’QUIN, Karen and BESEMER, Susan P. Using the Creative Product Semantic Scale as a Metric for Results-Oriented Business 1/34

YAMADA, Jin-Ichiro and YAMASHITA, Masaru. Entrepreneurs’ Intentions and Partnership Towards Innovation: Evidence from the Japanese Film Industry 3/258

PUCCIO, Gerard J., FIRESTIEN, Roger L., COYLE, Christina and MASUCCI, Cristina. A Review of the Effectiveness of CPS Training: A Focus on Workplace Issues 1/19

ZACKARIASSON, Peter, STYHRE, Alexander and WILSON, Timothy L. Phronesis and Creativity: Knowledge Work in Video Game Development

REHN, Alf and VACHHANI, Sheena. Innovation and the Post-Original: On Moral Stances and Reproduction 3/310 RICKARDS, Tudor and MOGER, Susan. Creative Leaders: A Decade of Contributions from Creativity and Innovation Management Journal 1/4 SCHWEIZER, Tanja Sophia. The Psychology of Novelty-Seeking, Creativity and Innovation: Neurocognitive Aspects Within a Work-Psychological Perspective 2/164 SLUTSKAYA, Natalia. Creativity and Repetition

2/150

SØRENSEN, Bent Meier. Identity Sniping: Innovation, Imagination and the Body

2/135

STYHRE, Alexander. Organization Creativity and the Empiricist Image of Novelty 2/143 TALKE, Katrin, SALOMO, Sören and MENSEL, Nils. A Competence-Based Model of Initiatives for Innovations

Volume 15

Number 4

2006

4/373

4/419

BOOK REVIEWS ABETTI, Pier A. Winning by Jack Welch (with Suzy Welch). 2/218 BURTON, Jamie. Product and services management by Avlonitis, G. and Papastathopoulou, P. 4/430 DA SILVA GOMES, Jorge F. Dilemmas of Leadership by Rickards, T. and Clark, M. 1/117 HARKINK, Erik. Innovation in Technology Alliance Networks by Lemmens, Charmianne. 1/118 HOSPERS, Gert-Jan. The Flight of the Creative Class: The New Global Competition for Talent by Florida, Richard. 3/323 HOSPERS, Gert-Jan. The Rise of the Creative Class: And How It’s Transforming Work, Leisure, Community and Everyday Life by Florida, Richard. 3/323 KRAAIJENBRINK, Jeroen. Knowledge Spillovers and Knowledge Management by Karlsson, Charlie, Flensburg, Per and Hörte, Sven-Ake. 4/431


No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

Recommend Documents