No title

EDITORIAL

63

Editorial

T

his second issue of volume 12 is in fact our first issue as new editors of Creativity and Innovation Management, with its sparkling new cover. As editors our initial challenge is to continue and strengthen the course and tradition that has been established by Tudor Rickards and Susan Moger. For the past 11 years they have done a great job, in close cooperation with the Blackwell team. We are very grateful for all their work and for their confidence in our efforts to continue their entrepreneurial initiative and practical and scientific focus. We are fully committed to further developing the depth and range of contributions that are published in the journal. In their editorial for this volume’s first issue Susan and Tudor introduced us as coming from the University of Twente in the Netherlands, and they mentioned that our contact with them had been developed within the context of the European creativity community. We would now like to take this opportunity to introduce ourselves to you more fully. Olaf Fisscher is Professor of Organization Studies and Business Ethics, and Petra de WeerdNederhof is Associate Professor of New Product Development, both in the Department of Technology and Organization of the Faculty of Business, Public Administration and Technology. Our research is guided by the notion that the organizational design of innovation processes can only be successful if it is based on a good insight into and understanding of social dynamic processes, stressing room for creativity and learning, the importance of an innovative climate, professional autonomy and shared leadership, collective mind, personal development, and social responsibility; and we feel that the organization of innovation should go hand in hand with the innovation of organization. Almost two years ago we were the guest editors of a Creativity and Innovation Management special issue that had the theme of Strategic Flexiblity (volume 10, issue 4). The articles in this issue reported on several of the projects from our faculty’s research programme ‘Organizing for Innovation’. The first and last articles in the current issue are other examples of the research projects within this programme.

© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.

In light of our own research projects and interests we feel very much at home with this journal, and the contents and scope set out by Susan and Tudor. We are very dedicated to continuing the intense and fruitful interaction between practice and theory, between doing and reflection, between knowledge and insight on the one hand and application on the other. Both scientific knowledge and the orientation on experiences and application in practice, must continue to be combined, as two sides of one coin. With respect to application of creativity and innovation, we are open for publications from industry, service, profit and not-for-profit, as well as from the field of consultancy. It will come as no surprise, from the characterization of our research interests above, that as well as aiming to contribute to advancing knowledge of and insights into individual capabilities to operate in a creative and innovative way, we want to stress the need to develop collective organizational competencies in the field of creativity and innovation management, for instance in new product development, research and development and the implementation of new technologies. In order to create such collective capabilities, an open, creative and innovative organizational method will be needed with a focus on human resources, social dynamics, leadership and corporate social responsibility, juxtaposed and intertwined with structural issues such as information infrastructure. The current issue contains five articles more or less bound by the theme of Innovation with a sense of reality. They discuss alignment mechanisms in innovation practice – doing justice to real life innovation experience in modelling innovation processes – and the importance of a practical, entrepreneurial orientation for innovation success, addressing ‘soft’ theories of change and last but not least providing guidance to experienced change consultants, while reflecting on the chaotic character of change processes. The first article is about research institutes taking the initiative for product innovation in strategic cooperation with industrial firms. The main research question concerns the configurations of suitable alignment mechanisms that should be applied

Volume 12

Number 2

June 2003

64

CREATIVITY AND INNOVATION MANAGEMENT

to the practice of product innovation. We report on two case studies in the field of energy research, following the so-called process-based contingency approach. In the second article Jan Buijs gives an extensive overview of the history of the Delft Design School. Based on his own in-depth involvement for many years in the field of new product innovation, his article explains and explores the main successive approaches in modelling the product innovation process. Taking as a starting point models representing the (normative) logical linear order, recent models try to show the more chaotic character corresponding with real life experience. We hope that you as our readers feel challenged by this article from Jan Buijs to write about your own institutes’ development in thinking about creativity and innovation processes. Helen Salavou and Spyros Lioukas focus on small and medium sized enterprises, and more specifically on the Greek food and beverages industry. Based on quantitative analysis they argue that the entrepreneurial orientation is a crucial factor in the success of radical product innovations in small and medium sized enterprises. In the next article Frido Smulders, Léon de Caluwé and Olivier van Nieuwenhuizen argue that in order to implement the new product successfully in operations more attention should be given to ‘soft’ theories of change. Next to the traditional focus on project management methods and on control, issues like learning and cooperation need to come in

Volume 12

Number 2

June 2003

the forefront. Also, the intervention strategies in the implementation process need to be well chosen for new product development to be successful. Implementation and organizational change is also the subject of the contribution by Klaasjan Visscher and Arie Rip. Following on from Jan Buijs’ discussion of the existing chaotic processes in new product design, in their article Visscher and Rip reflect on the chaotic character of change processes. They have developed a typology based on literature and a series of interviews with experienced change consultants which may help change the working practices of consultants and managers. With these articles we think we have compiled an issue with practical value as well as academic soundness. To do credit to the academic identity of Creativity and Innovation Management we aim to strengthen the review process as a double blind review procedure while maintaining the fast response time for submitted papers. The new procedures are described in the notes to contributors on the inside back cover. Last but not at all least we would like to draw your attention again to the 8th EACI conference on Cross Cultural Innovation, which will take place from 2 to 6 September 2003 in Mainz, Germany, hosted by Professor Horst Geschka. Please refer to their website: www.8ecci.org Petra de Weerd-Nederhof Olaf Fisscher

© Blackwell Publishing Ltd, 2003

ALIGNMENT AND ALLIANCES FOR RESEARCH

65

Alignment and Alliances for Research Institutes Engaged in Product Innovation. Two Case Studies Petra de Weerd-Nederhof and Olaf Fisscher Research institutes mainly engage in product innovation for the purpose of applying, testing and usually also transferring knowledge or technology. Managing product innovation processes in this type of environment leads to various problems with the establishment and managent of alliances and alignment. In this paper a systematic analysis of product innovation at two energy research institutes is presented, paying explicit attention to the strategic alliances in terms of alignment mechanisms. The cases illustrate clearly how the research institutes manage their external networks for product innovation, allowing comparison of the strengths and weaknesses of the two research institutes, and an indication of lessons to be learned from each other. The systematic analysis contributes to the identifying of the appropriate product innovaiton objectives to be pursued through strategic alliances, as well as the determination of suitable alignment mechanisms for product innovation. Lessons learned are presented in the area of context and contingency influences, crossing organizational barriers, differences in culture and the balancing of operational effectiveness and strategic flexibility.

Introduction

A

research institute, which leads the management of a product innovation process, faces several problems. Issues of alignment and alliances arise, including how to establish and manage external networks through a specific project. Research institutes initiate and participate in processes of product innovation beyond prototype development mostly for the purpose of applying, testing and transferring knowledge or technology. The primary output of these research institutes then consists of scientific products, of which the actual product (usually a prototype) to be developed is only one aspect. If these products are brought to market at all, it would be through the collaboration with external parties: marketing and manufacturing do not usually occur in-house. MESA+ for example, at the multidisciplinary research institute affiliated with the University of Twente in the areas of micro-system technology, micro-electronics and materials science, present a special page with so-called commercial partners on their website. Sixteen com© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.

panies, all active in the field of micro system technology, sensors, micro actuators, micro analysis systems and micro-electronics, are presented in a table indicating application areas and core products as well as services (development-prototyping-manufacturingtesting) provided. According to the website: ‘The partners can help you from your idea on paper to your readily packaged product’ (www.mesaplus.utwente.nl). This illustrates that for most product innovation or development activities research institutes have to form alliances with multiple external parties to be able to go through the whole trajectory. These alliances at the same time may serve the purpose of generating funding for the current product innovation activities as well as for future research, the latter for example through licensing. In terms of the different types of collaboration distinguished by Tidd, Bessant and Pavitt (1997), the partnerships that research institutes engage in for product development may be referred to as strategic alliances. From their overview of survey-based and case-based studies into patterns of collaboration, Tidd, Bessant and Pavitt typify strategic

Volume 12

Number 2

June 2003

66

CREATIVITY AND INNOVATION MANAGEMENT

alliances as having the following characteristics: they tend to have a flexible duration; they are often based on the rationale of low commitment between partners (in contrast with, for example, joint ventures or specially established research consortia), allow acess to markets, and have disadvantages, in the form of transaction costs, of potential lock-in and knowledge leakage. Here it should be noted that nearly all literature on types, rationales and (dis)advantages of technological collaboration is written from the viewpoint of the industrial company in the partnership. In contrast with this literature, this paper describes strategic alliances from the viewpoint of research institutes taking the initiative for collaboration in product innovation. When using the term ‘research institute’ we refer to the category of research organizations labelled as scientific institutional research organizations by Jain and Triandis (1990) based on Brooks (1968). According to Jain and Triandis ‘these organizations have a mission defined primarily in scientific terms; for example, advancement of high energy physics or molecular biology. Such research organizaitons follow some sort of a coherent program adapted to changing frontiers in their area of interest’ (Jain and Triandis, 1990, p. 10). Unlike socalled mission-oriented research organizations, which typically are industrial research laboratories directed to fulfilling the objectives and the mission of the parent organization rather to the development of science perse, or academic research organizations, which are usually concerned with small-scale basic research carried out in academic departments of universities, these research institutes tend to be more independent and are often at least partly funded by government. NIAB, a (since 1996 privatised) UK-based company providing research, services and information to the agricultural and food sector, described in Visser et al. (2001), is an example of such a research institute funded by both governmental and non-governmental research funders. At their website, NIAB mention the UK Ministry of Agriculture Fisheries and Food, the Food Standards Agency and the European Commission as some of their funders (www.niab.com). It is important to note that the research institutes’ objectives for the alliances should not be considered to be exclusively focused on what may be called the operational effectiveness of the product innovation process. Operational effectiveness refers to the degree to which product development contributes to realising goals currently set (i.e. a concrete new product). Operational effectiveness has a product concept (fit with market demands; fit with competencies) as well as a process (speed;

Volume 12

Number 2

June 2003

productivity; flexibility) component. The tendency for the contribution of the research institute to the product innovation efforts may be to over-emphasise product functionality and building technical competencies as performance factors, at the cost of, for example, speed (overall time to market) and the improvement of project management skills. Furthermore, research institutes will often be particularly interested in the contribution to their own strategic flexibility: the ability to adapt to, anticipate or even create future performance requirements (tomorrow’s innovations). This may easily lead to a complex situation within the alliance where conflicting demands and short- and long-term considerations have to be continuously balanced. The main assumption1 (and rationale) for this paper is that further improving the competence to manage product innovation in external networks, if only for reasons of efficiency, is an important factor for future overall performance of research institutes, in terms of operational effectiveness as well as strategic flexibility. From this starting point in the next sub-section the objective for this paper can be formulated as follows: to provide an insight into the way research institutes form alliances and design and manage alignment mechanisms for product innovation, through a systematic analysis of two case studies where research institutes take the lead in a network of parties developing a concrete product. The ‘lessons learned’ from these case studies should provide valuable input for both practice and academia in two ways. In the first place they are expected to contribute to the identification of the specific objectives (the WHY question) to be pursued through strategic alliances; specifically taking into account the context and contingencies relevant to the product innovation efforts and the parties engaged in the network. The second contribution is to the determination of suitable alignment mechanisms (WITH WHOM and HOW) to be used in these alliances, leading to operational effectiveness and strategic flexibility of product innovation processes and the research institutes involved.

Method and Conceptual Frame The systematic analysis of alliances and alignment of research institutes engaged in product innovation is rooted in a process-based 1

Based on the viewpoint that research institutes pursue sustained ‘competitive advantage’ or at least longer-term continuity.

© Blackwell Publishing Ltd, 2003

ALIGNMENT AND ALLIANCES FOR RESEARCH

contingency approach (Boer and Krabbendam, 1991; de Weerd-Nederhof, 1998) explicitly acknowledging that organizational systems do not operate in a vacuum: they depend on their context for their input, output and resources. Specific contextual elements relevant to a particular organizational subsystem can be identified through an inside-out examination of the input-output relationships of the primary, management and support processes. Alignment with the context is achieved through combinations of all elements of the product development system: goals, processes, technology and organizational arrangements. The process-based contingency approach is also one of functional equivalence: contingencies will determine the functions that the organisation must perform but not its specific structure (Gresov and Drazin, 1997). These starting points have led to the following systematic steps for description and analysis. The first step is the identification of relevant elements in the context (WHO) i.e. the process of partner search. Essentially, the specific objectives to be pursued largely determine the choice and characteristics of the partnership. Alongside the fact that research institutes need to find external partners that will carry out producer business functions such as manufacturing and product marketing/sales, the process of partner search is also influenced by the type of product and industry or application area. Partner search by research institutes also specifically serves the purpose of acquiring funding for the NPD trajectory (most funding bodies require proposals submitted by joint partners) which leads to another set of partner requirements. According to Hart and Baker (1994) the accommodation of third parties ensuring for example user relevance and supplier support is required. User input can come from direct contact with end-users or through branch organizations; direct contact and partnership with suppliers is more relevant in areas where scarce or unique components are needed. The need for societal embedding of a product innovation, for example in case of bio-technology for genetically modified food, again broadens the range of partners with which to align. Societal embedding captures a broader notion of success characterised by three dimensions: (i) integration in relevant industries and markets; (ii) admissibility according to the rules and standards set by government agencies for the sector; and (iii) acceptance by the public (Deuten, Rip and Jelsma, 1997).

© Blackwell Publishing Ltd, 2003

67

After identifying and briefly characterizing the context, the second step is the identification and description of the different alignment mechanisms used (goals, processes, technology (people and means) and organisational arrangements). Studies on determinants of new product success (e.g. Cooper, 1979; Cooper and Kleinschmidt, 1995, that cite among others Rothwell et al. on Project SAPPHO, 1972, 1974, and Montoya-Weiss and Calantone, 1994) list variables and factors pointing to alignment with the context: extensive customer–producer interfacing; better internal and external communication; better co-ordination of R&D, marketing and production; technological and marketing synergy; internal/external relationships (Cooper, 1979; Cooper and Kleinschmidt, 1995). Kahn (1996) lists a mixture of techniques, tools and organizational arrangements in his model of interdepartmental integration, that could be translated to collaboration within alliances. Collaboration, established through collective goals, mutual understanding, shared resources, common vision, informal activities and esprit de corps, according to Kahn’s findings represent a major factor for success. Interaction mechanisms such as meetings and exchanging reports also seem to influence success positively. These two systematic steps were applied in two case studies studied from the viewpoint of the research institute. The unit of analysis was the product innovation network initiated by the research institute. First, the specific objectives to be pursued with product innovation efforts were discussed with several people per case, in terms of operational effectiveness and strategic flexibility. Note that both performance dimensions have been described in terms of product concept effectiveness (fit with market demands, produceability and other environmental constraints) as well as process effectiveness (speed, productivity, flexibility), following the reasoning that strategic flexibility is an important prerequisite for future operational effectiveness. The same people were asked to make statements regarding the performance of the product innovation efforts, again in terms of operational effectiveness and strategic flexibility. Further withincase sampling (which activities, processes, events, times, locations, etc. to ‘sample’ within the case; see Miles and Huberman, 1994) took place following the already mentioned ‘insideout’ approach. The primary product development process of a specific project, and the people concerned in carrying it out, were first described. Then gradually the description was expanded to include the product innovation management and support processes, and the

Volume 12

Number 2

June 2003

68

CREATIVITY AND INNOVATION MANAGEMENT

contextual parties relevant for the alliance, always interviewing those people who were connected to the project. In order to enable some generalization, a number of questions were included asking for notable differences between the project being considered in detail and other projects in which the research institutes had been involved. For the actual data collection so-called mapping tools were developed (based on de Weerd-Nederhof, 1998). The characteristics of the alignment, to be described in the case studies, are summarized in the mapping tool presented in Table 1. The basis for the data gathering to complete the mapping tool was the following set of open questions, to be answered for each of the identified functions and parties in the product development context: • • • • • •

What kind of alignment is necessary? How does alignment take place? Through which alignment mechanisms? Is the alignment satisfactory? Are there any bottlenecks? Have there been any contextual changes recently; • If so, what have been their effects? Answers to these questions were obtained from researchers and managers of the casestudy companies and the partners involved in the projects. From the answers, the alignment mechanisms in use were first identified and described so that they could then be characterized in terms of interactions in internal and external networks as recommended by Hart and Baker (1994). Five characteristics suggested by Hart and Baker, based on Biemans (1994) can be recognised in the mapping tool: type of interaction (vertical/horizontal; competitive/ complementary); purpose of interaction (task performance or task simulation);

intensity of interaction; duration of interaction; formalisation of interaction. Next to these five characteristics, the mapping tool also includes an indication of the extent to which the alignment is formal or informal, which points to, for example, the embedding of an alignment mechanism in structure or culture. More detailed illustrations of the described alignment in the sample projects should also be listed in the mapping tool. In the data gathering these illustrations often took the form of stories and anecdotes. Furthermore, perceived strengths and weaknesses mentioned by the interviewees are explicitly listed in the mapping tool to facilitate linking the way the research institutes align with the partners in the alliance to the operational effectiveness and strategic flexibility of their product development processes. This linking was a crucial step in the systematic analysis of the data.

Case Studies: HeatCo and BearInc The two research institutes considered in this section, HeatCo and BearInc, are both active in the field of energy research. HeatCo is an energy research institute (founded in 1955), employing 900 people distributed between seven different business units. Each autonomous business unit covers a specific energy product-market combination. The case study was focused on one of these business units (fossil fuels). HeatCo’s overall mission is energy innovation: the development of technologies for a safe, clean and efficient supply of energy. Its main goals, in terms of which the different scientific products of HeatCo can be characterised, are threefold: (i) contribute to mid- and long-term energy policy (knowledge development);

Table 1. Mapping Tool for Alignment Mechanisms For each alliance with parties and functions in product innovation context describe: alignment strengths and weaknesses (as perceived) illustration (sample project)

Volume 12

Number 2

June 2003

– content (purpose and reason for alignment) – how: – mechanism(s), including type of alignment: combination of goals, processes, technology (people and tools & techniques), and organisational arrangements – frequency/duration – intensity – formal/informal

© Blackwell Publishing Ltd, 2003

ALIGNMENT AND ALLIANCES FOR RESEARCH

(ii) serve as a ‘knowledge base’: expanding knowledge on energy technology (knowledge development); (iii) contribute to concrete (short term) practical applications (product development). It is a specific company policy not to manufacture any product developed in-house, this has to be done by industrial companies licensing the technology from HeatCo. Related to HeatCo’s goals formulated above, there are three types of ‘customers’ or external target groups. First of all HeatCo receives a core subsidy from the government (about 40% of total income). The remaining 60% is earned through research contracts with governmental (national or European) agencies, often in co-operation with industry, through direct contracts with companies and through the licensing of in-house developed technologies. Basically, government funding is used for the longer-term activities (basic and applied research), for starting up product development activities and for complementing research contract funding (for example, some EU grants do not cover full costs). Note that up to the beginning of the 1980s HeatCo was almost 100% government-funded. When the annual subsidy was drastically decreased a reorganization, leading to a number of redundancies, was carried out, and HeatCo had to start earning income through co-operation with industrial parties. Towards the end of the 1980s, after a number of reasonably successful years, the financial situation was again becoming problematic and another turnaround was considered necessary. In 1990, the overall company structure was changed from matrix to business units, as part of a major ‘reorientation’ aimed at improving the marketorientedness of HeatCo. In the near future, further changes in government funding will take place, obliging HeatCo to establish formal alliances with industries covering most of their activities. BearInc’s parent company is a semiprivatized energy utility, subject to a high level of regulation, but with the majority of its business within a single energy sector. Its main customers are large energy utilities throughout the world. For this case study the focus is on BearInc which is a small business unit within the parent company, recently established as a specialist product development centre (at the time of the case study it had been in existence for less than one year). The main goals of this business unit are to develop products based on the parent’s technologies (physics based and process based). BearInc has a flat structure, and is much less hierar-

© Blackwell Publishing Ltd, 2003

69

chical than its parent. The number of staff is expected to grow rapidly and, with more than 50 people, a line structure is envisaged with discipline groups led by senior engineers. BearInc aims to have a ‘flexible workforce’: only some of the people are actually employed by BearInc, others are on contract or seconded to BearInc from the parent company or universities. BearInc has established an internal marketing function but, until now, for other functions such as purchasing, accounting and personnel management, and even for manufacturing of prototype components, BearInc has relied on the parent company. BearInc was set up as a profit centre, and in principle, all income should be generated by contracts with internal (sister companies or other business units or departments belonging to the parent) or external customers. In the following section the case comparison of the alignment and alliances for product innovation at both case-study companies is presented, which starts with a brief discussion of the differences between HeatCo and BearInc in terms of contingencies. The cases are compared in terms of the specific objectives to be pursued through strategic alliances (WHY), the actual alliances established (WITH WHOM) and the alignment mechanisms used (HOW) for interaction and collaboration, related to performance in terms of operational effectiveness and strategic flexibility.

Contingencies Table 2 summarizes some basic information about HeatCo and BearInc, giving insight in relevant contingencies. HeatCo and BearInc have a number of contingencies in common. They both develop scientific products, do not undertake any manufacturing themselves, are active in the energy sector, and rely on universities for basic knowledge. However, they are not exactly the same type of research institute. The main difference is that BearInc was explicitly set up as a product development centre, whereas, at HeatCo, the development of products (prototypes) is only one of their activities. In other words, for HeatCo product development is explicitly considered as a ‘vehicle’ for developing, applying and transferring knowledge, whereas for BearInc the process of product or prototype development is their core business, and also their core competence. Furthermore, HeatCo is much larger and has a business unit structure in place, whereas BearInc is very small and has as yet no hierarchy. There is also a difference in funding: BearInc predominantly relies on

Volume 12

Number 2

June 2003

70

CREATIVITY AND INNOVATION MANAGEMENT

Table 2. Relevant Contingencies for HeatCo and BearInc Contingencies Basic Information • Products • Mission/corporate strategy

• Status • # employees • Structure Competitive Environment • Customers • Suppliers • Other parties • Environmental constraints • Competitive position

HeatCo

BearInc

scientific; energy technology contribute to governmental energy policy; build energy knowledge base; develop practical applications of energy technology; no manufacturing foundation 900; 140 in Business Unit Business Units; within BU also product-based groups government (agencies); producers and users in energy sector universities for basic knowledge technologies can only be licensed within country the specialist in energy technology development

scientific; physics and process based prototypes worldclass centre of expertise for product development; exploitation of technologies; no manufacturing profit centre within parent company

(internal) customers paying for their prototype development activities, whereas, at HeatCo, funding is a combination of government subsidies and contracts with governmental agencies and industrial partners.

Specific Objectives For both research institutes, technological product performance is seen as an ‘order winner’, they take pride in ‘over-achievement’ in this area. BearInc, in their brochure, state that they have ‘a commitment to produce . . . products . . . which exceed client expectations’. The set of critical product development functions to be achieved through alliances for the two cases show some similarities but more differences. This is not surprising given the differences between the two companies already noted above. Concerning operational effectiveness, the differences are in terms of process performance: at HeatCo the short development time of an overall product development process is not so much a critical function as is the timeliness of the individual projects in which the process is usually divided. For BearInc however, a short devel-

Volume 12

Number 2

June 2003

30–35 all in NPD flat, no hierarchy; line structure envisaged mainly sister companies and other business units; sometimes external universities for basic knowledge ultimate benefit for parent company unique position

opment time is one of its selling points. Even though price and product cost were indicated as being of some importance to HeatCo and BearInc, they were not mentioned as specific objectives for product development. This may be related to the fact that both research institutes have a unique position in the market, and that they rely on their customers and endusers to give indications on manufacturability and price level. Finally, concerning strategic flexibility, the main difference between the two research institutes is that HeatCo largely develops its own technological competencies (sometimes in alliance with universities, but usually not for product development), whereas BearInc rely on their extensive networks within the parent company, and their ability to quickly absorb external knowledge (also from universities).

Alliances Table 3 lists the parties with which HeatCo and BearInc formally or informally align for their product development activities. In practice in both cases, alignment and collaboration within these alliances take place largely in the

© Blackwell Publishing Ltd, 2003

ALIGNMENT AND ALLIANCES FOR RESEARCH

71

Table 3. Parties involved in alliances for product development at BearInc and HeatCo BearInc

HeatCo

within company / parent

customer, (end) user, other BUs for specialists and e.g. calculations

other BUs for computational physics and prototype building / process development

external

universities, manufacturer

manufacturer, subsidising agency, customer, (end) user

same way: meetings (not many), reports for example describing the specifications or intermediate results, and via telephone, fax and email. At BearInc, the customer can officially be part of the product development team: sometimes these representatives are seconded to the centre for the duration of the project. For both cases it is very important to establish good contacts with customers, possible manufacturers and end-users etc. at the very start of the product development process. At HeatCo this is done via a systematic ‘target group approach’ and the ‘rule’ is that in each phase representatives of each target group should be involved with the product development activities. At BearInc, at least in the sample project, the early involvement of the manufacturer was more or less a coincidence where it should have been a deliberate policy.

Alignment Mechanisms In both cases (especially at BearInc), people were relied upon as the dominant alignment mechanism: the characteristics (skills, roles fulfilled) of product development workers and managers played a very important role in achieving fit with market demands and environmental constraints where operational effectiveness is concerned, and also for strategic flexibility. The people carrying out the product development processes are, in both cases, highly skilled academically, with a technical background. At BearInc however, people have somewhat more general skills: the specialists needed for a specific project are brought in from outside the centre (from the company or universities); whereas at HeatCo, in-house specialists carry out the development work. BearInc’s researchers are slightly more goal oriented (customer/end-product) compared with HeatCo, where we found curiositydriven technologists, with one of them always playing the product champion role persistently and enthusiastically.

© Blackwell Publishing Ltd, 2003

In the list of important alignment mechanisms, organizational arrangements are a close ‘second’. It is noteworthy that both HeatCo and BearInc preferred to refer to the (internal combined with external) networks in which they carry out product development as ‘multifunctional’, and to their informal, but well-communicated, way of working as a stage-gate procedure. For the most part however, the organizational arrangements through which alignment is achieved or stimulated are the innovative climate on the one hand, and the ‘lean’ overall structure on the other. Next to this, the socialdynamic activities in operational product development management (for BearInc more or less integrated in the leadership style of the managing director) are perceived to be crucial to performance. Related to the characteristics of people, the innovative climate and the relative lack of formal organizational arrangements, which are all very much in agreement with what is reported in literature as the characteristics of R&D organizations (e.g. Jain and Triandis, 1990), is the importance of support processes, particularly training. These are successfully in place at HeatCo, and perceived by BearInc to be one of the solutions to deal with the shortfall in future product development process effectiveness. Table 4 summarizes the alignment mechanisms used in both cases.

Operational Effectiveness and Strategic Flexibility Last but not least, product development performance as perceived by the managers of both companies, shows an almost similar perception of performance in terms of (operational) product concept effectiveness: excellent technical performance, but where price and manufacturability can be difficult points especially when products have to be produced in large quantities. In terms of process perform-

Volume 12

Number 2

June 2003

72

CREATIVITY AND INNOVATION MANAGEMENT

Table 4. Alignment Mechanisms Contributing to Operational Effectiveness and Strategic Flexibility BearInc People

Organizational arrangements

Processes

generalists within their disciplines; specialists brought in from outside formal multidisciplinary teams; no formal meetings except reviews; stage-gate procedure (not formally imbedded); innovative climate & lean structure training: no formal programme (workshop on procedure at start up) operational management

ance, BearInc at the moment ranks itself relatively high in terms of speed. The operational process flexibility is also considered to be quite satisfactory, and not only within one project, since people can be switched between projects more easily than at HeatCo because they have more generalist skills. In terms of strategic flexibility, both units are well satisfied with technological performance and learning; HeatCo seems to have the edge because new knowledge can be shared by more specialists and thus they are less reliant on one or two individuals. Organizational learning about the process of product development is said to be the core competence of BearInc and, up to now, they consider themselves to be performing quite well, although they rely heavily on informal individual learning. There is a great danger that, if the centre grows too fast, they will be unable to maintain the learning pace. At HeatCo the general opinion is that learning from (organizational) mistakes should be systematized and improved, starting for example with official post-project or programme reviews.

Discussion and Conclusion The purpose of applying the framework for the description and analysis of alignment mechanisms in product development in two cases was to learn about the use of these alignment mechanisms in strategic alliances as used by research institutes in developing new products. The ‘lessons’ to be learned should be valuable in both practice and academia, and may be found in the following areas.

Volume 12

Number 2

June 2003

HeatCo specialists; curiosity and technology driven; product champions multidisciplinary teams not officially established; regular meetings; procedure imbedded through training; innovative climate and ‘lean’ structure training: formal programme, e.g. for project management and commercial skills operational management

Influence of Context and Contingencies Initially, in both cases, we saw the strong influence of the context on the organization of product development activities, the specific objectives to be pursued, and specifically the alignment mechanism used to interact and collaborate. In the past, research institutes, such as HeatCo and BearInc’s parent company, were exclusively focused on the development of knowledge and technology. The potential of a wide range of applications, sometime in the future, seemed to be enough for the government to continue funding (especially in the case of HeatCo). Real marketing effort to obtain this funding was not required; this function was fulfilled by making the ‘right’ promises and telling legitimizing stories.2 Especially for HeatCo, the external legitimization context has changed dramatically, especially in terms of the basis on which funds are allocated. As a consequence of these changing contingencies, product development has become more important. In the sample projects, we have seen the establishment of separate units. BearInc is in fact such a unit. In HeatCo an extensive reorganization has led to a business-unit structure. Despite these 2

Note that in the interaction and collaboration with target groups in the sample projects, the role of stories (narratives) has also proven to be very important. Deuten (1994) explains the influence of narratives on the goal directedness of product development projects. He stresses the need for robust ‘Start up’ stories, ‘Project’ stories and the availability of ‘Repair’ (Anticipative) stories, as well as the need for a balance between prospective and retrospective tales (see de Weerd-Nederhof, Fisscher and Kerssens-van Drongelen, 1995).

© Blackwell Publishing Ltd, 2003

ALIGNMENT AND ALLIANCES FOR RESEARCH

structural solutions there was still a need for deep changes in culture and strategy. The need for a strategic approach to partnering seems to have been particularly inevitable. At HeatCo, specifically for product development, the introduction of the systematic target-group approach is an example of a relatively successful improvement activity. Both research institutes, in a growing number of cases, are now partners in the whole production chain and are faced with the challenge of fulfilling a crucial role. Crucial to the technical realization of functionalities, to the realization of short and above all reliable lead times, and in the ability to communicate and co-operate with external partners in marketing and manufacturing.

Crossing Organizational Barriers HeatCo and BearInc are confronted with the need to cross organizational barriers in order to engage successfully in product development. It is necessary to define, to organize and to manage strategic alliances not only between the research institutes and their external partners but also internally, for HeatCo with, for example, other business units, and BearInc with departments of the parent company. The fact that both case-study companies now refer to the internal combined with external networks in which they carry out product development as ‘multifunctional teams’ is already an indication that the companies are at least aware of this necessity. The growing stress on legitimization in advance of product development projects, and evaluation afterwards, may also be seen as further indication. However there still seems a long way to go. In addition to improved processes for partner search, the actual management of the alignment in the product development process also needs facilities and supportive mechanisms that extend the scope of the individual project. The actual alignment at both case companies takes place through meetings, e-mail, telephone and the exchange of documents etc. According to Kahn (1996), as discussed in the section on alignment mechanisms in the first part of this article, these are all facilitating interactions rather than collaboration.

Balancing Operational Effectiveness and Strategic Flexibility In terms of product concept effectiveness we may say that, to some extent at BearInc and certainly at HeatCo, operational effectiveness performance is not considered to be optimal. Although they are both aware of the need for

© Blackwell Publishing Ltd, 2003

73

further improvement in their abilities to meet criteria of functionality, cost and time, so far both case-study companies feel that they have not been fully successful. Despite the improved alignment through such actions as changing the organizational arrangements; the extensive reorganization within HeatCo (the business unit structure), the establishment of BearInc by its parent company; and intensified and formalized strategic alliances; both institutes regard their product development efforts as not fully competitive with industrial companies. On the other hand, future product concept effectiveness (strategic flexibility), especially in terms of creating a solid knowledge base seems to be ensured. Both companies, but especially HeatCo, demonstrated their competitiveness in this respect. Where process effectiveness is concerned it seems that considerable improvement in both operational effectiveness and strategic flexibility is needed. Improved performance in particular could be achieved by introducing alignment mechanisms aimed at collaboration, such as the setting of common goals. Overall, from our case studies, it can be seen that there is a need for a balance between operational effectiveness and strategic flexibility but that it is difficult to achieve this balance.

Differences in Culture The case-study comparison of the most relevant alignment mechanisms (people and organizational arrangements) reveals the influence of what may be called cultural differences. Research institutes such as HeatCo can be characterized as more or less a ‘monoculture’; relating primarily to their own professional field, people are focused on the development of knowledge. Within the organization, researchers are not challenged by manufacturing and marketing people, and traditionally neither by a strong directive management. BearInc, on the other hand, was established to break the monoculture of the parent company by establishing what may be called a more ‘productive tension’ between different cultures within the company. For example customer representatives are brought into the team, which thus becomes truly multifunctional. This produces a stronger need for communication across functional and organizational barriers, which brings specific objectives such as shorter lead times, awareness of cost and functionality demands, to the fore. The sample projects however show that the culture at BearInc is only relatively better than at HeatCo. Compared to the culture of the industrial companies with which BearInc

Volume 12

Number 2

June 2003

74

CREATIVITY AND INNOVATION MANAGEMENT

aligns, their culture can still be characterized as discipline and profession oriented with a relatively low commitment to industrial product development targets. The further integration of these industrial criteria into their own culture is crucial to improving future performance of product development at both companies. The focus should not be on creating a new monolithic industrial culture, but rather in creating the ability to cope with cultural differences.

Conclusions Concluding, in line with the stated objective, we have provided a systematic analysis of two case studies where research institutes take the lead in a network of parties developing a concrete product by considering: • the specific objectives to be pursued (WHY) in strategic alliances for both operational effectiveness and strategic flexibility in product development carried out by research institutes. An example from the case studies is the uncovering of the increasing importance of speed in order to achieve the right fit with market demands; • strategic alliances (WITH WHOM) incorporated in the product development system across organizational barriers; • alignment mechanisms (HOW) used in the strategic alliances and their impact on operational effectiveness and strategic flexibility; especially the importance of people and organizational arrangements. Thus insight was achieved into how research institutes align with their partners during product innovation, linked to performance in terms of operational effectiveness and strategic flexibility. Research institutes will benefit from uncovering their gaps and bottlenecks by following the same systematic steps when forming alliances and designing and managing their alignment mechanisms. The lessons learned from these two cases alone point to the possibility of uncovering successful configurations for product development as carried out by research institutes, which is one of the ultimate goals of the research underlying this paper (de Weerd-Nederhof, 1998). One envisaged track of further research is to analyse a larger variety of cases exploring the differences between, and variety in, specific objectives; type of strategic alliances and alignment mechanisms. This would be aimed at formulating testable propositions concerning (partially) successful product innovation configurations, and building a database of noteworthy cases, explicitly including contin-

Volume 12

Number 2

June 2003

gencies such as corporate strategy and the (competitive) environment. Such a database would be a contribution that in practice could provide indications as to how to close gaps and overcome bottlenecks.

Acknowledgements The authors want to thank the interviewees at HeatCo, BearInc and their partners for their co-operation. The case study research at HeatCo and BearInc was carried out jointly with Rob Bartholomew from Manchester Business School’s R&D Research Unit. It was funded by the British Council and the Dutch foundation for Scientific Research NWO.

References Bartholomew, R.A. and P.C. de Weerd-Nederhof (1997), Appraisal, Selection and Organisation of New Product Development Projects. Case Studies of two European Energy Sector Companies, 4th EIASM Product Development Conference, Stockholm, Sweden, May 26–27. Biemans, W. (1992) Managing Innovations within Networks, Routledge, London. Boer, H. and Krabbendam, J.J. (1991) Organising for market-oriented manufacture, POMS Conference, New York City, USA, November 11–13. Brooks, H. (1968), The Government of Science. The MIT Press. Cambridge, MA. Brown, S.L. and Eisenhardt, K.M. (1995) Product development: past research, present findings and future directions, Academy of Management Review, 20, 2, 343–78. Chiesa, V., Coughlan, P. and Voss, C.A. (1996) Development of a Technical Innovation Audit, Journal of Product Innovation Management, 13, 105–36. Cooper, R.G. (1979) Identifying industrial new product success, Industrial Marketing Management, 8, 124–35. Cooper, R.G. and Kleinschmidt, E.J. (1995) Benchmarking the Firm’s Critical Success Factors in New Product Development, Journal of Product Innovation Management, 12, 374–91. Deuten, J.J. (1994) Ironie en produktontwikkeling, Doctoral Thesis University of Twente, Enschede (confidential, in Dutch). Deuten, J.J., Rip, A. and Jelsma, J. (1997) Societal embedding and product creation management, Technology Analysis & Strategic Management, 9, 2. Gresov, C., and Drazin, R. (1997) Equifinality: functional equivalence in organisational design, Academy of Management Review, 22, 2, 403–28. Hart, S.J. and Baker, M.J. (1994) The multiple convergent processing model of new product development, International Marketing Review, 11, 1, 77–92.

© Blackwell Publishing Ltd, 2003

ALIGNMENT AND ALLIANCES FOR RESEARCH

Jain, R.K. and Triandis, H.C. (1990) Management of Research and Development. Managing the unmanageable. John Wiley & Sons, New York. Kahn, K.B. (1996) Interdepartmental Integration: a definition with implications for product development performance, Journal of Product Innovation Management, 13, 2. Miles, M.B. and Huberman, A.M. (1994) Qualitative Data Analysis. An expanded sourcebook., 2nd edition, Sage Publications, Thousand Oaks, CA. Montoya-Weiss, M.M. and Calantone, R.J. (1994) Determinants of new product performance: a review and meta analysis. Journal of Product Innovation Management, 11, 5, 397–417. Rothwell, R. (1972) Factors for success in Industrial innovations, from Project SAPPHO- A comparative study of success and failure in industrial innovation, SPRU, Brighton. Rothwell, R., Freeman, C., Horsley, A., Jervis, V.T.P., Robertson A.B. and Townsend, J. (1974) SAPPHO updated-project SAPPHO Phase II, Research Policy, 3, 258–291. Tidd, J., Bessant, J. and Pavitt, K. (1997), Managing Innovation. Integrating Technological, Market and Organizational Change, John Wiley and Sons, Chicester.

© Blackwell Publishing Ltd, 2003

75

Visser, S., Kerssens-van Drongelen, I., WeerdNederhof, P.C. de, and Reeves, J. (2001) Design of a Research Performance Measurement System: The Case of NIAB. Creativity and Innovation Management, 10, 4, 259–268. Weerd-Nederhof, P.C. de, (1998) New Product Development Systems. Operational Effectiveness and Strategic Flexibility, Twente University Press, Enschede. Weerd-Nederhof, P.C. de, Fisscher, O.A.M. and Kerssens-van Drongelen, I.C. (1995) Managing quality in R&D: Social-dynamic aspects and the importance of generating (new) organisational configurations, 1995 International PDMA conference, Minneapolis, USA, October 11–14.

Petra C. de Weerd-Nederhof is Associate Professor New Product Development, and Olaf A.M. Fisscher is Full Professor of Organisation Studies and Business Ethics. Both work at the School of Business, Public Administration and Technology, University of Twente, the Netherlands

Volume 12

Number 2

June 2003

76

CREATIVITY AND INNOVATION MANAGEMENT

Modelling Product Innovation Processes, from Linear Logic to Circular Chaos Jan Buijs Product innovation is the focal point of the Delft Design School in the Netherlands. During its more than thirty years of existence different models of the product innovation process were and are used for education and for research. This paper will describe the development of these models. The first models tried to describe the product innovation process in a logical linear order, but recently this logical order has come under discussion. The most recent models try to show the more chaotic character of the product innovation processes in real corporate life. Although this chaotic model better reflects the product innovation practice, for educational purposes it seems to be less useful than the original logical ones. For our teaching we propose the two versions (logic and chaos) of our innovation model as two sides of one coin. This innovation coin is without proper value with one side left blank.

Introduction

T

his paper describes the development of modelling product innovation processes and its use in education and research. It starts with the introduction of models from all over the world, which were taught at the Delft Design School (second section). That triggered internal discussions and led to the first models of their own, which were used for educational purposes only (third section). Parallel to the developments by the Delft Design School, new insights were also being produced abroad, as well as by Delft graduates after entering the professional world (fourth section). In the fifth section a large-scale empirical research project is described, in which more than 150 Dutch companies empirically tested a four-stage innovation model. In the sixth section an excursion is made to some interesting developments in modelling the creative problem solving process. Scholars in the creative problem solving field have been questioning the logical sequence of their models of the creative process and made some openings to more circular and more circumstantial models. The seventh section describes the first such attempts within the innovation field. Finally, the latest developments in Delft are described introducing a double model, which questions

Volume 12

Number 2

June 2003

the logical order of all the stages of the innovation process. Under the double model, there is no strict order: every stage can be the starting point of an innovation process and every stage can be followed by any other stage. Chaos from practice is now made visible. Teaching innovation using to this model has proved to be less useful: students prefer the more (chrono)-logical models. So the other side of the double model is a circular logical step-by-step version. We now teach them to use these two different model versions describing the same product innovation process: (1) a four-stage model describing the chaotic product innovation process on a very abstract level, and (2) a twenty-step model on a very concrete level which describes the product innovation process on a orderly and more (chrono)-logical way. And we also teach how these two model versions are intertwined and linked together. It is not chaos or logic, but it is chaos and logic – two sides of one coin.

Early Delft Period The Delft Design School, officially the Delft School of Industrial Design Engineering, had its first students in 1963. Right from the begin-

© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.

MODELLING PRODUCT INNOVATION PROCESSES

ning the curriculum was designed around a certain model of the product innovation process. Modelling innovation processes was seen as the way to become an accepted academic discipline. Modelling product innovation processes originates from the design research movement in the UK on its first conferences on Design Methods (see for instance Gregory, 1966; Jones 1970). The ideas developed in that period inspired the founding fathers of the Delft Design School in such a way that the core curriculum was based on those methods and models (see de Wilde, 1997). In those early days design researchers and design educators were interested in developing a systematic approach to product design, product development and product innovation, primarily to help professionals in the field to structure and lighten their work, and also to find better ways of teaching new design professionals. Hoping to be accepted as a serious academic domain, they also wanted to become as scientific and rigorous as possible, which set the tone for logical reasoning. The first thing design researchers did, based on their own practices as professional product designers, was to cut the product design/innovation process into little pieces, which they ordered in a kind of logical way. The pieces or steps were as small as possible, in order to be handled with the existing design techniques. The beginning of the product design process was usually seen as a product idea, followed by a stage in which this idea was analysed and developed into a product concept. In the next

77

stage this concept was developed and materialised into a working model or prototype. This prototype was then engineered for manufacturing, and finally the new product was launched on the market. Discussions among the design researchers and educators were about the number of stages, the names of the stages and the activities within those stages. The reasoning behind this division into stages was considered also to be important. For instance, some used the activities as the basis for the division, others looked for the departments within the company where these activities should take place. Under to a department-based division the steps include R&D department, design department, engineering department, production department, and finally marketing department. Using an activity-based division the product design process the steps include idea generation, idea screening, commercial evaluation, technical development, testing and commercialisation (see Figure 1). In the two models two more or less separated ways of thinking are visible – on the one hand, the engineering-based view, and on the other the more commercial-based view. The engineering-based view paid little or no attention to elements like market needs or market analysis. The commercial-based view took the technological aspects more or less for granted. In the two examples in Figure 1 it is clearly visible that the department-based view has an engineering background (four out of the five steps have a strong technological emphasis), and the activity-based view has a commercial background (only one out of the six stages is

Source: Saren, 1984

Figure 1. Two Early Models of the Product Innovation Process. Above is the Department-Based View, and Below, the Activity-Based View

© Blackwell Publishing Ltd, 2003

Volume 12

Number 2

June 2003

78

CREATIVITY AND INNOVATION MANAGEMENT

technology based). Both ways of thinking look at the product innovation process in isolation. There was little or no contact between these two worlds – the engineers were talking about the product creation process; the marketeers were talking about new product development. Nobody made explicit the relationships between the product innovation process and the company or its competitive environment. This was the situation when I entered the Delft Design School as a student in 1970. Design Methodology was considered to be the scientific core of product design or product innovation education. But the Delft Design School also tried to bridge the gap between the engineers and the marketeers. In its compulsory study programme both engineering courses and marketing courses were offered. Product design or product innovation was considered to be the integration of all different kinds of knowledge fields, and the Delft product designer was educated as a generalist. In those days the Delft Design School was not even a separate institute. It was an Interdepartmental School founded by the School of Mechanical Engineering (responsible for the engineering and technology-based courses), the School of Architecture (responsible for design methods, aesthetics and styling courses) and the School of General Sciences (responsible for courses in marketing, consumer behaviour, ergonomics, economics, management sciences and psychology). This generalistic point of view is still in vogue at the now independent Delft Design School in the year 2003. The Delft Design School based its design methodology courses on ideas from abroad – including influences from the Royal College of Art in London (which had firm links with the original design research movement), from the Hochschule für Gestaltung in Ulm (Germany) (which considered itself as the modern successor of the famous Bauhaus School) and partly from the Verein Deutscher Ingenieure (VDI, the German Society of Engineers) (de Wilde, 1997). The single most influencial design theorist for Delft was Bruce Archer from the UK. He had taught in Ulm and was at that time professor at the Royal College of Art in London. In 1971 he published a six-stage model of the product innovation process, and inside the stages he placed different steps. He did not start with a product idea or an ideation stage, but was one of the first scholars to introduce the idea that product design has to fit within the corporate strategy of a company. He also merged the ideas from the engineering worlds with the ideas from the commercial worlds:

Volume 12

Number 2

June 2003

his model is one of the first integrated product innovation models. The model is set out in Figure 2. This model is important for the overall development of modelling the product innovation process. It tried not only to balance both the engineering and commercial aspects of new product development, but also to link new product development with the strategic planning of the company. From this moment on, new product development was, in theory, not an isolated process for product designers or marketeers only, but could be handled as a corporate tool for growth and prosperity.

First Models From Delft In Delft, Norbert Roozenburg (the first student of Delft Design School, also its first graduate, in 1971) was becoming one of the leading design theorists. He analysed the different models of the product design process and came with the notion of a basic design cycle. First the designer has to analyse the design situation, then s/he has to synthesise possible solutions, followed by a simulation in which to judge the possible solutions against the original design situation, after which the designer is able to evaluate his or her (partial) design solution. This analyse-synthesise-simulate-evaluate sequence is still the kernel of the present Delft Design Method. This basic design cycle is present in every design stage and every design step (see Figure 3). Another basic idea is that every stage, phase or step has two different activities: first a divergent activity, followed by a convergent activity. The divergence is oriented towards getting as many alternatives as possible. The convergence is oriented towards choosing the best or most promising of those alternatives. Specific design tools are taught to improve the performance of the different activities. For instance, creativity techniques like brainstorming or morphological boxes were introduced to improve the quality of the divergent idea generation. Systematic programmes of requirements are developed to help the convergent idea screening. This alternating sequence of divergence and convergence is shown in Figure 4. Johannes Eekels joined the Delft Design School in 1976. In 1973 he had a book published on new business development, in which he links product innovation models with models of strategic planning. He came up with a model more or less similar to the Bruce Archer model. In 1980, Eekels was appointed

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

79

Figure 2. The Innovation Process According to Archer (1971)

© Blackwell Publishing Ltd, 2003

Volume 12

Number 2

June 2003

80

CREATIVITY AND INNOVATION MANAGEMENT

• policy formulation, which results in a product policy; • idea finding, which results in a new business idea; • strict development, in which three parallel processes take place; product designing resulting in a product design, marketing planning resulting in a marketing plan and production development resulting in a production plan; • realization, in which three main activities take place: production; distribution and sale; and (product)-use. These names of the four stages are used in the right-hand side of Figure 4. Policy formulation and idea finding are considered to be part of product planning; product planning and strict development form product development. Product development and realization form the total product innovation process (see Figure 5).

Modelling Activities Outside Delft Design School Figure 3. The Basic Design Cycle (Roozenburg, 1977)

Source: Roozenburg & Eekels, 1995

Figure 4. Divergence and Convergence in the Innovation Process as full professor in Product Policy and Product Planning. He and Norbert Roozenburg developed the first versions of the Delft Innovation Model. They divided the total product innovation process into four stages:

Volume 12

Number 2

June 2003

Parallel to these Delft developments, the German Society of Engineers (VDI) was investigating the starting point of new product development (Brankamp, 1976, 1978). As early as 1971, Klaus Brankamp had published a book on product planning in which he introduces the notion of so called ‘Suchfelder’ (‘search areas’). These are instruments to help to investigate the (competitive) environment of the company for new product/business opportunities. He had not yet made the connection to the corporate strategy. His original Suchfelder are more or less engineering-type tools to help the product idea generation process, mainly based on a kind of hierarchy of (product)-functions. Later these Suchfelder were developed in such a way that they became an excellent linking mechanism between the corporate strategic planning process and the product innovation process. Another parallel development took place in the Innovation Consulting Group, also located in Delft, but at TNO (The Netherlands Organisation for Applied Scientific Research). This commercial consulting agency was helping small and medium-sized firms with product innovation. After my graduation from Delft Design School in May 1976 I was asked to become their first innovation consultant. I was not educated by Eekels, Roozenburg or Brankamp, but I knew their publications and knew them personally. Trying to use their models in practice proved to be much more

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

81

Figure 5. The Product Innovation Process According to Roozenburg & Eekels (1995, original Dutch version 1991)

difficult than I had thought when I was studying them. I experienced a wide gap between the academic world, in which models of the product innovation process are developed, and the professional world in which people have to apply these models in real-life situations. Triggered by my practical experience with small and medium-sized enterprises (SME) I gradually developed an improved model of the product innovation process, in which a very easy-to-understand terminology was used. Within the academic design research community there is a tendency to invent new terms, without checking their usefulness, application or understanding in professional practice. People in industry want one simple, easy-to-understand and easy-to-use model of the (product) innovation process. In my improved model I tried to combine the ideas of Archer, Brankamp, Eekels and Roozenburg (and all their predecessors), but added some extra ease of use and people’s business to it. We, as SME-consultants, used colloquial language for describing the steps in the model. For instance we never used the term ‘strategy’ because SME people never use that term. Instead we were talking about finding the future direction for the company (in Dutch: koersbepalen), which was meaningful for them. In our (Innovation Consulting Group) search for easy-to-use models we had come across a four-stage model of the (product) innovation process developed inside Proctor

© Blackwell Publishing Ltd, 2003

& Gamble by Carlsson, Keane and Martin (1976). Their model is based on Kolb’s model of experiential learning (Kolb, 1976). According to David Kolb, experiential learning occurs in a sequence of four stages. First, people experience a concrete event, then they reflect on this experience and decide wether they like that event or not. If they like it, there is no need for learning; if they dislike it, there is a need for learning. So in the next stage they start thinking about new concepts they could apply to get better experiences. In stage four they start experimenting with that concept. By experimenting they will get a new concrete experience, on which they will be able to reflect in order to decide if this second experience was better than the first one. If so, they will apply the new concept in future situations; if not they have to design another concept and do a new experiment, and so on. The four learning stages are ‘concrete experience’ (CE), ‘reflective observation’ (RO), ‘abstract conceptualization’ (AC) and ‘active experimentation’ (AE) as shown in Figure 6. Kolb designed his model to describe the learning process of individuals. Carlsson, Keane and Martin experimently applied it to describe R&D processes within Proctor & Gamble. Using the premiss that the product innovation process (which in their opinion is equal to the R&D process) is similar to a learning process, because coming up with new products and services is the answer (learning) of a company reacting on its changing competitive environment.

Volume 12

Number 2

June 2003

82

CREATIVITY AND INNOVATION MANAGEMENT

Figure 6. Kolb’s Experiential Learning Model

Figure 7. The Step-by-Step Innovation Model Developed by the Innovation Consulting Group (1978)

In 1978 we introduced our four-stage product innovation model – the Step-by-step Innovation Model (Figure 7). The four stages are: • • • •

strategy formulation; design brief formulation; product development; product launch and use.

We were intrigued by the experiment of Carlsson, Keane and Martin, because they introduced for the first time insights from psychology and learning into the field of product innovation. As consultants trying to help small and medium-sized companies with their inno-

Volume 12

Number 2

June 2003

vation processes, this looked to be an interesting way. The relation between the Experiential Learning Model and the four-stage Step-byStep Innovation Model is as follows: • Stage 1 ‘strategy formulation’ equates to ‘reflective observation’; • Stage 2 ‘design brief formulation equates to ‘abstract conceptualization’; • Stage 3 ‘product development’ equates to ‘active experimentation’ and; • Stage 4 ‘product launch and use’ is the concrete experience the company will get.

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

In Figure 7; the box ‘present business’ represents the concrete experiences of the past, while the box ‘present and new business’ represents the new future concrete experiences after completing the full innovation, i.e. the learning cycle. Within our clients, SMEs, we were used to working with innovation teams as the main medium or tool for achieving change. The Proctor & Gamble experiments brought another element from Kolb’s theory into the open for helping to find the best team composition for innovation. This additional element of Kolb’s theory is the notion of different learning styles. Kolb had found out that not every person is equally good in all learning stages. Some people are better at concrete experience, others at abstract conceptualization, etc, so Kolb developed the Learning Style Inventory, which allows people to find out their learning style preferences through a selfassessment test. The prefered learning style is defined by a combination of two scores – one on the axis ‘active versus passive’ and the other one on the axis ‘concrete versus abstract’. There are four basic learning styles; in colloquial language they are dreamers (good at concrete experience + reflective observation), do-ers (active experimentation + concrete experience), testers (abstract conceptualization + active experimentation) and thinkers (reflective observation + abstract conceptualization) (see Figure 8). Carlsson, Keane and Martin used the Learning Style Inventory to measure the qualities of the innovation teams they were investigating. Innovation teams with all the four learning styles present are better than teams with an

83

unbalanced learning style profile. On a conceptual level this is easy to understand: if the innovation process equals the learning process then the innovation team has to go through all four stages to achieve an innovation. With an innovation team with dreamers, thinkers, testers and do-ers available that is easier to achieve than with an innovation team with only dreamers, or only do-ers. For our innovation consulting practice this proved to be an excellent instrument, not only to select team members, but also as an instrument and a language to better discuss problems inside less-well performing innovation teams. Finally Carlsson, Keane and Martin made some suggestions for using different management tools for overcoming difficulties in the four different stages. For instance if dreamers are not available then achieving divergence can be a problem for the team; by applying a technique like brainstroming the lack of natural divergence can be overcome. They also suggested that team leaders or consultants helping the team should always be one stage ahead of the team. So if the team is diverging, the leader/coach/consultant should already be thinking about how to achieve convergence in the next stage. These tips were very inspirational for our innovation consulting practice. In applying Kolb’s theory, we were able to integrate ideas from learning and change into the world of product innovation. Our experiences as management consultants helping SMEs to innovate showed that implementing innovations is very difficult. People inside small companies and probably also in larger

concrete

do-er

dreamer passive

active tester

thinker

abstract

Figure 8. Kolb’s Four Ideal Learning Styles

© Blackwell Publishing Ltd, 2003

Volume 12

Number 2

June 2003

84

CREATIVITY AND INNOVATION MANAGEMENT

organiszations share a common idea: innovation is fine, as long as nothing is going to change. By introducing learning and change as essential elements for stimulating innovation processes we were able to help people in companies better. Rational arguments do not work to reduce the effect of emotions, which are counter-productive. Dealing with those emotions does. The integrated innovation consulting method we developed at the Innovation Consulting Group therefore has five basic ingredients: • a simple four-stage step-by-step model of the innovation process (see Figure 7); • a coaching type of behaviour of the consultants (also known as process consulting, Schein, 1969, 1987); • a multi-disciplinary internal innovation team (including the CEO), composed by using the Learning Style Inventory for balancing the team; • the use of creativity techniques (for stimulating both divergent ànd convergent thinking); • the stimulation of external orientation (not only of competitors, but also new technologies, new (government) regulations, changes in the educational system, etc.).

Empirical Evidence In 1980 the Innovation Consulting Group was asked by the Dutch Ministry of Economic Affairs to participate in a large national programme to stimulate innovation in SMEs (The Project Industrial Innovation; Pii-project). They were asked to teach other consultants their integrated innovation consulting method and to do research on the effectiveness of this method. One hundred and fifty SMEs participated, and more than 60 management consultants from the professional consulting world were involved. The project was a great succes. After four years, 72 per cent of the participating SMEs had a concrete innnovation (at least a working prototype; so they were in stage 3 ‘product development’ of the model in Figure 7), 40 per cent of the innovations were already introduced on the market; stage 4 ‘product launch and use’) and 71 per cent had achieved a learning effect (the SME knows the methods and its pittfalls). In 14 per cent of the cases no success was achieved at all (Buijs, 1984, 1987, 1993). The empirical research also showed that the sequences and duration times of the sub-steps and stages of the product innovation process

Volume 12

Number 2

June 2003

were different for different groups of companies. In total, eight different patterns were discovered – in most cases just normal variations on the four-stage product innovation model. These results were congruent with earlier research carried out by Bob Cooper. He had discovered that seven different patterns of the innovation process could be distinguished in real new product development- life (Cooper, 1983). Two of our eight patterns are special, in the view of this paper, because they did not show any difference between the divergent and convergent activities either in stage 1 ‘strategy formulation’ and/or in stage 2 ‘design brief formulation. We labelled these patterns the parallel innovation processes: the backward and forward loopings are so frequent that no separated pattern is recognizable (see Figure 9). At the time of the research (Buijs, 1984, 1987) we did not pay much attention to this; it will be shown as an interesting result later in this paper. After my ‘return’ to the Delft Design School in 1986, to become the successor of Johannes Eekels, the four-stage step-by-step innovation model became the standard model for teaching about product innovation. We relabelled some terms (because design students are not innovation professionals yet), but essentially it is still the original model. We present the fourstage innovation model together with a way to handle, to act and play with the model. This handling and playing is strongly based on views on leading multi-disciplinary innovation teams in a facilitative leadership style (see Buijs & Valkenburg, 2000; Hohn, 2000). We now use different variants of the model for different purposes. To be more precise, we use different grades of detailing of the basic model for different purposes. For instance when we teach about innovation processes in general we use only the four-stage model (strategy formulation, design brief formulation, product development, product launch and use). When we are teaching about the relationship between the corporation, its brands and the kind of new products the company should develop, we use a very detailed description of the first stage of the innovation model. The strategy formulation stage is then sub-divided into five activities: (1) analysis of the present situation, which leads to the strategic situation of the company; (2) internal analysis; (3) external analysis; (4) search area generation; (5) search area evaluation; and (6) search area selection (Figure 10). Based on an analysis the strategic situation of the company is formulated. The strategic need for innovation is made explicit by esti-

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

voorbereiding 10 20 koersbepaling 1

koersbepaling 2

30 40 50 60 70

doelbepaling 1

doel-bepaling 2

80 90

voorbereiding 10 20

koersbepaling 1

30 40 koersbepaling 2 50 60 70 80

doelbepaling 1

doel-bepaling 2

90

1 Figure 9. The Two Parallel Innovation Processes (Above for Both the stages ‘strategy formulation’ and ‘design brief formulation’ (Buijs, 1987), below only parallel for the ‘design brief formulation’ stage (Buijs, 1984)

mating the future corporate situation when no strategic changes are made. If there is a gap between the situation when nothing changes and what is really wanted, the internal innovation team feels the urge to do something about it and the innovation process will be continued. To put it into Kolb’s terms, if the reflection of the company on its present situation leads to an unhappy feeling, there is a need to think (to learn) about new concepts for new business activities. During the internal analysis, the strategic strengths, the core competences are defined. In the external analysis, the competitive envrionment is analysed and the opportunities and threats are made explicit. During the search area generation, the

© Blackwell Publishing Ltd, 2003

85

SWOT-matrix (strengths, weakenesses, opportunities, threats) is established with an emphasis on the search areas (strategic ideas for innovation or potential new business opportunities). A search area is a combination of a strategic strength of the company with an external opportunity. For innovation we emphasise only the S- and O- combinations within the SWOT-matrix. Our use of search areas is a strategic application of the original, more technological-based idea from Brankamp. During search area evaluation, these strategic innovation ideas are checked with the outside world by interviewing experts, by looking at patents, by observing potential clients (users), etc. In search area selection a definite choice is made. The selected search areas form the starting point for the next stage: design brief formulation. For this stage and the two that follow, we have developed more detailed descriptions (see Buijs & Valkenburg, 2000). The complete detailed model of the product innovation process consists of 17 different steps in a given order (Figure 11). The Delft step-by-step product innovation model includes the following elements: • • • • • • • • • • • • • • • • • • • • • •

strategic situation of the company internal analysis evaluation external need analysis generating product ideas design brief market development product design market introduction distibution, promotion and sales evaluation external analysis generating search areas choosen search area internal analysis of bottlenecks evaluation product development developing manufacturing evaluation manufacturing product launch product in use

The 17 elements shown as the circles in the figure are activities or (sub)-processes; the other five elements relate to the strategic situation of the company, (choosen) search area, design brief, product design and product launch are the (intermediate) results of those processes and are shown as the squares in the figure. When we developed this model over the years we were fully aware of the existence of all different kinds of models of the product

Volume 12

Number 2

June 2003

86

CREATIVITY AND INNOVATION MANAGEMENT

Figure 10. The Strategy Formulation Stage

innovation process. Some were used for inspiration, for others we provided inspiration. For instance, in Asimow’s 1962 model, a stage is labelled ‘planning for retirement’. In our step ‘product in use’ we emphasise not only the use of the product but also its maintenance, its repair and its recycling. Andreasen and Hein have said that their model from 1985 was inspired by the Eekels/Roozenburg model in which three parallel stages (see Figure 5: production development, product designing and marketing planning) are recognised in the strict development stage. The Andreasen/Hein model introduces a three-band parallel thinking in all stages of the model. The three bands are market, product and manufacturing. They visualize their model in a horizontal mode. This led to our idea, in our vertical mode, to have the internal aspects on the left-hand side

Volume 12

Number 2

June 2003

of the model (like manufacturing), the innovation project elements in the middle (all the product elements) and all the external elements on the right-hand side (market). In an analysis of the reasons for differentiating product innovation processes Van der Zee (2003), a Delft Design School research student, made a comparison between 90 different innovation models from all over the world and from the early 1950s to the present day, including our model. He found out that in total, more than 1,248 different terms were used to descibe specific innovation activities. He converged these 1248 to a set of 54 innovation activities. Cooper had already indicated in his research in 1983 that different patterns of innovation processes exist. Some of the innovation activities are carried out in parallel, others sequential. According to Cooper the minimal NPD-process will have at least

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

strategic situation of the company

external analysis

internal analysis

generating search areas

evaluation

search area

internal analysis of bottlenecks

external need analysis

generating product ideas

design brief

product development

market development

product design

evaluation

market introduction

distribution, promotion & sales

manufacturing

two different sequential activities (product design & development and market launch). According to Van der Zee the maximum process will have 54 different activities. Cooper’s maximum process scored nine different activities. Our model is in the middle of these numbers. Van der Zee’s explanation is that activities like ‘appointing a project leader’ or ‘formation of the innovation team’ are necessary activities seen from a company’s point of view (and are included in his 54 activities), but are different activities to ‘concept testing’, ‘building a prototype’ or ‘patenting a new technology’. Our seventeen steps (activities) adequately cover the total product innovation process, but leave out all necessary elements for organising for innovation (like selecting a project leader, forming a team or communicating to suppliers).

Modelling Developments Inside the World of Creative Problem Solving

evaluation

developing manufacturing

87

product launch

evaluation

product use

Figure 11. The Complete Product Innovation Model: the Delft Step-by-Step Innovation Model

In the meantime, researchers in other domains were questioning the logical sequence of their models. Most practioners do not recognise separate steps in their daily practice. They do when they talk (logically) about it, but they seldom experience it that way. In several studies on design behaviour, experienced designers were asked to draw their own design process. Very few drew straightforward steps in a given order or sequence; most of them drew circles and loops from start to finish. It sometimes looks more like iteration than progress in the process. Some people consider designing as a special application of creative problem solving, and some interesting developments took place in this related field. Isaksen and Dorval reported on their new view on the CPS process at the Darmstadt (Germany) Innovation & Creativity Conference in 1993 (Isaksen & Dorval, 1993). They noted that most self-reported creative processes were spiral types, in contrast to the linear types from theory. They mentioned three developmental waves in their own thinking. First, the original development of a logical seven-stage CPS process model, including divergent and convergent activities, followed by a second period of expansion and strenghtening of the model. Finally, in the third wave they questioned the logical sequence of the stages and the visualization of the model. At the conference they suggested a new way of looking at the CPS process, the ecological view. There are four major activities: • understanding the problem;

© Blackwell Publishing Ltd, 2003

Volume 12

Number 2

June 2003

88

CREATIVITY AND INNOVATION MANAGEMENT

• generating ideas; • planning for action; • task appraisal.

searching for a renewed view on the innovation models.

Dependent on the fourth activity, task appraisal, one of the three other activities is tackled first, and depending how this task is executed the task will be ended, postponed or restarted again, or one of the other activities is dealt with (Figure 12). Their new visualisation still puzzles me. Another creativity scholar, Dan Couger, came up with the same idea in a slightly different way. He suggested a five-stage CPS model:

Circularity in Modelling the Innovation Process

• opportunity delineation and problem definition; • compiling relevant information; • generating ideas; • evaluating and prioritizing ideas; • developing implementation plan (Couger, 1995). He places the five stages on a circle and connects also all the stages with each other in a star form, also suggesting that all the elements/stages are neccessary but the sequence can be varied. You can more or less freely jump from one activity to another (Figure 13). These examples from the CPS world show that the logical sequence of the CPS model is questioned, and that more circular models seem to capture the real-life world better. These developments within an adjoining scientific domain were a strong inspiration for

After an extended literature search into the ‘fuzzy front end’ of innovation Sommeling (2002), a Delft Design School research student, discovers a paper from a group of innovation practitioners who reject the notion of a logical sequence in the innovation processes. These fourteen practitioners from eight multinational companies compared their own practices with the literature to find the best practices of the ‘fuzzy front end’ (FFE) of innovation (Koen et al., 2001). They compared the FFE with the new product and process development (NPPD). They conclude that the NPPD process is a logical, structured, disciplined and goal-directed activity with a project plan. In contrast to the FFE, which is experimental, chaotic, difficult to plan and unpredictable. Looking to their analysis, their view of the FFE is the same as the stages 1 (strategy formulation) and 2 (design brief formulation) of the Delft step-by-step innovation model, and their view of the NPPD process as the stages 3 (product development) and 4 (product launch and use). Koen and colleagues call their new model of the expanded FFE the new concept development model. It consists of three key parts:

Figure 12. The Ecological Creative Problem Solving model (Isaksen & Dorval, 1993)

Volume 12

Number 2

June 2003

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

89

Figure 13. Couger’s Creative Problem Solving Model (1995)

Figure 14. The New Concept Development Model (Koen, 2001) • a part with the five elements from FFE: idea generation, idea selection, opportunity analysis, opportunity identification, and concept and technology development; • a centre part, which they call the engine that drives the five FFE elements. The engine is fuelled by the leadership and culture of the organization; • an outer part which contains the influencing factors such as the competitive environment, the business strategy, organizational capabilities, etc. The linking pin between this NCD model and the NPPD process is the concept and technology development element. They do not see a

© Blackwell Publishing Ltd, 2003

standard sequence between the five FFE elements. According to the needs and the forces from the influencing factors one of the FFE elements will be the starting point – based on the result, one of the other elements is taken care of. It seems that ‘these needs and influencing factors’ act the same way as the task appraisal element of the Isaksen and Dorval CPS model (Figure 12). Koen c.s. found out that the best way of working with the NCD model is a chaotic procedure, without a logic sequence – a very ad-hoc organized activity (Figure 14). The new product and process development activities are, according to Koen et al., still organised in the traditional, more logical way.

Volume 12

Number 2

June 2003

90

CREATIVITY AND INNOVATION MANAGEMENT

Most Recent Developments of the Delft Product Innovation Model Inspired by these recent analyses of Koen and Sommeling, the two CPS process model remakes of Isaksen and Dorval, and Couger, and remembering the empirical results from the project industrial innovation (Pii) from 1984/1987 (the two patterns of the parallel innovation processes, see Figure 9), we came to the conclusion that we had to adapt the four-stage innovation model. In line with the Koen et al. view, the first two stages should be executed as in the new concept development model. And although Koen et al. suggest that the NPPD part of the innovation process is executed in a more orderly and logical sequence, we now think that the total look is much more like the latest CPS models (Figures 12 and 13). During the exit interviews in the Pii project in 1984 and the long-term effect interviews in 1987 with consultants and companies who reported the parallel processing of the steps in the innovation processes, I was very suspicous about their reporting. Originally I thought they were just lazy or unwilling to report correctly. How could you do an internal analysis and on the same moment do an external analysis? (for these steps see Figure 10). Re-reading the interviews with the consultants convinced me that these parallel actions are not only possible, but also very natural and useful. For instance, if the innovation team wanted to value one of the companies strengths to be of strategic importance (for becoming one of the backbones of the future innnovation), they have to compare this strength with those of the competitors. And by doing so (i.e. executing an external analysis) they discover certain essential characteristics of their competitors’ behaviour they did not have taken into account during the assessment of their own company (i.e. internal analysis), and then they restart their search for new strategic strengths of their own company. Here we see the effect of the pitfall of the way we visualize our models of the innovation process: without all the re-iterations. We all know about feed forward loops, feed back loops, side steps and doing steps over again. But as soon as they are reported we do not want to accept them. It was also very convincing to see that, because of the compulsory teamwork (under the Pii-framework), different team members were able to execute these steps in parallel. All the models of the product innovation process are described as if one individual is carrying out all actions. In reality it is teamwork, and the different steps will be executed in a parallel fashion. Team member A is generating

Volume 12

Number 2

June 2003

product ideas, while team member F is, on the same time, conducting interviews with potential customers, and team member C is looking for patents during his analysis of the state of the art of the technology. After meeting each other the next day they share their information and come up with an idea of how to proceed. So sequential steps from the logical innovation models can be executed in parallel in real-life innovation team situations. This means that all four stages are executed in the circular new concept development/CPS style (see Figure 15). In Figure 15 we relabelled and reshaped the ‘engine’ element from the original Koen new concept development model into a ‘heart’, because leadership, culture, emotion, motivation, risk-taking and passion are the true ingredients of innovative behaviour. Furthermore, the label ‘engine’ gives too much of an impression of a mechanistic worldview, while innovation is much more a people’s business. Depending on the qualities of the ‘heart’ and the ways the innovation team thinks about the company and its outside world, the innovation team selects out of the four stages one to begin with. Dependent on the results, it selects another one to proceed with. Of course, to do a prototype test in the product development stage, a prototype has to exist. That is the logical order. But the results of a prototype test can result in better insights in the needs of the potential clients, which can lead to new information for a strategy formulation stage. In the same way, actual use of the product in the product launch and use stage can lead to a

Product launch and use

Product development

Design brief formulation

Strategy formulation

Figure 15. First Circular Version of the Delft Innovation Model (February 2002)

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

reformulation of the design brief, which can lead to the generation of a second version of the original product innovation. Or it could lead to an improved promotion campaign for the present product innovation. Playing with both the logical sequences of the innovation process steps and the parallel usage of every (intermediate) result from the process as an opportunity to dive back and forth in one of the four main stages is an important task for future innovation teams. Due to the fact that different innovation team members are working in parallel, and the duration times of the different tasks are varied, some of the results will come earlier than the logical order of the process. And these results will of course influence the other tasks which are still being executed. Thus logical order and chronological order will be separated, which will give the team members a feeling of a chaotic product innovation process. Inspired by this circular version of the fourstage innovation model we went back to our linear model of Figure 11. Product innovation

91

processes are intended to help companies to design and introduce new products, which customers are willing to buy, and happy to use. Therefore, in product use the innovation process ends, but at same time forms the starting point of a new product innovation process. We therefore curved our linear model to come up with a new circular model which shows the product innovation process on the detail level (see Figure 16). Visualizing the process as a circular model suggests that there is neither beginning nor end, which is true in the sense that after introducing a new product on the market as a result of a product innovation process, the succesful use of this product will lead to reactions of competitors, for instance by introducing their new, and better performing products. This in turn will cause the original innovating company to start the next new product innovation process to regain its competitive advantage. Usually, as shown in Figure 11, strategy formulation is considered to be stage 1 – and seen

Figure 16. The Detailed Circular Model of the Product Innovation Process (March 2003)

© Blackwell Publishing Ltd, 2003

Volume 12

Number 2

June 2003

92

CREATIVITY AND INNOVATION MANAGEMENT

from the innovating company’s viewpoint that is correct. If we look at the overall objective of an product innovation process, designing, developing and introducing a new product are only means to have a commercial succesful and easy-to-use product in the market place, and which users are happy with it. Therefore, in Figure 16, the product-use step is placed at the top, and strategy formulation forms the next stage, placed to the right, followed by the design brief formulation, product development, and product launch and use stages. We then finish where we were started. To make this change from the linear version in Figure 11 to the circular version in Figure 16, we added some extra steps into the product-use stage to make the full circle. Starting from this use of the product the 26 innovation elements in the detailed model are the following: • • • • • • • • • • • • • • • • • • • • • • • • • • • •

product use evaluation (commercial) of product use evaluation (technological) of the product strategic product position evaluation strategic situation of the company external analysis internal analysis generating search areas evaluation choosen search area external need analysis internal analysis of bottlenecks generating product ideas evaluation design brief product development market development developing manufacturing product design evaluation market introduction manufacturing distibution, promotion and sales product launch evaluation

After this last evaluation the new product is in use, which will lead eventually to the start of a next product innovation cycle. The model is showing that elements on the inside have to do with internal aspects of the company (like manufacturing), and elements on the outside have to do with the commercial and competitive environment of the company (like sales or market research). Elements on the central circle line are the core product innovation activities and results.

Volume 12

Number 2

June 2003

Conclusions Looking back on this more than thirty-year history of the development of the Delft product innovation model it seems that at first we tried to capture the chaotic innovation reality in linear logical models, but that we are now accepting this chaos in our models. What we try to do at the Delft Design School is to teach the students that there are different types of models of the product innovation process. Dependent on your own situation, you have to choose for a specific model and or to make a choice for more or less detailing. We teach our students the two versions of our innovation model as two sides of the same coin. One side is the abstract four-stage model, where you can jump from one stage to another without any given sequence (Figure 15). The other side is the concrete detailed model, with a very logical order, including parallel parts followed by sequential parts (Figure 16). The problem with these much more openended, circular and chaotic types of models is that during education, students experience less help than with the more traditional strict, linear and logical models. For instance, while you are executing the first step from the strategy formulation stage to come up with a description of the strategic situation of the company, you have to analyse the commercial and competitive results of the product in use, which in the model is a stage earlier. So teachers and students complain that the model has the wrong sequence. Seen from the logic of the overall product innovation process the sequence is right, but seen from the activities a student (or the company’s internal innovation team) has to perform in chronological order, it looks like a wrong sequence. But models of the product innvation process should be logical and not chronological, because in the latter case the different duration times of the different stages and steps will ruin all the insights that the model is built for. Design educators have to choose their own models depending on the teaching situation (freshmen, final-year students or practitioers). The same is true for innovation researchers: depending on their goals and objectives they have to choose a specific model, a specific version of the model and or a specific level of detailing. It is not (yet) possible to give a concrete advice for which model or which level of detailing is neccesary for which goal and under which circumstances. Personal preferences will also play an important role. On occasion, it may be neccesary to use multiple modelling to get a good grip on the situation.

© Blackwell Publishing Ltd, 2003

MODELLING PRODUCT INNOVATION PROCESSES

There is probably no one best way to modelling the product innovation process. Doing your own thing, looking back on what you have achieved and deciding how to continue, to change, to go back or to find a complete new way of behaving is much more the likely way that innovators behave. This notion of reflective practice was developed by Donald Schön (1983). In our most recent empirical research on design activities in product innovation this locally driven behaviour seems to describe design behaviour in such a way that practitioners get the feeling that it is really describing what they are doing (Dorst, 1997; Valkenburg, 2000). Furthermore, having a portfolio of models and detailing available will help innovation students and practitioners to better deal with this reflective behaviour. It will provide multiple frames for reflection, and reflection is essential for learning, and learning is essential for (product) innovation.

References Andreasen, M.M. and Hein, L. (1985) Integrated Product Development. IFS (Publications) Ltd/ Springer Verlag, Bedford, UK/Berlin, Heidelberg. Archer, B.L. (1971) Technological innovation- a methodology. Inforlink, London. Asimow, M. (1962) Introduction to design. Prentice Hall, Englewood Cliffs, NJ. Brankamp, K. (1971) Planung und Entwicklung neuer Produkte. De Gruyter, Berlin. Brankamp, K. (ed.) (1976) Systematische produktplanung- ein Mittel zur Unternehmungssicherung. VDI-Verlag, Düsseldorf. Brankamp, K. (ed.) (1978) Arbeitshilfen zur systematischen produktplanung. VDI-Verlag, Düsseldorf. Buijs, J.A. (1984) Innovatie en interventie. Kluwer, Deventer. Buijs, J.A. (1987) Innovatie en interventie, tweede uitgebreide druk. Kluwer, Deventer. Buijs, J.A. (1993) Creativity and innovation in the Netherlands: Project Industrial innovation and its implications. In: Isaksen, S.C., Murdock, M.C., Firestien, R.L. and D. Treffinger, Nurturing and developing creativity: the emergence of a discipline. Abblex Publishing Company, Norwood. Buijs, J.A. and Valkenburg, R. (2000) Integrale productontwikkeling, tweede uitgebreide druk. Lemma, Utrecht. Carlsson, B., Keane, P. and Martin, J.B. (1976) R & D organizations as learning systems, Sloan Management Review, Spring. Cooper, R.G. (1983) The new product process: an empirically-based classificationscheme. R & D management, 13, january, no. 1, pp 1–13. Couger, J.D. (1995) Creative problem solving and opportunity finding. Boyd & Fraser Publishing Company, Ferncroft Village. Dorst, C.H. (1997) Describing design, a comparison of paradigms, Delft, Doctoral Thesis, Delft University of Technology.

© Blackwell Publishing Ltd, 2003

93

Eekels, J. (1973) Industriële doelontwikkeling. Van Gorcum, Assen. Gregory, S.A. (ed.) (1966) The Design Method. Butterworths, London. Hohn, H. (2000) Playing, leadership and team development in innovative teams; a reflection on theory confronted with the perspectives of experienced leaders. Eburon, Delft. Isaksen, S.G. and Dorval, K.B. (1993) Expanding views of CPS: a synergy methodology. In: Geschka, H., Moger, S. and Rickards, T. (eds), Creativity and innovation, the power of synergy, Darmstadt, Proceedings of the 4th European Conference on Creativity and Innovation. Jones, J.C. (1970) Design Methods, seeds of human nature. Wiley Interscience, London. Koen, P., Ajamian, G., Burkhart, R., Clamen, A., Davidson, J., D’Amore, R., Elkins, C., Herald, K., Incorvia, M., Johnson, A., Karol, R., Seibert, R., Slajkov, A. and Wagner, K. (2001) Providing clarity and a common language to the fuzzy front. Industrial Research Institute, Washington. Kolb, D.A. (1976) Management and the learning process, California Management Review, XVIII, 3. Roozenburg, N.F.M. (1977) De basis ontwerpcyclus. In Collegedictaat Ontwerpmethodologie, TUD-IO. Roozenburg, N.F.M. and Eekels, J. (1995) Product Design: fundamentals and methods. John Wiley & Sons, Chichester. Saren, M.A. (1984) A classification and review of models of the intra-firm innovationprocess. R&D Management, 14, 1, 11–24. Schein, E.H. (1969) Process consultation, its role in organization development. Addison Wesley, Reading, MA. Schein, E.H. (1987) Process consultation, volume II, lessons for managers and consultants. Addison Wesley, Reading MA. Schön, D.A. (1983) The reflective practitioner. Basic Books, New York. Sommeling, A.T. (2002) Samen nieuwe combinaties creëren in het Fuzzy Front End, Delft University of Technology, School of Industrial Design Engineering, unpublished MSc Thesis. Valkenburg, A.C. (2000) The Refelective Practice in product design teams. Delft, Doctoral Thesis, Delft University of Technology. Wilde, H. de (1997) Passie voor productontwikkeling, Delft, Doctoral Thesis. Zee, A. E. Van der (2003) Proceskeuze in de productontwikkeling, Delft University of Technology, School of Industrial Design Engineering, unpublished MSc Thesis.

Jan Buijs is professor in Product Innovation and Creativity at the School of Industrial Design Engineering (Delft Design School) of the Delft University of Technology. He is chairman of the European Association for Creativity and Innovation (EACI), a notfor-profit organization which is bridging the gap between academics and professionals in the field of innovation and creativity.

Volume 12

Number 2

June 2003

94

CREATIVITY AND INNOVATION MANAGEMENT

Radical Product Innovations in SMEs: The Dominance of Entrepreneurial Orientation Helen Salavou and Spyros Lioukas The present study focuses on strategic factors underlying the adoption of radical product innovations in SMEs. It investigates whether market focus, technological posture and entrepreneurial orientation lead to the adoption of more radical product innovations. The study provides some new evidence on this issue drawing upon data collected from Greek SMEs in the food and beverages industry. More specifically, a logistic regression model is applied to analyse the choice between radical as against incremental product innovations. Empirical results appear to support the claim that it is mainly entrepreneurial orientation that favours the choice of radical product innovations. This suggests that in SMEs the notion of entrepreneurial-push outweighs both market-pull and technology-push arguments. The findings are discussed in the context of Greece, taking into account the specific conditions prevailing. Apart from providing some new evidence in the important area of SMEs, they have also important implications for managers and policy-makers. In addition, they encourage further theoretical and empirical investigation.

Introduction

T

he present paper focuses on strategic orientation of small and medium-sized enterprises (SMEs) and its effect on product innovations. In particular, it explores the importance of strategic aspects related to market orientation, technology policy and entrepreneurial orientation in adopting radical as against incremental product innovations. There is prior empirical evidence that these strategic factors are important contributors to innovation at large. This is mainly based on investigations built on perspectives, such as ‘market-pull’, ‘technology-push’ and ‘strategic choice’ respectively, as well as on related extensions. It appears that all these factors are, in general, drivers to innovation. However, almost all previous evidence comes from large enterprises. We do not know if similar effects hold for SMEs. More specifically, existing knowledge is not sufficient to verify which of the strategic variables related to market orientation, technology policy and entrepreneurial orientation make SMEs select radical product innovations. Exploring such relationships

Volume 12

Number 2

June 2003

in the context of SMEs will enrich our understanding of the relative importance of these strategic forces in explaining innovative performance. SMEs are generally considered to have behavioural advantages that may justify their significant share in innovation (Dutta and Evrard, 1999). Despite their disadvantages, most often attributed to resource constraints (Freel, 2000), SMEs are somehow forced to utilize product innovations as a means of competitive strategy to a higher degree than large firms (Fritz, 1989; Sweeney, 1983). Product innovations, however, vary in a range from incremental improvements to radical changes. Radical product innovations may take a prominent place in the case of SMEs in their efforts to compete with large firms. It would be interesting, therefore, to focus our analysis on the above mentioned strategic factors as potential determinants of radical product innovation adoptions. The present study draws upon data from food and beverages, a traditional sector of the Greek economy that includes many domestic firms. This industrial sector of Greece, like the

© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.

PRODUCT INNOVATION IN SMES

respective sectors of most economies in the EU, is undergoing rapid transformation. The ‘rise’ of new technologies, new ways to add customer value and localization of globalized trends (often called ‘glocalization’) have created a state of flux. The results of this new competitive landscape are often mirrored in never-ending introductions of product innovations. Greece presents an interesting case given the dominance of SMEs (more than 95% of firms) and the processes of integration in the euro area which are under way. The traditional behaviour of semi-protected family-type SMEs is challenged as they confront both European integration and global competition. As such, competitiveness through radical product innovations is an important concern for both SMEs and policy-makers, especially when it relates to traditional sectors, such as food and beverages, with a vital contribution to economic growth. Survey data were collected by a structured questionnaire through personal interviews with top managers in 69 randomlyselected manufacturing SMEs. A logistic regression model was applied to the sample in order to empirically test the effects of market orientation, technology policy and entrepreneurial orientation on the decision on whether or not to adopt radical product innovations. In order to assist in the formulation of hypotheses and the interpretation of results, additional evidence was collected related to the specific conditions prevailing in the Greek business environment in which our sampled firms operate. To this purpose, interviews were conducted with experts in a sectorspecific support organization.1 The opinions elicited provide further qualitative information concerning Greek SMEs’ strategic orientation and its impact on product-related innovative activity. The present study would also contribute to theory as the strategic determinants considered are related to the traditional debate about the market-pull vs. technology-push question. On top of that, the present analysis helps to broaden this debate by adding the notion of ‘entrepreneurial push’, which is often assumed to be a driving force of strategic importance, particularly in SMEs. The article is organized as follows. First, the groundwork is laid out by presenting the the1 ETAT S.A. is sector-specific Industrial Research and Technological Development company of the General Secreteriat of Research and Technology (Ministry of Development) for the food and beverages industry.

© Blackwell Publishing Ltd, 2003

95

oretical model and hypotheses. This is followed by a brief presentation of the research methodology and the survey. Research findings and their implications are discussed next, and a concluding section summarizes the results.

Conceptual Model and Hypotheses Innovation researchers have, to a large extent, debated whether product innovations are driven by market demand or by technological advances. The market-pull argument suggests that firms innovate based on market needs, whereas the technology-push argument claims that change in technology is the primary driver of product innovation (Chidamber and Kon, 1994). Extending the research into the underlying driving forces behind product innovations in SMEs would be very useful. Given the strategic importance of innovative new products in SMEs, an interesting investigation would be to consider whether market orientation and technology policy influence the adoption of radical product innovations. This investigation could also embody the potential effect of entrepreneurial orientation. A rationale for including entrepreneurial orientation would be found by extending the ‘strategic choice’ perspective (Bourgeois, 1984; Child, 1972; March and Simon, 1958) which emphasizes the role of ‘upper echelons’ or ‘top managers’ or ‘strategic leadership’ in determining organizational outcomes (Guth and Ginsberg, 1990; Hambrick and Mason, 1984; Miller, Droge and Toulouse, 1988). Entrepreneurial style would be a key determinant of radical product innovations, especially in SMEs, where the top management (including the CEO) appears to be most influential. The conceptual model adopted (see Figure 1) assumes a direct linkage between market orientation, technology policy and entrepreneurial orientation on product innovativeness. The argument we intend to explore is whether market-focused, technology-prone and entrepreneurially-oriented SMEs are more inclined to adopt radical as against incremental product innovations. We define product innovativeness along with the development of the hypothesised relationships in the model.

Product Innovativeness The product innovativeness concept is an important classifier of new products reflecting a choice, either explicit or implicit, of product strategies (Song and Montoya-Weiss, 1998). The significance of such a classification has

Volume 12

Number 2

June 2003

96

CREATIVITY AND INNOVATION MANAGEMENT

Firm

Strategy-related Characteristics: Market Orientation Technology Policy Entrepreneurial Orientation

Product Innovativeness: Radical/Incremental Product Innovations

Figure 1. Conceptual Model

been recently addressed (Danneels and Kleinschmidt, 2001). Previously, most studies generally did not make any distinction between different levels of innovativeness in new products (Balachandra and Friar, 1997; Blythe, 1999; De Brentani, 2001). Instead, they considered as new product anything that was introduced into the market by the firm, regardless of the extent and type of newness. Product innovativeness, though, reflects the level of newness in product innovations that can vary broadly ranging over a wide spectrum (Balachandra and Friar, 1997). Such variations most often reflect the degree of changes related to product innovations (Hull, Hage and Azumi, 1985). Our analysis considers a dichotomy of this spectrum, namely radical and incremental product innovations. These two categories are mutually exclusive and reflect product newness relative to the firm, as perceived by managers (Song and Montoya-Weiss, 1998). Therefore, the higher the likelihood of adopting radical innovations, the lesser the likelihood for incremental innovations. Product innovativeness, the dependent variable, is, thus, a binary one expressing a choice of high as against low degree of newness in new products. Under the assumed conceptual model, our aim is to isolate and test the impact of market orientation, technology policy and entrepreneurial orientation on the choice among the two levels of product innovations, i.e. radical versus incremental. This study is in line with certain recent research studies positing that there are major differences in factors depending on whether the innovation is radical or incremental (Balachandra and Friar, 1997). In addition, it responds to the need for crosslevel research on organizational innovation (Atuahene-Gima and Ko, 2001; Drazin and Schoonhoven, 1996) by investigating the impact of characteristics at the firm level on innovativeness at the product level. Next, the components of the conceptual model are discussed in more detail. Given the lack of related studies on SMEs, hypotheses

Volume 12

Number 2

June 2003

are developed taking into consideration (a) the extant research which draws mainly from large firms, (b) evidence, where available, referring to the specific national context and related debate, and (c) the viewpoints of industry experts who understand conditions and the factors at work in the specific environment of Greek SMEs in the food and beverages sector. However, the hypothesized relationships will be tested statistically to confirm if, and to what extent, they apply to the particular case of Greek SMEs.

Effects of Market Orientation Market orientation has attracted enhanced interest in both management and marketing literature. Nonetheless, little attention has been paid to the relationship between market orientation and innovation, as the market orientation literature has only recently begun to acknowledge the role of innovation in the context of market orientation (Han, Kim and Srivastava, 1998). Researchers who dealt with this issue tried to explore the impact of market orientation on innovation, either direct or indirect. Jaworski and Kohli (1996), for example, regard innovation as an outcome of market orientation stating that innovation has been inappropriately absent in models of market orientation. Nonetheless, Slater and Narver (1994) suggest that the market-driven business is well positioned to anticipate the developing needs of customers and respond to them through the addition of innovative products. In a similar vein, Gatignon and Xuereb (1997) provide support for a positive impact of the firm’s market orientation dimensions on the way product innovations are marketed and acquire strong relative advantages. Overall, however, results of those studies dealing with the impact of market orientation on different types of product innovations are rather contradictory, failing to yield consistent conclusions. Davis (1993), for example, found a positive effect of market orientation on radical but not on incremental product innovations. On the contrary, Atuahene-Gima

© Blackwell Publishing Ltd, 2003

PRODUCT INNOVATION IN SMES

(1995) concluded that the impact of market orientation was greater on incremental rather than on radical product innovations. Similarly, Christensen and Bower (1996) suggest that market orientation stifles the development of original new products, instead encouraging the development of product modifications. A more recent study, on the other hand, found positive effect of market orientation on both product innovations, i.e. radical (‘new to the market’) and incremental (imitations) (Sandvik, Grønhaug and Øgaard, 2000). Because researchers’ views are not uniform, this relationship is explored here in the light of empirical data. However, the prevailing view based mainly on evidence from large firms in the relevant literature is that market-oriented firms may be more innovative, because they are more responsive to rapidly evolving customer needs. To our knowledge, very few studies have explored the same argument in the context of SMEs. So, in the absence of such evidence it is useful to resort to qualitative evidence from the Greek context. The industry experts believe that the adoption of the market orientation concept is necessary for SMEs, given the strong competition within the food and beverages sector. When interviewed, they argued in favour of market orientation as an enabling factor of product innovations in SMEs. In their view, persistence in satisfying their customers’ changing needs and wants is a source of radical product innovations which is nurtured by SMEs’ inherent advantages of flexibility, adaptability and closeness to their customers. Taking all the above into account the following hypothesis can be advanced: H1: Market orientation is likely to lead to radical as against incremental product innovations.

Effects of Technology Policy Many empirical studies have interpreted their results pinpointing the impact of technology policy on firms’ innovative behaviour. To this direction, Ettlie (1983) asserts that technology policy becomes a central variable in the emerging modification of existing innovation theories. Firms’ strategic orientation including technology policy has often been investigated as an innovation determinant (Wilson, Ramamurthy and Nystrom, 1999). Technology policy undoubtedly affects the development of new ideas, new processes and new products. Ettlie and Bridges (1983), for example, suggest that technology policy reflects the innovative attitude of an organization and its commitment to innovation. It involves such things as recruiting technical personnel, com-

© Blackwell Publishing Ltd, 2003

97

mitting funds to new technology development and building or maintaining a tradition of being at the forefront of a technological area in a particular industry. Innovative firms may be proactive in acquiring new production technologies, and employ such technologies for developing new products (Cooper, 1984, 1994). To this direction, Ettlie, Bridges and O’Keefe (1984) assert that an aggressive technology policy is required for supporting the adoption of radical innovations. In a similar vein, Ettlie and Rubenstein (1987) suggest that changes in technology are more likely to lead to more radical product innovations. In an empirical study, Gatignon and Xuereb (1997) investigated the way that technological orientation affects product innovations, and more particular their characteristics. They further supported the view that the more technology-oriented the firms are, the more radical their product innovations. The aforementioned discussion supports the argument that technology policy significantly affects product innovations, and in particular the more aggressive it is, the more radical the product innovations marketed by firms. However, it remains unclear whether this argument holds true for all firms, irrespective of size. Additional evidence related to the specific national context and related debate is partially in line with this viewpoint. On the one hand, industry experts believe that resource constraints may direct SMEs’ efforts towards innovations that require less capital investment and a lesser reliance on qualified staff. As such, they argue that SMEs often introduce innovations based not so much on core production technology which requires high R&D budgets, but on other aspects (e.g. delivery tuning, value-added services, etc.) along the value chain. On the other hand, European Community technology and innovation policy initiatives, in conjunction with national policies in support of SMEs, are intensively working on the creation of a fertile breeding ground for innovative ways of doing business.2 The business environment, aided by the Community Support Framework programmes, is making aggressive technology policies easier for SMEs. Therefore, despite reservations expressed by the industry experts, we think that the dominant argument prevailing in large enterprises would also be transferred in the context of SMEs. Hence, we hypothesise that: 2

Recent evidence indicates that among the member states Greece showed one of the highest growth rates in R&D investments (European Commission Research, 2002).

Volume 12

Number 2

June 2003

98

CREATIVITY AND INNOVATION MANAGEMENT

H2: Aggressive technology policy is likely to lead to radical as against incremental product innovations.

Effects of Entrepreneurial Orientation Previous research studies suggest that entrepreneurial orientation drives innovative activity (Covin and Slevin, 1991; Kitchell, 1995; Russell and Russell, 1992). Moreover, innovative firms exhibit a greater willingness for risktaking and proactive market leadership (Khan and Manopichetwattana, 1989). In a similar conceptualization, being reactive is not sufficient to compete successfully in a dynamic world of aggressive innovators, and most innovative firms are more willing to look hard at risky opportunities and confront them (Souder, 1987). However, few research studies have empirically explored the influence of entrepreneurial orientation on product innovation. Among them, Miller and Friesen (1982) argue that entrepreneurial firms innovate boldly and regularly while taking considerable risks in their product-market strategies. In a similar vein, Miller, Kets de Vries and Toulouse (1982) suggest that substantial product innovations require greater amount of risk-taking and proactiveness from companies. Saleh and Wang (1993) support the view that more innovative compared to less innovative companies would take more risks and adopt a proactive strategy (leading rather than following competitors) in anticipating the need for change and new opportunities. Whilst the preceding literature suggests that entrepreneurial orientation promotes innovative activity by affecting the introduction and implementation of product innovations within firms, there is no explicit empirical evidence concerning the influence of this organizational phenomenon on product innovativeness. It is interesting, therefore, to explore whether enhanced entrepreneurial orientation demonstrated by greater amount of proactiveness, and risk-taking assumed by top management influence the level of newness in product innovations marketed by SMEs. Besides, top-management style is frequently playing an important role in determining strategic directions while shaping ‘outcomes’ (Hambrick, 1989). This is particularly true for Greek SMEs characterized by concentration of power and control in the hands of top management, which is most often identical with family ownership (Lioukas and Makridakis, 1996). The industry experts interviewed also believe that entrepreneurial attitude is salient in SMEs, and accounts for initiatives related to radical product innovations. Recent research also indicates that during 1980s and 1990s significant changes occurred

Volume 12

Number 2

June 2003

in the Greek management practices, especially as a new generation of well-educated ownersmanagers takes responsibility (Bourantas and Papadakis, 1996; Makridakis et al., 1997; Spanos, Prastacos and Papadakis, 2001). The tradition of entrepreneurship in the food and beverages sector, coupled with enhanced emphasis on creativity exhibited by new top managers (Makridakis, Papagiannakis and Calogirou, 1996) is likely to amplify the capabilities needed for implementing aggressive innovative behaviour. Given the above considerations, we would hypothesise that in the case of SMEs: H3: Enhanced entrepreneurial orientation is likely to lead to radical as against incremental product innovations. Control Variables In considering control variables for SMEs, experience and size come as obvious candidates for inclusion. The former can be expressed through firm age. Several studies support a negative relationship between firm age and innovative behaviour. It is argued that the older the firm, the more bureaucratic and the less receptive it is to innovation (Hurley and Hult, 1998). On the contrary, recently established firms are more inclined to innovation, probably due to the necessity of conquering space in the market (Garvin, 1983). Based on the above, the relationships between product innovativeness and firm-specific strategic characteristics are re-examined after controlling for age. Size was finally excluded from the analysis, because it was highly correlated with firm age.

Research Methodology Sample and Data Collection The data pertain to one of the most dynamic industries of the Greek economy, namely food and beverages, in which the vast majority are SMEs. This choice is dictated by (1) the importance of this sector for the Greek economy in terms of manufacturing employment and GDP share (23% and 25% respectively), and (2) the opportunity it provides for studying the proclivity of firms within a traditional sector towards the adoption of radical product innovations in the new competitive scene. Following the EU definition of SMEs,3 the relevant population is defined as all inde3

Published by the Official Newspaper of the European Communities on 30 April 1996.

© Blackwell Publishing Ltd, 2003

PRODUCT INNOVATION IN SMES

pendent firms with less than 250 employees and less than € million annual turnover. According to these criteria, a population of 745 SMEs was identified using ICAP database (Gallup’s subsidiary). Subsequently, 69 firms which have introduced at least one new product (or new product category) during the last three years (1995–1997) were selected by random sampling.4 Data were collected by a structured questionnaire through personal interviews with the top management (42% managing directors, 26% marketing & sales managers, 17% financial managers and 15% others). All interviews were carried out personally by the authors during 1998. The estimation data set included 66 firms that provided complete data for all necessary variables. With respect to the type of new products (or new product categories) introduced by the sampled firms, 28% were described as incremental product innovations (i.e. improvements and modifications of existing products) while 72% as radical product innovations (i.e. new product lines, new-to-the-market products).

Additional Qualitative Information In order to help the hypothesis development and explain the results of this study, qualitative data were collected through semistructured interviews with experts. More specifically, experts from the food and beverages sector support organization (ETAT S.A.) were asked to express their opinions concerning Greek SMEs’ strategic posture in terms of market orientation, technology policy and entrepreneurial orientation as well as its impact on product-related innovative activity. Although these viewpoints are confined in the context of Greece, they nevertheless are useful in understanding the conditions prevailing and the particularities of the effects found regarding the food and beverages industry. Thus, opinions expressed by experts are supplementary and shed additional light on the extent to which specific dimensions of strategic importance affect (or not) the adoption of radical product innovations. This evidence is embodied both in hypotheses formulation and the discussion following the empirical findings.

4

Given that such a criterion required phone calls to 745 firms and the related cost was extremely high, a sample was randomly selected from the whole population. Firms not meeting this criterion were replaced by others that were not selected in the first run.

© Blackwell Publishing Ltd, 2003

99

Variables Measurement Product innovativeness, our dependent variable, has been operationalized by many previous empirical studies from an internal, firm perspective in terms of product’s newness relative to the firm, as perceived by managers (Booz, Allen and Hamilton, 1982; Brouwer and Kleinknecht, 1996; Cooper, 1979; Song and Montoya-Weiss, 1998). Likewise, product innovativeness is measured in the present study using a dichotomous categorisation as defined from the firm’s perspective, i.e. based on managers’ perceptions of newness (Song and Montoya-Weiss, 1998). In particular, respondents were asked first to identify the most important, in terms of sales turnover, new product (or new product category) introduced by their firms in the last three years (1995–1997) and then to classify it according to its newness (i.e. radical or incremental). Thus, in the binary construct, 1 represents radical product innovations (i.e. new product lines, new-to-the-market products) while 0 refers to incremental product innovations (i.e. improvements and modifications of existing products). Product innovativeness is, therefore, a perceptual measure reflecting the degree of similarity of the new product with those already marketed by the firm. Market orientation is measured in this study by a version of the scale developed by Ruekert (1992). This version is similar to the instruments proposed by Kohli and Jaworski (1990), Narver and Slater (1990) and Shapiro (1988). This modified instrument is defined as a set of specific behaviours and activities reflected by the extent to which a firm understands and responds to customer needs. More specifically, market orientation captures in this study four dimensions, i.e. customer information, customer responsiveness, market-driven strategy formulation, market-driven pricing policy (see Appendix A). Technology policy is tapping the degree of the firm’s long-term commitment to activities involving technological issues. This variable is measured in this study by a scale proposed by Ettlie (1983), based on the earlier work of Ettlie and Bridges (1982) (see Appendix A). Entrepreneurial orientation is measured by a multi-item, semantic differential scale suggested by Naman and Slevin (1993) reflecting top-management’s behaviour in taking strategic decisions and operating management philosophies. In the present study, this variable captures two dimensions, i.e. proactiveness and risk-taking. More specifically, proactiveness is conceived by top management’s willingness to carry out actions before competitors, while risk-taking is compre-

Volume 12

Number 2

June 2003

100

CREATIVITY AND INNOVATION MANAGEMENT

hended by top management’s preference to undertake activities of (high/low) risk (see Appendix A). Finally, firm age, the control variable, is measured by the number of years passed since the firm’s foundation (Da Rocha, Christensen and Paim, 1990; Heunks, 1998).

Estimation and Results Appendix A presents inter-item reliability coefficients of the scales adopted which are acceptable according to standards suggested by Van de Ven and Ferry (1980). All multi-item scales measuring independent strategy-related variables were factor-analysed (see Appendix A) in order to assess their ‘factorial validity’ which is a form of construct validity (Allen

and Yen, 1979). Then, averages of items pertaining to factors extracted were used to form independent variables for further statistical analysis. Tables 1 and 2 report descriptive statistics of all the independent variables, together with their correlation coefficients. As mentioned above, product innovativeness, the dependent variable, is dichotomous. In a way, it expresses the likelihood that a new product will suggest radical as against incremental innovation. A logistic regression model was therefore employed to deal explicitly with that type of dependent variable (Kleinbaum, 1994). So, the estimated model takes the following form: Prob (radical product innovation) = 1 (1 + e - Z ) ,

Table 1. Descriptive Statistics of Independent Variables Variable

Min.

Max.

Mean

SD

Market Orientation: Customer Information Customer Responsiveness Market-driven Strategy Formulation Market-driven Pricing Policy Technology Policy

3.33 4.50 3.67 2.00 2.00

7.00 7.00 7.00 7.00 7.00

5.78 6.34 5.48 5.40 5.25

0.79 0.65 0.83 1.05 1.17

Entrepreneurial Orientation: Proactiveness Risk-taking Firm Age

2.00 1.50 1.00

6.60 6.25 136.00

4.82 3.93 27.25

1.07 1.12 23.77

Table 2. Spearman’s rho Correlations among Independent Variables Variable

1. 2. 3. 4. 5. 6. 7. 8.

Variable

Customer Information Customer Responsiveness Market-driven Strategy Formulation Market-driven Pricing Policy Technology Policy Proactiveness Risk-taking Firm Age

1.

2.

3.

4.

5.

6.

7.

8.

1.00 0.34** 0.41** 0.32** 0.19 0.07 0.15 -0.11

1.00 0.39** 0.31* 0.34** 0.23 -0.00 0.16

1.00 0.18 0.28* 0.30* 0.13 -0.10

1.00 0.28* -0.00 -0.04 0.26*

1.00 0.52** 0.37** -0.17

1.00 0.47** -0.02

1.00 -0.12

1.00

Notes: ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

Volume 12

Number 2

June 2003

© Blackwell Publishing Ltd, 2003

PRODUCT INNOVATION IN SMES

101

where Z = f (Xi, C), i.e. a linear combination of independent variables (Xi) and a constant (C). Table 3 provides the statistics of the model. First, it appears that the model does a good job in explaining variation in product innovativeness (Nagelkerke R2 = 0.386). The classification table (see Figure 2) provides an additional way to assess the model fit by comparing predictions to observed outcomes. More specifically, this table compares the observed and predicted types of product innovations, when firms with a predicted probability of 0.5 or greater are classified as having radical product innovations. In other words, this table shows only whether the estimated probability is greater or less than one-half. The off-diagonal entries of the table show how many firms were incorrectly classified. Overall, 74.24% of the firms examined were correctly classified. Table 3. Statistics for Model containing the Independent Variables -2 Log Likelihood Goodness of fit Cox & Snell – R2 Nagelkerke – R2

58.454 54.551 0.270

Table 4 contains the estimated coefficients (under column heading B) and related statistics of the logistic regression model. The model predicts radical product innovations from the variables customer information, customer responsiveness, market-driven strategy formulation, market-driven pricing policy, technology policy, proactiveness, risk-taking, firm age and a constant. The 95% confidence interval of Exp (B), which is the exponential of the coefficient, is also reported. Intervals which include the value 1 are not statistically significant at the 0.05 level. Inspection of Table 4 reveals that only two of the explanatory variables have significant coefficients. These are the entrepreneurial orientation components, notably proactiveness and risk-taking. The first is statistically significant at the 0.01 level while the second at the 0.10 level. Both these coefficients have the expected signs. These results support H3 and indicate that the higher the entrepreneurial orientation, the higher the probability for a firm to launch radical product innovations, and, hence, the lower the probability to adopt incremental product innovations. On the contrary, in the light of present empirical evidence, H1 and H2 must be rejected. Also, the coefficient of the control variable, firm age, is not significant.

0.386

Discussion Chi-Square

Significance

20.778 20.778 20.778

0.007 0.007 0.007

Model Block Step

Research findings suggest that the effect of market orientation (H1) is insignificant. Thus, they support neither of the previous claims, that market orientation has a positive impact on radical product innovations (Davis, 1993; Sandvik, Grønhaug and Øgaard, 2000) or that it has a negative impact on radical product

Predicted:

Observed: Incremental Product Innovation Radical Product Innovation

Incremental Product Innovation

Radical Product Innovation

% Correct:

36.84% 7

12

5

42

89.36%

Overall

74.24%

Figure 2. Classification Table Note:

The Cut Value is 0.50

© Blackwell Publishing Ltd, 2003

Volume 12

Number 2

June 2003

102

CREATIVITY AND INNOVATION MANAGEMENT

Table 4. Parameter Estimates of the Logistic Regression Model Variable

Constant

B

S.E.

Sig.

Exp (B)

95% CI for Exp (B) Lower

Upper

1.553

0.590

0.009

Market Orientation: Customer Information Customer Responsiveness Market-driven Strategy Formulation Market-driven Pricing Policy Technology Policy

–0.354 –0.440 –0.048 0.422 0.132

0.334 0.378 0.353 0.345 0.371

0.288 0.244 0.893 0.221 0.722

0.702 0.644 0.954 1.524 1.141

0.365 0.307 0.478 0.776 0.551

1.349 1.350 1.903 2.996 2.361

Entrepreneurial Orientation: Proactiveness Risk-taking Firm Age

1.494 0.673 –0.008

0.477 0.403 0.015

0.002*** 0.095* 0.599

4.454 1.960 0.992

1.750 0.890 0.964

11.336 4.312 1.021

Notes: *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level.

innovations (Atuahene-Gima, 1995; Bennett and Cooper, 1981; Christensen and Bower, 1996). So, it appears that when controlling for entrepreneurial orientation, market orientation appears to play a lesser role or no role at all. The question then arises, why market orientation does not affect the radicalness of product innovations in SMEs. Market orientation has rightfully received much attention as it is based on management practices (Slater and Narver, 1999). However, it may be that in our case negative effects, like those pointed out by Christensen and Bower (1996), may counteract and cancel out the expected positive effects of market orientation in SMEs. Another plausible explanation may be related to the observed limited variability across firms as regards market orientation (see Table 1). This may be attributed to particularities in the context of Greek SMEs in the food and beverages industry. It is interesting to report the opinions of industry experts, who emphasised that market orientation does matter for SMEs mainly because of the strong competition within the food and beverages industry. Thus, Greek SMEs are challenged and somehow forced to be market-oriented to effectively cope with demanding conditions underpinned by European integration and globalisation. This seems to account for high average scores given by respondents to market orientation dimensions (see Table 1). In other

Volume 12

Number 2

June 2003

words, since market orientation appears to be generally adopted by all firms, its effects may not show up in the degree of newness incorporated into product innovations. Despite recent evidence (Baltas and Salavou, 2000) that Greek SMEs tend to offer more new products when adopting stronger market orientation, the present study goes a step further and provides additional insights on their innovative activity. In particular, the present findings indicate that market orientation does not help to explain the level of innovativeness embodied in new products. However, future research in SMEs of other sectors is needed to verify whether similar results hold true, since the market orientation concept has been foremost based on large organizations (Appiah-Adu and Singh, 1998). The insignificance of technology policy (H2) contradicts previous assertions involving a positive relationship between aggressive technology policy and radical product innovations (Ettlie, Bridges and O’Keefe, 1984; Ettlie and Rubenstein, 1987; Gatignon and Xuereb, 1997). However, this finding may be attributed to constraints faced by SMEs, as in our case. More specifically, it is often argued that SMEs cannot easily ensure resources which are usually required to implement radical product innovations (Ettlie and Rubenstein, 1987). This argument appears to be strong in the case of food and beverages SMEs in Greece. The

© Blackwell Publishing Ltd, 2003

PRODUCT INNOVATION IN SMES

industry experts expressed their reservations by referring to the resource constraints that the majority of SMEs face regarding their ability to retain highly qualified and competent personnel in production and to afford heavy investments in production technology. As a result, these firms often innovate by relying on aspects other than technological along the value chain (e.g. delivery tuning, value-added services, etc.). Thus, the link between technology policy and product innovation is less strong or constrained to areas outside core technology. This particular finding seems to corroborate relevant evidence that Greek firms strive to differentiate themselves by emphasising non-technological techniques to mimic the image of established foreign rivals (Droukopoulos and Thomadakis, 1993). This relationship should, in any case, be subject to more empirical scrutiny given the little empirical evidence on the effects of technology policy in small firms (Zahra and Covin, 1993). Results provide support of a positive effect of entrepreneurial orientation on product innovativeness for both of its components, proactiveness and risk-taking. These findings support hypothesis H3. This suggests that SMEs, in which top managers behave proactively and are risk-takers, favour radical as opposed to incremental product innovations. These results are in line with international literature arguing in favour of a positive relationship between firms’ entrepreneurial orientation and product innovation. In particular, they corroborate the work of Miller, Kets de Vries and Toulouse (1982), who suggest that greater amount of proactiveness and risktaking is needed for substantial product innovations. They are also consistent with the empirical findings of Saleh and Wang (1993) which support broadly that more proactive and risky postures adopted by top managers are indicative of firms’ behaviour displaying radical product innovations. A justification of these results may, however, be tied to the nature of our sample consisting of SMEs in Greece, a country endued with unique capabilities in the entrepreneurial act. This may explain why these firms preserve their traditional behaviour which is in general characterized by family-type management dominated by one individual or a small number of individuals, usually identified with ownership (Bourantas and Papadakis, 1996). That is, top-management team is permeated by the entrepreneurial attitudes, the values and style of the owner(s). These findings converge with experts’ opinions. They emphasized that entrepreneurs in SMEs, who are both owners and managers, do play an impor-

© Blackwell Publishing Ltd, 2003

103

tant role in undertaking initiatives related to radical product innovations. Overall, our results provide support of entrepreneurial-push as an important driving force of innovative behaviour. This force seems to outweigh the traditional forces of both market-pull and technology-push in guiding radical product innovation adoptions by SMEs. This not only justifies our investigation of entrepreneurial orientation but would also trigger off further theoretical and empirical investigation on this issue.

Conclusions The purpose of this study was to investigate strategic drivers of radical product innovation adoptions in SMEs. Based on extensions of the ‘market-pull’, ‘technology-push’ and ‘strategic choice’ perspectives, it considers the potential effects of market orientation, technology policy and entrepreneurial orientation. So, this article provides evidence on these drivers of innovation from the area of SMEs, which is considerably under-researched. The specific findings support the dominance of entrepreneurial orientation. Both its dimensions (proactiveness and risk-taking) were found to be statistically important in explaining the choice of SMEs to adopt radical as against incremental product innovations. Technology policy and market orientation, on the other hand, appear to play an insignificant role. So, it appears that radical product innovations are more likely to be introduced by SMEs as a result of proactive behaviour and risk-taking posture assumed by top management. These empirical results regarding SMEs differ with those in the literature of large enterprises, which give a role to all three strategic forces investigated. These results have wider implications for both theory and policymaking. First, in relation to the debate of whether market orientation or technology policy affects the choice of high-level product innovativeness, the specific findings suggest that neither of these two drivers is important in adopting radical product innovations. On the contrary, they justify an extension of this debate to include the notion of ‘entrepreneurial-push’. This lends support to a wider argument which underlines that entrepreneurial orientation in SMEs may be more important than market orientation and technology policy in promoting aggressive innovative behaviour. Stated differently, the argument of ‘entrepreneurialpush’ is elevated, and seems to outweigh the traditional debate of market-pull and technology-push.

Volume 12

Number 2

June 2003

104

CREATIVITY AND INNOVATION MANAGEMENT

From a practical perspective, the results of this study have useful implications, especially for SMEs in Greece. Given the strong entrepreneurial tradition in the food and beverages sector, policy-makers and managers need to examine the extent to which proactiveness and risk-taking leverage radical product innovations. These capabilities appear more important than reliance on market orientation and technology policy. Thus building a fertile entrepreneurial environment would accelerate innovativeness and adaptation to the emerging new competitive landscape. Furthermore, the present analysis could be used to qualify on policies, both national and in the framework of EU innovation and entrepreneurship support programmes. Enhancing dimensions of entrepreneurial environment, which amplify SMEs’ innovative activity, especially through radical product innovations, may be crucial to designing better policies. Our findings, suggesting that it is mainly entrepreneurial orientation that favours radical as opposed to incremental product innovations, provide empirical support to such policy directions. Taking it further, this would lend support to the wider cultivation of entrepreneurship in the society. We should close, however, with a note of caution. Our study provides some new evidence drawing upon data from SMEs in the food and beverages industry of Greece. As such, it helps to project to the international literature a view from a national context other than those of large countries (such as United States, Japan, Australia) which dominate the literature. Nonetheless, it would be interesting to see whether similar relations obtain in SMEs from other industries and/or national contexts of similar set-up. This would help to confirm and extent the present results and to direct policies accordingly.

References Allen, M.J. and Yen, W.M. (1979) Introduction to Measurement Theory. Brooks/Cole, Monterey, CA. Appiah-Adu, K. and Singh, S. (1998) Customer orientation and performance: a study of SMEs. Management Decision, 36, 6, pp. 385–94. Atuahene-Gima, K. (1995) An Explanatory Analysis of the Impact of Market Orientation on New Product Performance. Journal of Product Innovation Management, 12, pp. 275–93. Atuahene-Gima, K. and Ko, A. (2001) An Empirical Investigation of the Effect of Market Orientation and Entrepreneurship Orientation Alignment on Product Innovation, Organization Science, 12, 1, pp. 54–74. Balachandra, R. and Friar, J.H. (1997) Factors for Success in R&D Projects and New Product Inno-

Volume 12

Number 2

June 2003

vation: A Contextual Framework. IEEE Transactions on Engineering Management, 44, 3, pp. 276– 87. Baltas, G. and Salavou, E. (2000) An Empirical Investigation of the Determinants of Product Innovation in Small and Medium Sized Firms, 29th EMAC Conference, Rotterdam. Bennett, R.C. and Cooper, R.C. (1981) The Misuse of Marketing: An American Tragedy, Business Horizons, 25, pp. 51–61. Blythe, J. (1999) Innovativeness and Newness in High-tech Consumer Durables. Journal of Product and Brand Management, 8, 5, pp. 415–29. Booz, Allen and Hamilton (1982) New Product Management for the 1980’s. Booz, Allen & Hamilton, Inc., New York. Bourgeois III, L.J. (1984) Strategic Management and Determinism. Academy of Management Review, 9, pp. 586–96. Bourantas, D. and Papadakis, V. (1996) Greek Management: Diagnosis and Prognosis. International Studies of Management and Organization, 26, pp. 13–32. Brouwer, E. and Kleinknecht, A. (1996) Firm Size, Small Business Presence and Sales of Innovative Products: A Micro-econometric Analysis. Small Business Economics, 8, 3, pp. 189–201. Chidamber, S.R. and Kon, H.B. (1994) A Research Retrospective of Innovation Inception and Success: The Technology-Push, Demand-pull Question. International Journal of Technology Management, 9, 1, pp. 94–112. Child, J. (1972) Organisational Structure, Environment and Performance: The Role of Strategic Choice. Sociology, 6, pp. 1–22. Christensen, C.M. and Bower, J.L. (1996) Customer Power, Strategic Investments, and the Failure of Leading Firms. Strategic Management Journal, 17, pp. 197–218. Cooper, R.G. (1979) The Dimensions of Industrial New Product Success and Failure. Journal of Marketing, 43, pp. 93–103. Cooper, R.G. (1984) The Strategy-Performance Link in Product Innovation. R&D Management, 14, pp. 247–59. Cooper, R.G. (1994) New Products: The Factors that Drive Success. International Marketing Review, 11, 1, pp. 60–76. Covin, J.G. and Slevin, D.P. (1991) A Conceptual Model of Entrepreneurship as Firm Behavior. Entrepreneurship: Theory and Practice, 16, 1, pp. 7–26. Danneels, E. and Kleinschmidt, E.J. (2001) Product Innovativeness from the firm’s perspective: Its dimensions and their relation with project selection and performance. Journal of Product Innovation Management, 18, pp. 357–73. Da Rocha, A., Christensen C.H. and Paim, N.A. (1990) Characteristics of Innovative Firms in the Brazilian Computer Industry. Journal of Product Innovation Management, 7, pp. 123– 34. Davis, D.C. (1993) Marketing in High-technology Industries: The Biotechnology Example. PhD dissertation, Kent State University, Ohio. De Brentani, U. (2001). Innovative versus incremental new business services: Different keys for

© Blackwell Publishing Ltd, 2003

PRODUCT INNOVATION IN SMES

achieving success. Journal of Product Innovation Management, 18, pp. 169–87. Drazin, R. and Schoonhoven, C.B. (1996) Community, population, and organization effects on innovation: A multilevel perspective. Academy of Management Journal, 39, pp. 1065–83. Droukopoulos, V. and Thomadakis, S. (1993) The share of small and medium-sized enterprise in Greek manufacturing. Small Business Economics, 5, pp. 187–96. Dutta, S. and Evrard, P. (1999) Information technology and organization within European small enterprises. European Management Journal, 17, 3, pp. 239–51. Ettlie, J.E. (1983). Organizational Policy and Innovation Among Suppliers to the Food Processing Sector. Academy of Management Journal, 26, 1, pp. 27–44. Ettlie, J.E. and Bridges, W.P. (1982) Environmental Uncertainty and Organizational Technology. IEEE Transactions on Engineering Management, EM-29, 1, pp. 2–10. Ettlie, J.E. and Rubenstein, A.H. (1987) Firm Size and Product Innovation. Journal of Product Innovation Management, 4, pp. 89–108. Ettlie, J.E., Bridges, W.P. and O’Keefe, R.D. (1984) Organization Strategy and Structural Differences for Radical versus Incremental Innovation. Management Science, 30, 6, pp. 682–95. European Commission Research (2002) Towards a European Research Area, Science, Technology and Innovation, Key Figures 2002, available at www.cordis.lu/ Freel, M.S. (2000) Barriers to Product Innovation in Small Manufacturing Firms. International Small Business Journal, 18, 2, pp. 60–80. Fritz, W. (1989) Determinants of Product Innovation Activities. European Journal of Marketing, 23, 10, pp. 32–43. Garvin, D.A. (1983) Spin-off and the New Firm Formation Process. California Management Review, 25, 2, pp. 3–20. Gatignon, H. and Xuereb, J.M. (1997) Strategic Orientation of the Firm and New Product Performance. Journal of Marketing Research, Vol. XXXIV, pp. 77–90. Guth, W. and Ginsberg, A. (1990) Guest Editor’s Introduction: Corporate Entrepreneurship. Strategic Management Journal, 11, pp. 5–15. Hambrick, D.C. (1989) Putting Top Managers back in the Strategy Picture. Strategic Management Journal, 10, pp. 5–15. Hambrick, D.C. and Mason, P. (1984) The Organization as a Reflection of Its Top Managers. Academy of Management Review, 9, pp. 193– 206. Han, J.K., Kim, N. and Srivastava, R.K. (1998) Market Orientation and Organizational Performance: Is Innovation a Missing Link? Journal of Marketing, 62, 4, pp. 30–45. Heunks, F.J. (1998) Innovation, Creativity and Success. Small Business Economics, 10, pp. 263– 72. Hull, F.M., Hage, J. and Azumi, K. (1985) R&D Management Strategies: America versus Japan. IEEE Transactions on Engineering Management, EM-32, 2, pp. 78–83.

© Blackwell Publishing Ltd, 2003

105

Hurley, R.F. and Hult, G.T.M. (1998) Innovation, Market Orientation, and Organizational Learning: An Integration and Empirical Examination. Journal of Marketing, 62, pp. 42–54. Jaworski, B.J. and Kohli, A.K. (1996) Market Orientation: Review, Refinement, and Roadmaps. Journal of Market Focused Management, 1, 2, pp. 119–35. Khan, A.M. and Manopichetwattana, V. (1989) Innovative and Noninnovative Small Firms: Types and Characteristics. Management Science, 35, 5, pp. 597–606. Kitchell, S. (1995) Corporate Culture, Environmental Adaptation, and Innovation Adoption: A Qualitative/Quantitative Approach. Journal of the Academy of Marketing Science, 23, 3, pp. 195– 205. Kleinbaum, D.G. (1994) Logistic Regression: A SelfLearning Text. Springer-Verlag, New York. Kohli, A.K. and Jaworski, B.J. (1990) Market Orientation: The Construct, Research Propositions, and Managerial Implications. Journal of Marketing, 54, pp. 1–18. Lioukas, S. and Makridakis, S. (1999) Globalisation and Competitive Strategies of Greek Firms (in Greek). Strategies, 1, pp. 18–28. Makridakis, S., Papagiannakis, L. and Calogirou, Y. (1996) Greek Management: Developments, Prospects, Trends (in Greek). EASE. Makridakis, S., Calogirou, Y., Papagiannakis, L. and Trivellas, P. (1997) The dualism of Greek firms and management: present state and future implications. European Management Journal, 14, 4, pp. 381–402. March, J.C. and Simon, H.A. (1958) Organization. Wiley, New York, USA. Miller, D. and Friesen, P.H. (1982) Innovation in Conservative and Entrepreneurial Firms: Two Models of Strategic Momentum. Strategic Management Journal, 3, pp. 1–25. Miller, D., Kets De Vries, M.F.R. and Toulouse, J.M. (1982) Top Executive Locus of Control and its Relationship to Strategy-making, Structure and Environment. Academy of Management Journal, 25, 2, pp. 237–53. Miller, D., Droge, C. and Toulouse, J.M. (1988) Strategic Process and Content as Mediators between Organisational Context and Structure. Academy of Management Journal, 31, pp. 544–69. Naman, J.L. and Slevin, D.J. (1993) Entrepreneurship and the Concept of Fit: A Model and Empirical Tests. Strategic Management Journal, 14, pp. 137–53. Narver, J.C. and Slater, S.F. (1990) The Effect of a Market Orientation on Business Profitability. Journal of Marketing, 54, pp. 20–35. Ruekert R.W. (1992) Developing a Market Orientation: An Organizational Strategy Perspective. International Journal of Research in Marketing, 9, 3, pp. 225–45. Russell, R.D. and Russell, C.J. (1992) An Examination of the Effects of Organizational Norms, Organizational Structure, and Environmental Uncertainty on Entrepreneurial Strategy. Journal of Management, 18, 4, pp. 639–56. Saleh, S.D. and Wang, C.K. (1993) The Management of Innovation: Strategy, Structure, and Organiza-

Volume 12

Number 2

June 2003

106

CREATIVITY AND INNOVATION MANAGEMENT

tional Climate. IEEE Transactions on Engineering Management, 40, 1, pp. 14–21. Sandvik, K., Grøhhaug, K. and Øgaard, H. (2000) The Impact of Market Orientation on Innovation and Profitability, 29th EMAC Conference, Rotterdam. Shapiro, B.P. (1988) What the Hell is ‘Market Oriented’. Harvard Business Review, 66, pp. 119–25. Slater, S.F. and Narver, J.C. (1994) Does Competitive Environment Moderate the Market OrientationPerformance Relationship? Journal of Marketing, 58, 1, pp. 46–55. Slater, S.F. and Narver, J.C. (1999) Market Oriented Is Not Enough: Build a Learning Organization. In R. Deshpandé (ed.) Developing a Market Orientation, SAGE Publications Inc., pp. 237–66. Song, M.X. and Montoya-Weiss, M.M. (1998) Critical Development Activities for Really New versus Incremental Products. Journal of Product Innovation Management, 15, pp. 124–35. Souder, W.E. (1987), Managing New Product Innovations. Lexington Books, Lexington, MA. Spanos, Y.E., Prastacos, G.P. and Papadakis, V. (2001) Greek firms and EMU: contrasting smes and large-sized enterprises. European Management Journal, 19, 6, pp. 638–48. Sweeney, G.P. (1983) New Entrepreneurship and the Smaller Firm. Campus, Frankfurt, New York. Van de Ven, A. and Ferry, D. (1980) Measuring and Assessing Organizations. Wiley, New York. Wilson, A.L., Ramamurthy, K. and Nystrom, P.C. (1999) A Multi-attribute Measure for Innovation

Volume 12

Number 2

June 2003

Adoption: The Context of Imaging Technology. IEEE Transactions on Engineering Management, 6, 3, pp. 311–21. Zahra, S.A. and Covin, J.G. (1993) Business strategy, technology policy and firm performance. Strategic Management Journal, 14, pp. 451–78.

BIOGRAPHICAL SKETCH Helen Salavou is a post-doctoral researcher at the Athens University of Economics and Business. Her research interests are in innovation management and business strategy. She holds a BSc in Business Administration, an MBA and a PhD from the Athens University of Economics and Business. Spyros Lioukas is Professor of business strategy at the Athens University of Economics and Business. He received a PhD from the London Business School. Formerly he was a lecturer at the London Business School. His current research interests are in strategy, business transformation and entrepreneurship. He has published many articles in leading academic journals including Management Science, Organization Science, and the Strategic Management Journal.

© Blackwell Publishing Ltd, 2003

PRODUCT INNOVATION IN SMES

107

Appendix A. Measures and items 1. Market Orientation (Cronbach Alpha Coefficient for scale: 0.73) To what extent does each of the following statements describe adequately your firm 7 ‘very much’)

(Response format: 1 ‘not at all’ to

Variables

Factor loadinga Factor 1

Customer Information: Listens to opinions of customers. Uses customer information to improve quality of products. Personnel has adequate information about customers and competitors.

Factor 2

Factor 3

Factor 4

0.86 0.70 0.69

Customer Responsiveness: Keeps promises made to customers. Respond to customer needs in delivery on time.

0.82 0.71

Market-driven Strategy Formulation: Company objectives are based mainly on customer needs. Uses customer information to develop new products. Company planning is organised by markets rather than products.

0.61 0.66 0.76

Market-driven Pricing Policy: Prices are determined by market conditions. Responds to customer needs in credit policy. Total Variance Explained: 67.25%

0.88 0.70

Notes: a Principal components analysis with varimax rotation.

2. Technology Policy (Cronbach Alpha Coefficient for scale: 0.87) Please rate the extent to which the following statements describe the technology policy followed by your firm (Response format: 1 ‘strongly disagree’ to 7 ‘strongly agree’) Variable

The policy of this firm has been to always consider the most up-to-date production technology available. We have a long tradition and reputation in our industry of attempting to be first to try out new methods and equipment. We spend more than most firms in our industry on new product development. We devote extra resources (i.e. time, money) to recruit the best qualified personnel in production. We devote extra resources (i.e. time, money) to technological forecasting. Total Variance Explained: 69.60%

Factor loadinga Factor 1 0.85 0.88 0.83 0.72 0.88

Notes: a Principal components analysis with varimax rotation.

© Blackwell Publishing Ltd, 2003

Volume 12

Number 2

June 2003

108

CREATIVITY AND INNOVATION MANAGEMENT

3. Entrepreneurial Orientation (Cronbach Alpha Coefficient for scale: 0.81) Please indicate the point that represents your opinion by comparing your company with your main competitors, on average for the last three years (1995–1997) for each of the following pair of sentences. Factor loadinga

Variable

Factor 1 Proactiveness Less new products. Changes in products have been mostly of a minor nature (i.e. incremental).

1234567 1234567

More new products. Changes in products have usually been radical.

Factor 2 Risk-taking

0.83 0.74

Please indicate the point that represents your opinion for each of the following pair of sentences. There is a strong emphasis on the marketing of true and tried products.

1234567

There exists a very strong emphasis on the development of new and innovative products.

0.69

There is a strong proclivity for low risk projects (with normal and certain rates of return).

1234567

There is a strong proclivity for high risk projects (with chances of very high return).

0.81

Owning to the nature of the environment, it is best to explore gradually via cautious, incremental behaviour.

1234567

Owning to the nature of the environment, bold, wideranging acts are necessary to achieve the firm’s objectives.

0.57

Typically we adopt a cautious, ‘wait and see’ posture in order to minimise the probability of making costly decisions.

1234567

Typically we adopt a bold, aggressive posture in order to maximise the probability of exploiting potential opportunities.

0.84

Typically we respond to actions which competitors initiate.

1234567

Typically we initiate actions to which competitors then respond.

0.69

We are very seldom the first business to introduce new products.

1234567

We are very often the first business to introduce new products.

0.78

We typically seek to avoid competitive clashes, preferring a ‘live-and-let- live’ posture. Total Variance Explained: 57.74%

1234567

We typically adopt a very competitive, ‘undo-thecompetitors’ posture.

0.50

Notes: a Principal components analysis with varimax rotation.

Volume 12

Number 2

June 2003

© Blackwell Publishing Ltd, 2003

INTERVENTIONS IN EXISTING OPERATIONAL PROCESSES

109

The Last Stage of Product Development: Interventions in Existing Operational Processes Frido Smulders, Léon de Caluwé and Olivier van Nieuwenhuizen We explored two questions in two empirical studies. In the first study we wanted to find out whether we could describe the interactions between New Product Development (NPD) and Operations by using a generic theory of change management and interventions. In the second study we explored such interventions during implementation and wanted to find out to what extent these interventions followed the theories of change. Some of the interactions are interventions from NPD to Operations in order to change the work of Operations. We found that these interventions and intended changes fit the theories of change management quite well. This means that the change literature gives good insights into the causes of success and failure of these interventions and could mean that, in some respect, we can apply the knowledge of these theories of change management and interventions to the field of NPD. This insight provides a new and additional perspective: NPD to be seen as a process that uses interventions to facilitate, or provoke, change within operational processes. We see too much attention paid to the project management methods and ‘hard’ theories, for example rationality, control, hierarchy, planning, predicting and prescribing. There is good reason to add the more ‘soft’ theories of change to NPD practice, with extra and explicit attention to learning, trial and error, monitoring, tell and sell, empathy, and co-operation. NPD practitioners are also very much focused on the product and less (or not at all) on the actual changes that have to take place in production (Operations) related to the implementation of these processes. For successful implementation interventions need to be tuned towards the receiver. The interventionist (i.e. NPD practitioners) must be aware that there is a range of choice if different intervention strategies in order to tune their intervention efforts more effectively towards the receiver. In order to be able to do this, NPD practitioners, as well as the people from Operations, need to become skilled in using a generic theory of intervention and change, such as the one described in this paper, during their interactions.

Introduction

T

his paper focuses on the interface between New Product Development (NPD) and Operations. The NPD process is defined as the process that leads to the creation of product and process descriptions, whereas Operations are defined as the transformation of these descriptions into physical products. Companies nowadays are forced to react quickly to external changes by adapting their products and their Operations to those changes. These adaptations result in a continuous stream of process and product innovation projects as © Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.

well as the day-to-day operational processes. The focus of this paper will be on the main stream of new products like line extensions, next versions, etc. that have to be implemented in operational processes. These dayto-day new products are not believed to be disruptive in any sense. Excluded from this research are breakthrough products. For this stream of new products the interface between NPD and Operations a vital link. In order to be effective in their innovative efforts, companies need to understand all the interactions between NPD and Operations. This is not always the case.

Volume 12

Number 2

June 2003

110

CREATIVITY AND INNOVATION MANAGEMENT

Boer and During (2001) have found in their extensive research, that people have a tendency to overlook the organizational adaptations that are necessary to implement the object of innovation. This was found in all three different types of innovation processes they included in their research, i.e. product innovation, process innovation and organizational innovation. This misfit in perception regarding the necessary adaptations is one of the symptoms that call for a better understanding of the actual process that takes place while implementing the object of innovation. What Boer and During (2001) found is not really new, earlier research by Leavitt (1965) and Ruttan and Hayami (1984) found in some respect similar results. What is surprising however, is that during all these years of research in the field of innovation, these problems of implementation still occur. This calls for additional research dedicated to the subject of implementation. Boer and During (2001) speak about organizational adaptations, which means in our perception that some changes have to be made to it is organization in order to implement the object of innovation. Because the focus of this paper is on the product innovation process, the object of innovation will be the newly developed product. However, the literature on NPD, as we will show later, does not mention the organizational adaptations that are necessary to implement the results of the NPD process. What is mentioned and described, like Clark and Fujimoto (1991), is the development of the operational processes parallel to, or after the development of the product. However we have not found any authors that describe the process of changing the present operational processes in order to be able to produce the new product. In this paper we will explore this implementation phase by making use of theories of change management. We want to find out if we can see the interactions between NPD and Operations as interventions and if we can treat them as such. We will present the results of two empirical studies in three companies. In the first study we wanted to find out if we could describe the interactions between NPD and Operations during the implementation process by using a generic theory of change management and interventions. In the second study we explored the (micro) interventions during implementation and wanted to find out to what extent these interventions followed the theories of change. The paper is structured as follows. First we will discuss the existing theories on NPD regarding what they say about the transition

Volume 12

Number 2

June 2003

of NPD results to Operations. Then we will describe a generic theory of change and intervention. We go on to describe our perspective of the development process and more specifically our perspective of the last stage of development, i.e. the transition of the new product from development to Operations. We will illustrate that this phase, theoretically, can be seen as an intervention process that facilitate or provoke (micro) changes within Operations. Next, the results of the two empirical studies will be presented. Finally we discuss the answers to the earlier mentioned issues as well as some theoretical and managerial implications.

Implementation Phase of Product Development: What Do Theories Say? In this action we will discuss what the literature on NPD tells us about the last stage of the development process. First the formal models are discussed, followed by some specific NPD concepts that might include this transition to Operations. The transition of product and process descriptions from the development stage to the production stage can be called implementation. In this phase these descriptions will be implemented in the existing processes within Operations, like purchasing, production, sales, etc. Operations will convert the descriptions of the product into the physical product that in its turn can be delivered to the customer. Thus, this implementation phase can be seen as the last internal phase of the product innovation process. The importance of this phase is illustrated by the fact that during implementation senior management’s attention is at its highest level (Wheelwright & Clark, 1992). This is not surprising, because during that period investment is made in machines, production lines, tools, training programmes and sometimes even a new factory was to be built. Whatever the amount of investment or whatever the amount of necessary change is, internal implementation is the first ‘proof of the pudding’ regarding the quality of the developmental part of the product innovation process. Comparing the existing NPD models, the definitions and terminology used for this phase of transition from development to operational processes differs a lot. Some authors, such as Pahl and Beitz (1984), describe this transition as the final result of the NPD project by using respectively the terms ‘installation’ and ‘realisation’ for this phase. In the Stage-

© Blackwell Publishing Ltd, 2003

INTERVENTIONS IN EXISTING OPERATIONAL PROCESSES

GateTM model (Cooper, 1993) it is referred to as ‘testing & validation’ before the process is finalised by ‘full production & market launch’. Archer (1971) has defined it as the beginning of production and sales, and called it ‘manufacturing and marketing start-up’. Buijs and Valkenburg (2000), following Andreassen and Hein (1985), describe development as three parallel processes. Integrated product development (IPD) during the NPD process brings together three main disciplines, product development process development and market development. The last phase of these three parallel processes they define ‘market introduction’ as the moment to start production and the introdution of the new product to the market. Not included during this last phase are any changes to current Operations. Apart from differences in terminology all these models lack a well-described internal implementation phase. Most authors just indicate by means of arrows that there is a transition from development activities to operational activities (like production, sales and distribution, etc.) but in fact do not describe what is really happening during that transition. Of course, these models are very much focused on the development of the product and associated processes and not so much on implementing the changes in the down stream phases. That seems to be the territory of other researchers. Some authors, like Pugh (1991) and Ulrich and Eppinger (1995), focus on design for manufacture (DFM) to secure the transition from NPD to Operations. According to Pugh (1991) the main aim of DFM is to minimize the component, assembly costs, and development cycles, and to enable higher quality products to be made. DFM is just one of the many ‘design fors’. All these together are called ‘Design for X’. ‘X’ stands for any of the quality criteria that feed the product design specification. Others are, design for assembly (DFA), design for piece-part producibility, design for serviceability (DFS), design for packaging, maintenance, etc. All these various aspects (‘X’) of a product must be in balance and not applied in isolation. Cross-talk between the ‘design for’ disciplines and specialisms is necessary to prevent an imbalance of the characteristics of the newly developed product (Pugh, 1991). However, none of these Xs takes into account the ‘soft side’ of the process such as developing interventions and facilitating the necessary changes to the work force. We did not identify any ‘X’ that focuses on implementation of the new product, something like ‘design for implementation’ (DFI), nor did

© Blackwell Publishing Ltd, 2003

111

we find any ‘X’ that includes intervention and change processes, e.g. ‘design for change’ (DFC). Concurrent engineering (CE) is the last NPD concept that is discussed here. CE focuses on the interface between NPD and Operations by concurrent development of the product and associated processes in a cross-functional integrated way (e.g. Clausing, 1989). The broad concept of CE focuses on doing the development work in parallel as opposed sequentially. In other words CE integrates upstream and downstream development stages and brings as many downstream considerations as possible into early decision-making (Clark, Chew and Fujimoto, 1994). Cross-functional integration of NPD and the different operational processes seems a conditio sine qua non within the concept of CE. Defined as such the approach of CE looks similar to the approach of integrated product development (Buijs & Valkenburg, 2000). Some of the studies in the field of CE concentrate on the various contingent organizational factors to detect the importance of their role. On the other hand there seems to be over-attention to the development and application of all kind of (computer) tools that must support or facilitate the use of CE (Gonçalves, Roy and Steiger-Garçao, 2002). Nor, within the field of CE, did we find any authors that concentrate explicitly on intervention and change within the operational processes. As we may conclude from these different NPD theories, concepts and approaches there is very little focus on the actual transition from NPD results to Operations. One of the exceptions to this is the study of Clark and Fujimoto (1991). They describe this transition to manufacturing as a process in which a multi-functional team (or teams) work in parallel on the development of the operational processes and during the ramp-up of production. Ramp-up is defined as the period that the manufacturing process needs to reach from zero to full production. Although they also focus on implementation to production, they do not consider the changes that have to be made within the existing operational processes nor do they consider the interventions that are necessary to provoke these changes. And that is the focus of this exploratory study.

Generic Theory of Change and Intervention: Thinking in Five Different Colours In this section, we will describe in generic terms the theory of intervention and change.

Volume 12

Number 2

June 2003

112

CREATIVITY AND INNOVATION MANAGEMENT

They can be seen on different levels. An intervention can be regarded as a single sentence or as a major series of planned actions. We adopt the general definition of Argyris (1970), which covers these different levels. He states that: ‘to intervene is to enter an ongoing system of relationships, to come between or among persons, groups or objects for the purpose of helping them’. Cummings and Worley (1997) elaborated on this general definition. They say that an intervention is ‘a set of sequenced planned actions or events intended to help an organisation increase its effectiveness. Interventions purposely disrupt the status quo; they are a deliberate attempt to change an organisation or sub-unit toward a different and more effective state’. In the framework of this article we use the term intervention for the deliberate actions leading to the implementation of new products in Operations. The word change can be characterized as a container concept. A search for the underlying values and meanings of the word results in a whole range of meanings used by change agents. For example, the word change is used to describe the desired result: the result that has been realized or the effect that has been achieved. But the word change is also used to describe the process: the transition from one situation to another or the working method followed. These two meanings are common knowledge to many. In addition, there are definitions of the word change that are more or less unconscious: they concern ways of thinking and strategies hidden within the word itself. Five of these meanings predominate in the large body of literature on (planned) change and interventions. Caluwé and Vermaak (2003a) integrated this body of literature into a generic theory that uses five different colours to cover five different perspectives of change (see Table 1). Caluwé and Vermaak are aware of the six colours that De Bono (1985) uses for his socalled Thinking Hats. Regarding these colours of De Bono, Caluwé & Vermaak (2003a) say: ‘In his (De Bono) approach, ways of thinking are unravelled in order to use just one method or enable a group to exercise the different approaches. In our description of the colours, any relationship to the De Bono’s Thinking Hats is expressly avoided’. There is also no connection with the three colours that Rhodes and Thame (1988) use to discuss three different thinking styles. Although some parts of the concepts of De Bono (1985) and Rhodes and Thame (1988) may show some similarity and overlap with the colours of Caluwé and Vermaak (2003a), there is no direct connection between these concepts.

Volume 12

Number 2

June 2003

According to Caluwé & Vermaak (2003a) there is a world of difference between the five coloured perspectives in terms of: • the underlying assumptions; • the ideals and pitfalls; • the types of intervention and the people involved; • the competencies and role of the change agent; • the predictability of the result, the nature of steering, guaranteeing, etc. Table 1 provides characterisations of the five colour-print in key words. Apart from the keywords in Table 1, we will briefly describe each colour. The underlying train of thought behind yellow-print thinking is the search for common interests, the effective pursuit and implementation of complex goals in an environment of many (conflicting) goals and influences. The colour yellow symbolizes power ‘the sun’, ‘fire’ and the nature of coalition formation ‘hatching processes’ and ‘fireside chats’. The basis of blue-print thinking lies in recording the result in advance, specifying the requirements, the successive implementation of activities and adjustments in the light of the result to be achieved. In this case, change is especially aimed at material matters and form. The term ‘blue-print’ refers, therefore, to the previously determined design that is then implemented. Red-print thinking typically achieves change by stimulating people in the right way by deploying HRM (Human Resources Management) tools. These tools assist change in the ‘soft’ aspects of the organization, such as management style, talents and competencies. This aspect is concerned with people, the colour symbolizing human blood. People must be influenced, enticed and simulated. Green-print thinking concentrates on changing people or stirring them into action by motivating them to learn, bringing them into learning situations and increasing their learning potential. The result depends to a large extent on this learning potential. Greenprint thinking is aimed at getting people to move, ‘giving them the green light’. It is also concerned with ‘growth’ just as is represented by the colour green in nature. Finally, we describe white-print thinking. The philosophy behind this is that everything changes spontaneously and permanently. The change agent removes obstacles based on painstaking observation and interpretation of what is happening. White-print thinking draws on the inner strength and

© Blackwell Publishing Ltd, 2003

INTERVENTIONS IN EXISTING OPERATIONAL PROCESSES

113

Table 1. The Five Ways of Looking at Change Things / people will change if you . . . Yellow-print

Blue-print

Red-print

Green-print

White-print

– can unite the interest of the important players – can compel people to accept (common) points of view/opinions – can create win–win situations; can form coalitions – demonstrate the advantages of certain ideas (in terms of power, status, influence) – get everyone on the same wave-length – can bring people into a negotiating process – formulate a clear result/goal beforehand – lay down a concrete plan with clear steps from A to B – monitor the steps well and adjust accordingly – keep everything as stable and controlled as possible – can reduce complexity as much as possible – stimulate people in the right way, for example, by inducements (or penalties) – employ advanced HRM tools for rewards, motivation, promotions, status – give people something in return for what they give the organization (barter) – manage expectations and create a good atmosphere – make things attractive for people – make people aware of new insights/own shortcomings – are able to motivate people to see new things/to learn/to be capable of . . . – are able to create suitable (collective) learning situations – allow the learning process to be owned by the people involved and geared toward their own learning goals – start from people’s drives, strengths, and ‘natural inclinations’ – add meaning to what people are going through – are able to diagnose complexity and understand its dynamics – give free rein to people’s energy and remove possible obstacles – make use of symbols and rituals

Source: Caluwé & Vermaak, 2003a, p.45.

direction of people rather than on external safety or reassurance. Giving things meaning also plays an important role. The colour is white because white incorporates every colour and offers the most space for self-steering: everything is still open.

Our Perspective on the Implementation Phase: Intervention in Existing Processes In the paragraph on NPD theory we have seen that apart from the fact that some authors focus extensively on the manufacturability of the new product, there is very little focus on the actual implementation processes of the new product within the present operational processes.

© Blackwell Publishing Ltd, 2003

More or less similar to Clark and Fujimoto (1991) we identified three processes (Smulders, Buijs & Valkenburg, 1999) within the total process of product innovation, product development processes, operational processes and consumption/use processes. Later a fourth process was added, the product management process. These four processes (Figure 1) are described as follows: product managers within the product management process (1) base their ideas for new products on feedback that they receive from the market and ask NPD to develop (2) the new products and associated processes and come up with descriptions of these two; the operational processes (3) use these descriptions to manufacture the physical products and sell them to the consumers/clients in the market that will

Volume 12

Number 2

June 2003

114

CREATIVITY AND INNOVATION MANAGEMENT

(1) Product management

Product management (2) NPD processes

(3) Operations ‘O’

NPD Operations ‘O’ + ∂

time

(4) Market: using proces

Figure 1. The Four Main Processes of Product Innovation in a Schematic Overview

use (4) the new product. The feedback in its turn provides market information to the product management process (i.e. product manager). Figure 1 shows these processes in a schematic overview. The two processes in dotted lines, product management and customers use, will not further be addressed in this paper. In general, introducing the necessary changes and adjustments to the existing operational processes ‘O’ forms the implementation of a new product within the Operations of a company. There will always be changes necessary, because if nothing changes within Operations the present output, the present products, will stay the same. These changes (
) can range from very simple in terms of a newly shaped tool for injection moulding to very difficult, for example the assembly of an extremely complicated new product or the start-up of a radically new production process. But the operational process (‘O’) must undergo some sort of change (
) and become ‘O’ +
. In order to implement the change there must be interactions between NPD and Operations. In Figure 1 we have illustrated these interactions by an arrow from NPD to Operations. If we talk about interactions that have to facilitate or provoke some kind of change (
), then we might speak, according to the theory of change, about intervention and interventionists. We explored this idea in the empirical studies.

Number 2

We conducted two empirical studies, one for each question.

First Research Question, First Study

In the empirical studies we wanted to explore if we can apply the (meta-) theory on change as developed by Caluwé and Vermaak (2003a) to the interactions between NPD and Operations. More specifically, we wanted to explore two questions:

Regarding the first question we gathered empirical data in two different companies. One company produces high-end consumer electronics (e.g. TV, audio, etc.) and the other produces high-end lighting products (e.g. for discos, festivals, etc.). The products of the two companies were comparable in terms of product size, product technology, production methods and end assembly. In each company we interviewed eight people from production and from product development. The people chosen were involved with two recent product innovation projects for each company. The interviews were open, half-structured and taped. On our first research question the results show the following. In one of the companies a person from development is very much aware of the fact that he needs to come up with a product that is easy to produce and assemble. In order to be sure of some elements of his conceptual product he sends a preliminary (still conceptual) drawing to a technical person in production and asks for feedback on those elements. What he receives back is a drawing with ‘red all over’, all kind of remarks on elements and tolerances that were not asked for. The comments are far to detailed given the conceptual status of the drawing. Apart from the fact that he is not really getting an answer to his question, the next quote illustrates that the action of posing the question to production can be seen as an intervention:

1. Can we see (some of) the interactions between NPD and Operations as interventions?

‘problem is to convince them [production] to use time . . . well to get them out of their normal and everyday life (of production),

Empirical Studies

Volume 12

2. To what extent do the (micro) interventions during implementation follow the theories on change?

June 2003

© Blackwell Publishing Ltd, 2003

INTERVENTIONS IN EXISTING OPERATIONAL PROCESSES

to something that is totally new and then give feedback’ In fact the NPD person is struggling with choosing the right type of intervention in order to provoke the desired (micro) change within the activities of the man from production. Later in the development process, when the 0-series (
20 products) have to be done in the actual production area, NPD asks for 20 metal plates. ‘it is not easy to do 0-production up there, because they are focused on doing their normal production. So, uh . . . they rather run thousands plates’ Again this is an intervention from NPD within the existing production process in order to provoke change. After the 0-series there is in both companies a production ramp-up. This ramp-up towards full production volume, can take some 10–12 weeks and is on the real assembly line. At this time all people that will be working on that line will be involved. The assembly workers are trained during this period by the people that have been involved in the 0-series. This training is a typical green intervention. During the first two or three days of the ramp-up it might be that only one or two products are assembled, whereas the target of volume production is 80 products a day. People have to learn the complexity of the product and also have to learn to detect the defects on parts that they have to assemble. During the ramp-up the first series of products need to be tested. In the lighting company this is done by NPD. It often happens that because of this testing some modifications need to be done to the product. In such cases NPD orders (blue intervention) production to stop. Or as one of the production managers remarks: ‘they [NPD] don’t know what it means to stop production . . . we have 25 workers waiting . . . they want to deliver the new products . . . NPD doesn’t have the same insight [empathy] in production as we do’. This is again an intervention from NPD that according to the interviews is the rule rather than the exception during the ramp-up phase. From these cases we learned that implementation of a newly developed product in Operations is a process over time rather than a moment in time. One could say that implementation in terms of interactions starts as early as the concept phase and continues after

© Blackwell Publishing Ltd, 2003

115

the ramp-up. We also learned that most of these interactions could be seen as interventions from one process to another process. Of course there is one moment in time that the first product is made on the real production line, but it is necessary to prepare for this and therefore interactions between participants from the two processes are necessary. This preparation is done during the full length of the development period, and many (micro) interventions take place. In most cases these interventions are from NPD towards Operations, i.e. NPD is the one that intervenes in the process of Operations. The character of the interventions and the changes that these provoke seems to change over the period from concept development to ramp-up. During concept development, NPD seeks information and just wants somebody from Operations to step out of his daily work for a short while in order to concentrate on the question from NPD. This could be regarded as a micro intervention, just changing the behaviour of somebody else for a short period of time. Later in the process, during the zero-series, the intervention looks similar. Although, changing the attitude of Operations in running only 20 plates instead of thousand plates might be not that easy. Then, at the start of the ramp-up, the people from the production line need to learn how to manufacture and assemble the new product. Again an intervention, but this time it is towards the desired new state of making new products. This new state has a more permanent character. During the interviews this moment in time, the first days of the ramp-up where the first products are made, was not mentioned very often. It does not seem to give a lot of unexpected trouble. People sort of know about these start-up troubles and therefore expect and accept them. However, interventions from NPD later during ramp-up are experienced as much more disturbing and unwanted and therefore create a lot of friction. Concluding this part of our research, we can say that in the process of preparation the implementation and immediately after the beginning of the ramp-up many interventions take place, mainly from NPD towards Operations. As we will discuss later, from the perspective of the five-colour theory there might be options open to the person from NPD to choose a particular intervention. But first we need to know if we can apply this theory. The question that arises from this is, to what extent are these micro interventions following the theories of change? This is what we explored in our second empirical study.

Volume 12

Number 2

June 2003

116

CREATIVITY AND INNOVATION MANAGEMENT

Second Research Question, Second Study The second study was carried out within a large food company that produces fastmoving consumer goods. With this study we wanted to explore our second research question: ‘To what extent do interventions during implementation follow the theories of change?’ We asked ourselves the question how could we find out if the interventions and the changes follow the change management literature to some extent. First we decided to use as a reference the generic theory on change and interventions that we described earlier (Caluwé & Vermaak, 2003a), i.e. the five different colours. Next we thought that we needed to check if the effectiveness or ineffectiveness of a certain intervention follows the rules of this five-colour theory. In order to do so, our data must provide information on the colours of the interventions, the effectiveness of those interventions as well as information about the intervention (i.e. colour) preferences of the people involved. We defined the person that made an intervention as the subject and the person that needed to change his/her work as being the object. The idea was that we would answer our research question if were able to show that interventions – regarding the implementation of a new product – that are in a colour other than the (intervention) preference of the object are less effective then interventions that are in the preferred colour of the object. We still look at micro changes that have to be facilitated or provoked by the interventions we are interested in, i.e. interventions necessary to implement a new product within the operational processes. A total of 35 people involved in six product innovation projects were interviewed. We interviewed people from the development side, and people from the various departments within Operations. They were developers, marketers, purchasers, supply-chain planners, controllers, and people from the quality department. Again, the interviews were open, half-structured and taped. The transcripts of the tapes were analysed with WinMax.1 Apart from the interviews we used an ‘Alpha’ version2 of a psychological test that was developed to record the intervention preferences (in colours) of the test-person.

1 WinMax is a software program for qualitative analysis of text. 2 With an ‘Alpha’ version of this test we mean that the test has only recently been developed and is not yet validated.

Volume 12

Number 2

June 2003

Figure 2. Template to Describe the Effectiveness of an Intervention

Apart from filling in the test we had to find out during the interviews: • who (subject) was making an intervention on whom (object); • what was the content of the intervention; • what was the aim of the intervention; • how was the intervention made; • how effective was the intervention. Effectiveness was measured by comparing the result of the intervention, in terms of resulting change of the object, with the aim that the subject had in mind, i.e. the desired change of the object (see Figure 2). From the interview data we first identified the interventions. Then we identified the colours of the intervention as well as its effectiveness. From the psychological tests we got the intervention preferences (in colours) of the people involved, i.e. the preferences of the subject of the intervention and preference of the object. We identified 18 interventions, of which 7 were in the preferred colour of the object and 11 were in another colour (see Table 2). It was not hard to identify the interventions, because in the interviews the ineffective interventions were often the first frustrations to come up. An example: in one case the blue-thinking project manager made, in order to finish the project in time, a detailed timeline with actions for everybody in his project team, including the actions for the white-thinking developer. This is an extremely blue-coloured intervention. This intervention was, for the rest of project team highly effective, because they liked it, as blue thinkers. But for the developer, whose preference is white, the intervention had been ineffective. His response for being too late was: ‘I am a creative man. And creativity should not be bonded by rigid things, such as timelines. That is the reason why I have been too late’. Looking at the results of the data analyses we can see that when a subject intervenes using a colour that meets the preference of the objects, the chance of being effective is higher than when the colour of the interven-

© Blackwell Publishing Ltd, 2003

INTERVENTIONS IN EXISTING OPERATIONAL PROCESSES

117

Table 2. The Effectiveness of Interventions in Preferred and in Not-Preferred Intervention Colours Interventions (n = 7) in preferred colour object by object 2 failures 2 medium successful 3 successful

= 29% = 29% = 42%

tion is different from the preferred one of the object. According to the results from this second empirical study we find strong indications that the application of theories of change, or at least awareness of these theories might improve the effectiveness of interventions between NPD and Operations.

Discussion From the two empirical studies we have learned that (in some) cases we can see an interaction between NPD and Operations as an intervention that has to facilitate or provoke a (micro) change within the other process. We also found in our second study that the effectiveness of such an intervention depends largely on the match between the colour of the intervention and the colour preference of the object, the receiver of the intervention. These findings imply that in the case of an intervention, the interventionist (mostly NPD) has a choice among different types of interventions. In this part we will discuss two phases where interventions between NPD and Operations are likely, the interventions during the conceptual phase and interventions during the 0-series. During the conceptual phases of development the people from Operations are only involved in the development project parttime. During that period they act like consultants to the NPD-people. However, for each question that is put to them they have to step out of their day-to-day work in order to concentrate on it and provide valuable answers. In a way we could say that there are two necessary successive interventions. The first one is to get the people in the other process to spend time on such a question and the second intervention is to get them in the right ‘process’ mood in order to be able to interpret and answer the question properly. Solving day-to-day problems within the production process is different from concentrating on conceptual drawings. There are differences in the

© Blackwell Publishing Ltd, 2003

Interventions (n = 11) in other colour preferred

5 failures 5 medium successful 1 successful

= 45% = 45% = 10%

‘frame of thinking’ between the two processes. Within Operations the solutions to problems are mostly quite obvious and need to be directly implemented to prevent any extra production losses – a blue-print type of approach. Within the NPD-process one will find white- and green-print approaches. Thus to provide good answers to NPD, the person from Operations needs to adapt to this white and green frame of thinking. That needs some kind of intervention. Each time NPD has a choice as an interventionist on how to approach the people from production. For illustration purposes we will show a possible type of intervention for each colour. If NPD just sends an email with drawings as an attachment, then this is a blue intervention. Making an appointment to jointly discuss the drawings is an example of a red intervention. If such an appointment has the character of a negotiation about possible results, it is a yellow intervention. Coaching production to be able to answer the question is a green intervention, and leaving it for later and letting production do it their way, could be regarded as a white intervention. Not all types of intervention are equally realistic, or will be equally successful, but the purpose of describing them is to illustrate that NPD as an interventionist is able to choose between different interventions. And as we learned from the second study, the effectiveness of the interventions is dependent on making the right choice from the perspective of the object/receiver. The second phase that we will discuss here is the period around the 0-series. Let us consider that from the perspective of the production side. In both companies in our first study, people on the production side were taken away from other production lines and told (blue intervention) to go and work on the 0-series of the new product. But we noted a difference between the two companies. At that time in company A this was the first time production had seen the new product. In company B, the people chosen for the 0-series had already seen a prototype in an earlier

Volume 12

Number 2

June 2003

118

CREATIVITY AND INNOVATION MANAGEMENT

phase and therefore they knew what the product should look like. In company B there had obviously been an earlier intervention to accomplish that. This fact made it easier for them to assemble the first product. There is another difference; during the 0-series in company B there were always people from NPD present to help and coach the process. They all made notes of the problems that occured during assembly. Being there and making notes is a green intervention with red (motivating) elements. In company A, the first product of the 0series was done under the supervision of a technical assistant from Operations and sometimes with somebody from NPD. There is a difference in the kind of intervention if people from NPD are there or not. Being there and showing responsibility and interest gives a red glow to the intervention. This red glow will provide extra motivation to the people from production. However, if the technical assistant has not seen the product before and there is at that moment nobody from NPD, then this is a very white intervention: ‘Let’s see how we have to put this together?’. In such a case they more or less get a box with parts and have to figure out themselves how to fit these together. This will obviously be less effective in terms of enthusiasm for the new product, time needed to get know the new product, but also to ‘debug’ the new product. It would make a difference to learning if the technical assistant had seen a prototype and/or when somebody from NPD is present at the production side. As we can see from these examples, there is not only a choice between different interventions from NPD to Operations, e.g. to be present at the 0-series or not, but there is also something to be said for making an intervention much earlier in the process in order to have a more effective intervention later. This asks for much more conscious attitude, in terms of planning and choosing different kinds of interventions, from the side of the interventionist. For being able to plan such a series of interventions, the interventionist must be knowledgeable in the field of change management.

Conclusion In this paper we explored the application of theories of change management and interventions to the interactions that take place between NPD and Operations. In doing this we focused mainly on the interactions before, during, and after the start of production.

Volume 12

Number 2

June 2003

We found that some of the interactions are indeed interventions from NPD to Operations in order to change the line of work of Operations. We also found that these interventions and intended changes fit the theories on change management quite well. This could mean that in some respects we can apply the knowledge of these theories of change management and interventions to the field of NPD. In our opinion this gives a new and additional perspective: NPD can be seen as a process that uses (micro) interventions to facilitate (or provoke) changes within operational processes. This new perspective gives us the possibility to add ‘soft’ change management theories and tools to the mostly rational (‘hard’) models within the existing literature on NPD. From these two studies we also learned that people from NPD are very much focused on the product itself and less on the actual changes that have to take place in Operations. During the interactions with the people from production, they do not take in account that they are actually intervening in their processes, i.e. disrupting the status quo and change it to a different state (Cummings and Worly, 1997). Of course they know that the new product will cause some changes, but they do not recognize it as a (micro) change process with associated interventions. They just bring the blueprints with them (if they show up at all) and discuss the testing or other things that must be done. It is clear that alternative interventions are not only possible, but are sometimes even desirable. On the other hand, production is not explicitly aware of the fact that NPD is intervening and trying to provoke (micro) changes in the course of their work. Production seems to accept the fact that NPD is disturbing their work although they do not like it. They experience these interventions as rather annoying which often results in counter productive (re-) actions or misunderstanding of what actually is meant by NPD. NPD practitioners need to think in terms of the object or receiver of the intervention. In many cases this receiver has to go through some kind of change process. A successful intervention needs to be tuned towards this receiver. A choice has to be made between alternative interventions, some might be more successful than others. So, there is a need that participants in the NPD process become, in a way, skilled interventionists, able to choose feasible and desirable interventions to facilitate the change process. Especially in their actions towards those parties that are only involved in the product innovation process

© Blackwell Publishing Ltd, 2003

INTERVENTIONS IN EXISTING OPERATIONAL PROCESSES

part of the time. A more conscious application of theories of change and intervention, as well as more awareness of intervention preferences (colours) of the parties involved, could make these interactions more effective. This awareness of the preferences of the receiver calls for empathy. NPD must develop some sort of capacity to understand (to a certain level) the feelings of people from Operations. From social psychology we know that, to a certain extent, the feelings and behaviour of people are determined by characteristics of the process that they are in. So, not only empathy on participant level, but also empathy on process level might be necessary in order to be able to choose the right moment and the right type of intervention. We would even take a step further. We assume that it would be very useful for both workers in NPD and Operations to become aware of the different interventions and change processes and that they become skilled in using the theory of the five colours as a language in order to discuss the processes they are in or they are starting. ‘Will this be a blue process? Is this a negotiation exercise (yellow)? Is this a learning experience? Is this two-way communication (red) or can we take the initiative (white)?’. We see this as a very fruitful application that might play a role as an ‘interface language’ in addition to the desired functional ‘multilinguality’ of the NPD practitions as mentioned by Clark and Fujimoto (1991), but it needs to be tested and researched. In other contexts these applications work very effectively, for instance in the communication between organisational change agents (see Caluwé and Vermaak, 2003b). They become aware of the assumptions of others as well as their own assumptions and can talk about these assumptions without any form of conflict. In our future work we will map the fivecolour theory of Caluwé and Vermaak (2003a) on the different stages of the total product innovation process. So far we have only looked in an explorative way at the last stage of product development. We are also interested in finding out what balance in colours an apparently seamless product innovation process has among all the interventions that have to take place during the course of the process.

Acknowledgement The authors would like to thank Harry Boer and Poul H.K. Hansen from Aalborg University Denmark and The Netherlands Organiza-

© Blackwell Publishing Ltd, 2003

119

tion for Scientific Research (NWO) for their respective support of this research project.

References Andreasen, M.M. and Hein L. (1985) Integrated Product Development IFS (Publications) LtD/ Springer-Verlag, Bedford, UK/Berlin, Heidelberg. Archer, L.B. (1971) Technological Innovation: a Methodology. Inforlink, Frimley. Argyris, Ch. (1970) Intervention theory and method: a behavioral science view. Addison Wesley, London. Boer, H. and During, W.E. (2001) Innovation, what innovation? A comparison between product, process and organizational innovation, International Journal of Technology Management, 22, 1–3, 83–107. Buijs, J.A. and Valkenburg, R. (2000) Integrale Productontwikkeling. Lemma, Utrecht. Caluwé, L. de, and Vermaak, H. (2003a) Learning to change. Sage Publications, Thousands Oaks, CA. Caluwé, L. de and Vermaak, H. (2003b) Thinking about change in five different colours. In Boonstra (ed.) The psychological management of organisational change. John Wiley & Sons Ltd, Chichester. Clark K.B., Chew, B. and Fujimoto, T. (1994) Manufacturing in design: beyond the production/R&D dichotomy. In Susman, G. (ed.) Integrating design and manufacturing for competitive advantage. Oxford University Press, New York, pp. 178– 206. Clark, K.B. and Fujimoto, T. (1991) Product Development Performance: Strategy, Organization and Management in the world auto industry. Harvard Business School Press, Boston MA. Clausing, D. (1989) Concurrent Engineering. In Proceedings Symposium on Concurrent Product and Process Design. ASME, San Fransisco, December. Cooper, R.G. (1979) Introducing successful new products. European Journal of Marketing, 10, 6. Cooper, R.G. (1993) Winning at new products: Accelerating the process from idea to launch. Addison Wesley, Reading MA. Cummings, T. and Worley, C. (1997) Organization Development and change. West Publishing, Minneapolis. De Bono, E. (1985) Six thinking hats. Penguin Books, Middlesex. Gonçalves, R.J., Roy, R. and Steiger-Garçao, A. (eds) (2002) Advances in concurrent engineering. A.A. Balkema Publishers, Lisse. Leavitt, H.J. (1965) Applied organizational Change in Industry: Structural, Technological, and Humanistic Approaches. In March, J. (ed.) Handbook of Organizations. Rand McNally, Chicago, pp. 1144–1170. Pahl, G. and Beitz, W. (1984) Engineering Design; a Systematic Approach. Design Council, London. Pugh, S. (1991) Total design. Addison-Wesley, Reading MA. Rhodes, J. and Thame, S. (1988) The colours of your mind. Collins, London.

Volume 12

Number 2

June 2003

120

CREATIVITY AND INNOVATION MANAGEMENT

Ruttan, V.W. and Hayami, Y. (1984) Toward a Theory of Induced Institutional Innovation. Journal of Development Studies, 20, 203–223. Smulders, F.E.H.M., Buijs, J.A. and Valkenburg, A.C. (1999) Waarom is innoveren zo moeilijk? Het antwoord of blijven er nog uitdagingen over?, Management en Organisatie, juli/augustus, 5–18. Ulrich, K.T. and Eppinger, S.D. (1995) Product Design and Development. McGraw-Hill, New York. VDI 2221 (1987) Systematic Aproach to Design of Technical Systems and Products., Verein Deutscher Ingenieure, Düsseldorf. Wheelwright, S.C. and Clark, K.B. (1992) Revolutionizing Product Development. Free Press, New York.

Volume 12

Number 2

June 2003

Frido E. Smulders is assistant professor Management of Innovation at the faculty of Industrial Design Engineering, Delft University of Technology and consultant with Mosaic Management Consultants. Léon de Caluwé is professor in change Management at the Free University of Amsterdam and partner at Twijnstra en Gudde Management Consultants. Olivier van Nieuwenhuizen studied Industrial Design Engineering at Delft Technical University of Technology. He is now working as a product manager / account manager at ContiTech, The Netherlands.

© Blackwell Publishing Ltd, 2003

COPING WITH CHAOS IN CHANGE PROCESSES

121

Coping with Chaos in Change Processes Klaasjan Visscher and Arie Rip In their efforts to change organizations, managers and change consultants are time and again confronted with the limited controllability of organizations, the complexity and indeterminacy of change processes and the uncertain and ambiguous effects of their actions. In short, they are confronted with chaos. Some managers and consultants try to enhance their (illusion of) control over organizations by attempting to reduce chaos, while others accept and embrace chaos and base their change practice on it. This article focuses on the second group. Based on a study of literature and a series of interviews with experienced change consultants, a typology is developed, in which an enlightened modern, an ironic, and a postmodern way of coping with chaos in change processes is elaborated. The typology may help change consultants and managers with the development of their way of working and the articulation of their professional identity.

Introduction

‘O

ne wing-beat of a butterfly may cause a storm at the other end of the world. Is your commercial talent remarkable enough to make an enormous impact?’. With this slogan, printed on flyers and posters, Procter & Gamble tried to attract Dutch students to attend its 1999 business course. It used an icon of chaos theory, but turned its meaning upside down. According to chaos theory, butterflies may cause a storm, but not because they have a remarkable talent for causing storms and try to have meteorological impact. Through its ad, Procter & Gamble reinforced the widespread modernist view that changes in organizations are caused by actions of remarkable men and women who want to make a difference, despite chaos and contingencies. Managers, especially top managers, tend to treasure the view that their talents and interventions can have a significant impact on the future of an organization. They cherish their agency, or their ability to make a difference, and regard themselves as the principal change agents in their organizations. However, as change agents they are time and again confronted with the limited controllability of organizations, the complexity and indeterminacy of change processes and the uncertain and ambiguous effects of their actions. To cope © Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.

with these problems, managers increasingly hire change consultants to assist them in reducing chaos and realizing changes. What should consultants do if they are hired to assist managers in these change processes? They could try to reduce chaos and reinforce their client’s self-image, because managers have to manage, and therefore have to believe unconditionally in the effects of their actions. Or they could embrace chaos, advise modesty and help managers in a different way to change their organizations. This article focuses on consultants who have chosen the second route. These consultants do not try to reduce chaos for their clients in order to enhance their illusion of control. Rather they accept and embrace the chaos of change processes and base their consultancy practice on it. In this article, we develop a typology of change agents, presenting different ways of coping with chaos in change processes. This typology is based on a study of literature and a series of in-depth interviews with experienced change consultants. The typology may help change consultants with the development of their way of working and the articulation of their professional identity. In addition, it may be used by managers to reflect on their approach to organizational change and their identity as a change agent, as well as to find appropriate consultants to assist them in change processes.

Volume 12

Number 2

June 2003

122

CREATIVITY AND INNOVATION MANAGEMENT

The Illusion of Agency Managers tend to think highly of their own agency. Particularly when things go well, they are inclined to think that it was they who made the difference, while when things go wrong they blame external factors such as economic depressions, stubborn employees or just bad luck. The view of managers as agents, steering the course of history of an organization, is reinforced and cherished by businesspress journalists, management gurus and academics involved in the modernist management discourse. They attribute to managers the capacity to change organizations, and sometimes the world around as well, by purposeful action. The modernist view of change agents and change processes is as follows: based on an analysis of the opportunities and threats in the market and the strengths and weaknesses of their organization, combined with their experience-based intuition, managers create a vision and a hierarchy of strategic plans. The ‘vision’ is a statement of what is good for the organization, typically in terms of markets to serve, products to make, growth to realize and resources to employ. Once this vision is established, it soon acquires the status of a ‘grand narrative’ (Lyotard, 1984), stating what makes sense and what does not, defining success and failure, and what is to be regarded as progress or decline. Directed by this vision, managers make all kinds of plans, ordered hierarchically, to align the organization with the new vision. Once plans are established, managers direct the implementation in the organization, thus realizing the vision. Managers make a difference by creating a vision, making plans and implementing them. Consultants make a difference, in the modernist view, if they can contribute to this process. They may conduct SWOT-analyses, perform benchmarking studies, design elaborated plans, mobilize people to implement the plans, and help to prevent and overcome possible resistance in the organization. Based on their skills and expertise they contribute to the change process, reducing chaos and assisting the management to lead their organization to success, and thus showing their agency as a consultant. It is an illusion to think that this kind of agency is conclusive to change organizations. The impact of purposeful actions is uncertain and mostly limited, and the sense these actions make depend on the viewpoint of the one who judges them. The relations between visions, plans and effectuated changes in an organization are ambiguous, as is the role of managers and consultants. The

Volume 12

Number 2

June 2003

complexity and uncontrollability of social processes, the contextuality of order and the narrative character of agency make the modernist view of change processes and change agents appear as illusory. In order to elaborate this point more generally, consider the Battle of Borodino, as described by Tolstoy in War and Peace (Tolstoy, 1982; see also Latour, 1988). On 26 August 1812, a battle was fought between Napoleon’s troops and the Russian army, near the Russian village of Borodino. On and around the battlefield was a tangle of action. Soldiers acted in response to the concrete events that happened to them. Neither the causes, nor the effects of these events were known to them. They could not see the contours of the battlefield, they did not know the strategies of their generals and they had no idea whether they were winning or losing the battle. In the course of the battle it became clear that the actions of some small groups of soldiers were decisive, even though major charges were without any clear result. The commanders present, Napoleon Bonaparte and the Russian Field Marshal Kutuzow, had no overview over the battlefield and they were constantly troubled by contradictory messages and outdated plans of their general staffs. The messengers they sent out with orders for their men rarely arrived at the right spot, and if they did, their message was misinterpreted or ignored. Tolstoy described the battle as a muddled ball of fighting people, which neither commander could influence. The battle evolved unstructured, unpredictable and uncontrollable, and the influence of individuals with strategic and tactical plans was very small. Social processes, on battlefields as well as in organizations, are chaotic. Outcomes are emergent and rarely the result of someone’s great vision, well-elaborated plans and careful implementation efforts. The stream of events is without an a priori structure. During and after the action, people try to bring order into the events by telling stories about them: ‘Man . . . is the story-telling animal. Wherever he goes he wants to leave behind not a chaotic wake, not an empty space, but the comforting marker-buoys and trail-signs of stories’ (Swift, 1984, p. 62). In a story, an unstructured stream of events gets a beginning and an end, with certain persons and actions singled out. Separate actions are made visible and receive meaning in light of a plot. Causes are attributed to circumstances and people with intentions. A story tells whether something has changed, what has changed, who has made a difference and why. From the muddled ball of events and actions, storytellers pull threads and weave them into a meaningful whole.

© Blackwell Publishing Ltd, 2003

COPING WITH CHAOS IN CHANGE PROCESSES

‘Even though stories are the inevitable results of action, it is not the actor, but the storyteller, who perceives and “makes” the story’ (Arendt, 1958, p. 192). Stories are the accomplishment of storytellers and are not determined by the events. However, storytellers are to a certain extent bound to the rules of a discourse to tell their stories acceptably and convincingly. Reducing the complexity and contingency of the events by attributing agency to only a few purposeful acting individuals is a central rule in modernist discourse. As the mainstream of nineteenthcentury historians told their audience that Napoleon and Kutuzow won battles with their ingenious strategies, journalists, management gurus, executives and academics tell us that successful organizational changes are achieved through the brilliant strategies of talented managers, the Napoleons of our time. Formulated strategies and statements of CEOs are highlighted as the sources of success. Management storytellers reconstruct ‘actual’ strategies in hindsight from the changes that have taken place. And especially when managers and consultants anticipate a reconstruction, they may produce so many documents and statements, which may be interpreted in so many different ways, that it is almost always possible to reconstruct a suitable strategy or plan retrospectively. Managers may have different reasons for reinforcing modernist discourse and maintaining the illusion of agency. It helps them to justify their position to shareholders, so to mobilize financial and other resources for their plans. It fulfils the expectations of employees, external stakeholders and the public, gives them recognition, and strengthens their selfconfidence and self-esteem. Consultants may also have reasons to maintain the illusion of agency. It helps them to show their added value, since they can make a clear contribution to strategy formulation, organization design or implementation. Besides, one can earn a lot of money in this kind of management consultancy. But once a manager or a consultant has acknowledged the chaotic character and uncontrollability of change processes, the discourse dependent attribution of agency, and the limited and ambiguous effects of the process of making and implementing strategic plans, they cannot maintain the illusion of agency and have to find ways to cope with the unpredictability and complexity of the world, without the help of modernist storylines. Those consultants and managers cannot view a strategy as a ‘grand narrative’ anymore, the rationally established and indisputable basis for all justifiable plans. Strategies are fragile narrative constructions, and their

© Blackwell Publishing Ltd, 2003

123

definitions of progress and decline are contingent, bound to the specific context in which the strategy has been created. This also problematizes the hierarchy of plans and the implementation process. People with different opinions cannot simply be regarded as obstacles to progress. They view the process from a different context, which may be justifiable as well. Strategic change may involve the suppression of other people, while the justification for doing so can only be built, at best, on quicksand. To summarize the preceding, we formulate three characteristics of the change process that are especially relevant for the actions of change agents, viz. ‘chaos’, ‘attribution’ and ‘contextuality’. ‘Chaos’ is our shorthand to indicate that order emerges without there being an a priori structure, and that the actual changes are established in the complex course of the process. ‘Attribution’ indicates that the relation between effects and the actions of individuals is not a direct, causal one, but that this relation is created by storytellers, in stories where effects are attributed to actions and persons. ‘Contextuality’ means that the sense and the value of any action and any order is bound to a certain context and can be valued differently from different points of view. These characteristics show that agency is to a certain extent illusory, but also how an illusion of agency can occur. Agency is created in stories; it is not an ‘ability to make a difference’, which a change agent may or may not possess.

A Typology of Change Agents If ‘chaos’, ‘attribution’ and ‘contextuality’ are acknowledged, and not seen as something to be overcome or reduced, there are still several positions possible. Taking ‘contextuality’ as a starting point, three positions are possible: an enlightened modern, an ironic and a postmodern position. The enlightened modernist assumes that there are situations where the ‘contextuality’ of actions and order can be bracketed and where processes can be seen as determinate. In these situations, a modernist approach is thought to be possible. The ironist acknowledges ‘contextuality’, but participates in processes in which local and temporary certainties and stabilities may emerge (Rorty, 1989). The postmodernist sees everything as contextual and indeterminate – ‘anything goes’ – and refrains from attempts to bring about order. Taking one of these positions on ‘contextuality’ has consequences for the positions on ‘chaos’ and ‘attribution’. In practice, change agents tend to develop a coherent and consistent way of coping with all characteristics of

Volume 12

Number 2

June 2003

124

CREATIVITY AND INNOVATION MANAGEMENT

the illusion of agency. Enlightened modernists look for situations where they can set the illusion of agency aside. There they take normative positions, based on expertise and rational analysis, act intentionally to change organizations and claim the credits for successes. But they admit that it happens to be a fortunate accident that it is possible in some situations. Ironists see agency as an attribution after the fact, produced through actors who aim to make a difference. They can involve themselves in attempts to realize changes and claim credits without believing in the determinate character of such processes. Ironists show what they want to achieve and what they stand for, but they remain sensitive to alternative viewpoints and stories, and willing to be challenged in their temporary and local certainties. Postmodernists cherish the plurality of stories and views, and refrain from attempts to effect change by purposeful interventions. They only create reflexiveness among the participants in change processes.

Methodology This typology concerns both change consultants and managers who operate as change agents. To elaborate the types empirically, a series of in-depth interviews has been conducted with change consultants, in order to reconstruct the ways they work and cope, as part of their professional identity. We focused the interviews on change consultants rather than on managers, because we expected them to have elaborated a way of working and a professional identity, independent of the specifities of particular organizations. And since they are hired to assist in changing organizations, and have to win contracts, they have articulated their way of working and can present ‘who they are’. Furthermore, consultants are not only confronted with the illusion of their own agency, but also with the illusion of agency of their clients, and that makes it essential for them to be reflexive about the way they cope with illusions of agency. To acquire relevant data we interviewed seven senior consultants, who had shown in publications or otherwise that they recognized the problems and dilemmas with the illusion of agency. In the interviews, the consultants were not directly asked to state their way of coping with the aspects of the illusion of agency and their professional identity. We started with questions about their normal practice, asking them to tell typical stories about their way of working, through their own eyes, and through the eyes of clients and colleagues, since identities are constructed in and through concrete stories by different story-

Volume 12

Number 2

June 2003

tellers. Then, focussing on the agency-related dilemmas they did or would encounter, we asked them to reflect and to reconstruct their ways of coping. The following descriptions of the three types are illustrated with examples and (translated) quotations from the interviews.

Enlightened Modernists Enlightened modernists cope with ‘chaos’ by lumping and splitting. This means that they distinguish between chaotic, uncontrollable situations, in which they do not work, and controllable situations, where ‘chaos’ can be bracketed. In the latter, purposeful action is possible and results can be obtained. Enlightened modernists only engage themselves with problem fields where they can define the problem and the criteria for success, where they have a lot of experience and well-elaborated methods. There they have a reasonable chance for success. A consultant said, ‘I wouldn’t say to a client that our way is the only way, but we have experience with it and we achieve good results with it, and if that appeals to the client, we sell it’. In many cases, enlightened modernists have specialized in fields of organizational change with a history in industrial engineering, such as logistics or administrative organization. For enlightened modernists, it is of the utmost importance to be able to demonstrate that their actions have results, because their clients hired them for achieving results and will judge their added value on this. Therefore they try to obtain the role of expert in the stories of their clients and to receive the credit for (some of) the changes that take place in an organization. One consultant remarked that it is therefore important that some results and successes become visible in the short term. Otherwise, clients will probably forget the impact of the consultant. Enlightened modernists reinforce the view of the changeprocess as a rational design process, make a visible contribution to this process – preferably in the form of a written text with their name attached to it – and propagate stories about the process that acknowledge their agency. Enlightened modernists take normative standpoints towards their clients, telling them what is good for them, based on experience, rational analyses and knowledge of themarket, competitors and management literature. According to enlightened modernists, clients expect answers from them. Clients want certainty and consultants think themselves able to provide it. As one of the consultants said:

© Blackwell Publishing Ltd, 2003

COPING WITH CHAOS IN CHANGE PROCESSES

‘They know, because they tried to get things started for a while, that it is very difficult, that it will cost people their head and that careers are heavily dependent on the success of such a project. So what do they seek: certainty. They talk with people, and the consultant or firm that gives them the idea that “they can solve our problem” is hired.’ Enlightened modernists help create strategies with a clear notion of progress and success, which is not to be doubted within the organization, gives sense to all kinds of plans and justifies the implementation. Giving clarity and certainty is of the utmost importance. Not showing uncertainty is considered even more important. A consultant said: ‘One might ask: how sure are we [the consultants] that we are on the right track? Well, we do not know for sure that we are on the right track. But we are sure of this: it is better to have a somewhat inferior plan than no plan at all. So you should make a plan, and you can think it over a hundred times, and there will always be better plans, but in the end it is about creating certainty, giving everyone a marching route’.

Ironists Ironists cope with ‘chaos’ by using ‘guerrilla methods’. This means that they do not try to bring about change in organizations in a controlling manner, as do enlightened modernists, but by creating occasional incentives, tackling local problems and removing bottlenecks. Their work is not aimed at reaching a desired end-situation, but at escaping from an undesirable current situation. They initiate changes and start looking for solutions in constant interaction with their clients and other members of the client organization, mixing process-oriented and outcome-oriented techniques. Ironists have a more process-oriented view of organizational change than enlightened modernists. Organizations are seen as emerging patterns, potentially developed in a collaborative learning and sense-making process. Changing organizations is a prototyping process, a constant conversation with the situation, an alternation of doing and learning. In the process, a balance is sought between intervening and self-organization, between designing and letting emerge. One of the interviewees phrased his relation to design metaphorically: ‘I fly around “design” like a moth around a flame. I couldn’t do without it, because that

© Blackwell Publishing Ltd, 2003

125

would be the end of my profession, but I know that, if I come too close, my wings will burn’. Ironists are aware of the fact that agency can be attributed to people in different ways. Different stories are possible, but, nevertheless, ironists try to make their own story, looking for causes of success and failure with themselves and with the organization. They bring order into the situation, but recognize the fragility of this order. But if they would not try to do that at all, they would not be able to learn from their experiences. And especially in cases of assumed failure, this is important. Several interviewed consultants stressed that it is important to frame failure as a learning experience, because otherwise consultants would entirely loose their self-confidence and trust in their agency. Ironists have a different view of their impact to enlightened modernists. They think that their main impact is in initiating change processes, not in reaching objectives and achieving good results. The latter would not be possible for ironists because they do not know in advance where the change process will end. They help put things in motion and monitor constantly to see whether an organization is really changing. One of the consultants said about this monitoring: ‘Because of all kinds of circumstances a consultant may fail to get an organization moving, but it is inexcusable if he only finds out after a year and sending in a bill of a hundred thousand dollars’. Ironists may take a normative position towards their clients and tell them how the client should act. But they know and show the historical and paradigmatic context from which their standpoints originate. They base their normativity on local certainties, things they think to see clear in a certain place and time. But ironists are aware of the fact that they can be wrong and that they may do damage if they remain attached too strongly to their positions. They are flexible enough to adjust their standpoints if the situation requires it, or if their clients really disagree with them. As an interviewed consultant remarked: ‘It happens rather often that a consultant thinks “I know very well how those things go, so don’t nag about alternatives, we should proceed in that direction”. But it is part of your flexibility as a consultant to realize that what will really happen in an organization is not what the consultant wants, but what the organization wants’.

Volume 12

Number 2

June 2003

126

CREATIVITY AND INNOVATION MANAGEMENT

So ironists are more flexible in their opinions and actions than enlightened modernists, but they do have them, in contrast with postmodernists, who totally refrain from taking normative standpoints.

Postmodernists Postmodernists cope with ‘chaos’ by refraining from pursuing goals or initiating changes, thus avoiding the confrontation with the illusion of agency. It is the client who wants changes, who has to cope with ‘chaos’, not the postmodern consultant. The job of the postmodernist is to mirror the client organization in different ways, showing the premises of their problems and solutions, and to open up opportunities they had not seen themselves. Postmodernists may act as ‘reflectors’, asking questions to stimulate their clients to reflect. Or they may play the role of ‘provocateur’ or ‘wise fool’, who tries, by asking ‘naïve’ questions and by calling in question what is taken for granted, to challenge the way of thinking of clients, and to show them new perspectives. Genealogical analysis of strategies on a local level may be used as a technique to show clients their strategy in terms of its history and thus its ‘contextuality’. They may also use deconstruction techniques, showing the suppressed opposite that is always latently present in a strategy. If, for instance, a client stresses the bright side of a strategy, the postmodernist will show the dark side, if a client emphasizes order, the postmodernist emphasizes chaos. Postmodernists see that agency cannot be attributed unambiguously, and that impact or a lack of impact, success or failure are totally dependent on whom is asked at what moment. Therefore, it is not possible to make any clear evaluation of a change process. Postmodernists do not attempt to evaluate the effects of their work. It is sufficient if they have a good feeling about their work and if their clients keep coming back. Apparently, there are clients who think it pleasant or useful when a postmodernist mirrors their problems or brings up new perspectives. According to one of the interviewed consultants, his clients say he is ‘creative’. In this way postmodernists justify their way of working and their raison d’être. Postmodernists do not take any normative position towards their clients and the organizations. They help their clients by reframing their problems and showing possible solutions, not by telling them what to do. A consultant told us: ‘As a consultant, you can be useful by helping someone to get another perspective, or several other perspectives. And you

Volume 12

Number 2

June 2003

don’t have to tell what the correct perspective is, because there isn’t any. All perspectives are possible, but it is nice to see that some perspectives give someone a bad feeling and other perspectives give him a good feeling’.

Concluding Remarks Enlightened modern consultants, ironic consultants and postmodern consultants are apparently able to create added value for their clients. They have different ways of working, based on different assumptions, think differently about the status of their opinions and judgements, and expect (and get) different types of credit for their work. Table I summarizes their ways of coping with the illusion of agency. This typology makes explicit three aspects of the professional identity of change consultants and its internal coherence. Consultants develop their professional identity in concrete interactions with clients and colleagues. In practice, recurrent types occur, and by putting them in a typology, we have created three models that may help consultants to reflect on these interactions, and to articulate and further develop their identity. It may also make them aware that a positional choice on one of the aspects of the illusion of agency has consequences for their positions on the other two. This typology may also be used by managers who want to reflect on their personal way of working and their identity as a change agent, or those who want to hire an appropriate consultant to assist them in a change process. Managers tend to be modernists, but that does not mean that it is impossible for them to act as ironists or postmodernists in change processes. As postmodernists they might operate as if they were outsiders, and leave the change process mostly to the self-organization of their employees, while enhancing their employees’ reflexivity and creativity in that process. And as ironists, they might make the change process into a collaborative learning process with their employees and stakeholders, making the changes a joint responsibility. However, managers often have an all-encompassing responsibility for their organization, and the ironist and postmodern rewards for the initiation of change or enhanced creativity may not be enough for them. Besides, they have a history with an organization and cannot shape their identity as a change agent irrespective of their established managerial identity as a whole. Managers may compensate their own modernism by hiring ironic or postmodern change

© Blackwell Publishing Ltd, 2003

COPING WITH CHAOS IN CHANGE PROCESSES

127

Table I. Three types of change agents Enlightened modernist ‘Chaos’

‘Attribution’ ‘Contextuality’

• Bracketing chaos • Elaborate design methods • Credits for effects • Certainties • Occasional grand narrative

Ironist

• Fighting and embracing chaos • Collaborative learning • Credits for initiating change processes • Local certainties • Personal narratives

consultants. If they acknowledge the illusion of agency to a certain extent, by their own conviction or through a consultant’s persuasion, a cooperation with ironists or postmodernists may be very fruitful. An interviewed postmodern consultant told that he preferred working with modernist managers to ironists or postmodernists. Modernists are more likely to get stuck in their own frame of reference, because they do not see its ‘contextuality’ and are blinded for alternatives, and with them the added value of a postmodernist shows best. Postmodern consultants may also assist postmodern managers in their efforts to increase the reflexiveness of their organization, but a risk is then that consultant and client start deconstructing each other’s deconstructions and mirroring each other’s mirrors, leading to reflections ad infinitum and no changes at all. Ironic consultants usually work best with ironic managers. They need like-minded partners to walk the uncertain and ambiguous road of change. If ironists work with postmodern managers, who refrain from establishing any order, even locally and temporarily, they have to look for partners elsewhere in the organization. In such cases, consultant and client organization are cooperatively working on changes while being mirrored and deconstructed by respectively their client and their manager. Ironic consultants may also work with modernist clients, but then it is extra important to communicate what a manager can expect from them and what they expect from a manager. An interviewed consultant said that modernist managers tend to see ironists as cautious modernists, who just need more information of the situation to provide them with expert advice, certainty and guaranteed success. If the ambiguity and indeterminacy of the

© Blackwell Publishing Ltd, 2003

Postmodernist

• Avoiding chaos • Mirroring the client

• Credits for ‘good feeling’ or creativity • No certainties • Anything goes

process then prove fundamental, they are disappointed. For those managers an enlightened modern consultant would have been a better match. Enlightened modernists may also work for ironic or postmodern clients, but only in the rare cases where the clients agree that ‘chaos’ and ‘contextuality’ can be bracketed. In other areas, ironic and postmodern clients would consider modernist consultants arrogant, naïve or overconfident. In many cases, individual managers and consultants will not fit purely in one of the categories of the typology. They may, in fact, behave like chameleons, changing colour in changing contexts. In one context they may show themselves as enlightened modernist, while in another context they appear as ironist. Among the interviewed consultants was an enlightened modernist, specialized in manufacturing strategies, who turned into an ironist when he started to engage himself in making more general corporate strategies. A postmodern consultant, who had no certainties in the organizations of his clients, ‘changed colour’ into an ironist when he started to talk about the consultancy profession. An internal consultant, operating as an ironist, showed up as an enlightened modernist when he had to justify his work to the CEO of the company. ‘Changing colours’ can occur when the tasks are different, or when the ‘context of justification’ is different. If ‘changing colour’ becomes a conscious strategy, this might be seen as a fourth type: the chameleonic change agent. Such a change agent would need a broad repertoire of skills and stories, and sufficient flexibility to change at the right moment. On a meta-level, one may applaud such flexibility. A postmodernist who is willing to step from behind the mirror to aim for changes is taking up responsibilities.

Volume 12

Number 2

June 2003

128

CREATIVITY AND INNOVATION MANAGEMENT

But this flexibility may also be seen as inconsistency. If someone is known as a modernist and switches suddenly to an ironic or postmodern identity, or the other way around, she or he may lose credibility. ‘Changing colour’ can also lead to opportunism. Suppose that a consultant starts as an enlightened modernist, with a rational method, a perspective of bright results, and with a storyline to claim the credits for these results. But when things go wrong, she or he changes into an ironist, and says she or he only wanted to set the organization in motion, and it is the client, not the consultant, who is responsible for the results of the organization. The other way around, a consultant could start as an ironist, but, when things work out well, try to claim the credits for the success. This kind of chameleonic behaviour occurs, definitely, but it should not be considered a fourth type to be added to the typology. The ‘chameleonic change agent’ does not specify another way of coping with ‘chaos’, but refers to the consistency and flexibility of the change agent in action and identity. Referring to the Procter & Gamble advertisement, we could say that, in a way, change agents are like butterflies. They may have great impact, but then again they may not. They should not overestimate their ability to influence the success of organizations. The change agents we interviewed did not do so. They accepted that the effects of their actions are mostly limited, uncertain and ambiguous, and that they may not take their agency for

Volume 12

Number 2

June 2003

granted. Enlightened modernists can see the possibility of guaranteeing some success, but only in contexts where they think ‘chaos’ can be bracketed, and even then success is precarious. But that does not mean that change agents who accept ‘chaos’ cannot produce any added value for an organization. The opposite is true, but their added value shows in the change process itself and not in its results.

References Arendt, H. (1958) The human condition. The University of Chicago Press, Chicago. Latour, B. (1988) The pasteurization of France. Harvard University Press, Cambridge MA. Lyotard, J.F. (1984) The postmodern condition: a report on knowledge. Manchester University Press, Manchester. Rorty, R. (1989) Contingency, irony and solidarity. Cambridge University Press, Cambridge. Swift, G. (1984) Waterland. Picador, London. Tolstoy, L.N. (1982) War and peace. Penguin Books, Harmondsworth.

Klaasjan Visscher is Assistant Professor at the School of Business, Public Administration and Technology, University of Twente, The Netherlands. Arie Rip is Professor of Philosophy of Science and Technology at the School of Business, Public Administration and Technology, University of Twente, The Netherlands.

© Blackwell Publishing Ltd, 2003