EDITORIAL
223
EDITORIAL Strategic Flexibility
S
triving for strategic flexibility contributes to both incremental an...
17 downloads
333 Views
495KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
EDITORIAL
223
EDITORIAL Strategic Flexibility
S
triving for strategic flexibility contributes to both incremental and radical innovation, which are important prerequisites for sustained competitiveness. Technology based manufacturing and service industries are highly challenged by the dynamic tension between short and long-term effectiveness demands. Companies in these industries have to react very fast and have to perform according to many criteria at the same time. Competition and customers challenge them in terms of newness of concepts, functionalities and designs, high quality, low cost and short and reliable delivery times. Today’s operations have to be very effective, in terms of both customer demands and company profitability. Profitability, from the earliest stages of the new product life cycles, is a prerequisite for investments in new technology, new product and process development and companies’ competencies. In order to speed up this flying wheel and to shape a sound basis for future innovations and business success, companies need a strategically well-chosen ‘flexibility-mix’. Operational effectiveness, requiring operational flexibility in terms of for example being able to gather and rapidly respond to new technical and market knowledge as a project evolves, should go hand in hand with strategic flexibility to create room for new strategic options in future. Achieving strategic flexibility particularly also requires flexibility in different organizational functions, for instance in human resources policy and in product and process design. The combination of operational effectiveness and strategic flexibility is the focal point of the research program ‘Management of Innovation’ at the University of Twente. This program aims for the development of knowledge and the application of this knowledge in tools and managerial practices in the field of innovation management. A specific focus is put on research and product and process development, as the upstream stages in the production chain which are predominantly responsible for (technical) innovation. Close cooperation with technology-based industries is one of the preconditions in this program. In this special issue we intend to give an # Blackwell Publishers Ltd 2001. 108 Cowley Road, Oxford OX4 1JF and 350 Main St, Malden, MA 02148, USA.
impression of the research in this Twente research program on innovation management. Before introducing the separate contributions we will elaborate more on the concept of strategic flexibility related to innovation management. Creativity and innovativeness are rooted in highly competent and committed employees. But, to be innovative on the level of a project team, of an organization or on the interorganizational level of co-maker relationships, supply chain, network or even a virtual organization, an adequate organization design and leadership that fits to the situation is also needed. In other words, innovation has to be organized and managed adequately. At the same time, these organizational and managerial capabilities are a substantial part of companies’ competence base and therefore also the subject of processes of innovation and learning. Organization of innovation has to go hand in hand with innovation of organization, particularly of the organization of research and new product development functions. Strategic flexibility can be seen as a performance dimension referring to the readiness of an organization to adapt to, anticipate or even create future performance requirements. The need for strategic flexibility implies dynamics from within the organization: a firm’s goals and strategy may be changed proactively and with considerable frequency to improve performance, based on the development of (new) core competencies and the better development of key resources. Strategic flexibility as a prerequisite for operational effectiveness in the (near) future underlines the importance of searching for flexible organizational forms (innovating the organization), especially when looked upon from a social dynamical viewpoint. Given the need to perform under high time pressure, concerted action is required. This implies a framework of performance measurement embedded in effective feedback mechanisms. Design and implementation of such performance management is a difficult task in an environment characterized by creativity and professionalism. It can only be successful if it is based on a good insight in, and understanding of, social
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
224
dynamical processes, stressing room for creativity and learning, the importance of an innovative climate, professional autonomy and shared leadership, collective mind and personal development. Questions such as how to motivate people, how to stimulate them to cooperate within and between organizational entities, how to avoid the not-invented-here syndrome, have to be addressed. Designing the organizational structures and management systems have to go hand in hand with functional interaction patterns and style of leadership. The contributions in this special issue give an impression of the Twente research program on innovation management in which the above described issues are important characteristics. In all cases one or more of the authors are members of the Twente academic staff. The last article is written from the practice of designing a performance measurement system by two practitioners cooperating with two researchers from Twente and illustrates how tools and instruments are put into practical use. Hummel et al. show in a nuanced way the relevance of communication in interorganizational product development. Gomes et al. explore qualitatively as well as quantitatively the senior management support in the new product development process, both in the
Volume 10
Number 4 December 2001
Netherlands and in the UK. Van Riemsdijk and De Leede study flexible labour relations in the automotive industry. They executed four case studies together representing the production chain. The article is an example of the way the innovation of organization is explored. The article written by Fisscher et al is also focussed on the automotive sector. They explore the changes in the way the research and development function in a company is located and organized given huge changes in this function. In the last contribution Visser et al report from practice on the design of a performance measurement system for research and development at NIAB. In all articles strategic flexibility is an issue, explicitly or implicitly. Creating this strategic flexibility is enormously challenging. This is reflected in all five articles, where theory and practice are intertwined, where disciplinary based innovations go hand in hand with organizational and managerial experiments and renewal, where the individual and the collective performances are two sides of one coin. Learning and performing really need to be each other’s prerequisites: learning in performing and performing in learning in order to create operational effectiveness and strategic flexibility! Olaf Fisscher and Petra de Weerd-Nederhof
# Blackwell Publishers Ltd 2001
INTER-ORGANIZATIONAL COMMUNICATION
225
Types and Timing of Inter-organizational Communication in New Product Development Marjan Hummel, Wouter Van Rossum, Onno Omta, Gijsbertus Verkerke and Gerhard Rakhorst Managing the communication between the participants involved in inter-organizational product development is complex. The traditional models of new product development are not sufficient to gain insight in effective management practices in this respect. Our study explored the inter-organizational communication in a research and development project. Our results confirm Gersick’s model that looks upon new product development as being punctuated by periods of rapid change. In these periods, including the start-up, explorative prototype stage, and completion of the project, inter-organizational communication is essential about design objectives and project planning, contextual factors and the required resources, skills and knowledge.
Introduction
T
he development of innovative products can offer high financial rewards. However, development risks are high due to the imperfect knowledge about users’ needs, possible reaction of the competition, other technological solutions, and the required knowledge, skills and resources (Song & Montoya-Weiss, 1998). Alliances between users, knowledge institutions and industry are important to reduce these development risks (Schilling & Hill, 1998). External collaboration with users is valuable in order to gain understanding of their needs (Pitta & Franzak, 1996). Collaboration between knowledge institutions and industry yields partners knowledgeable in technological solutions and the market. It provides access to complementary knowledge, skills and resources, enabling functional specialisation. The specialisation of the participants involved in new product development increases the need for integration of their activities (Strieter, 1998). Managing the interdependent interfaces between these inter-organizational actors is complex. Accordingly, many inter-organizational development projects neglect environmental factors concerning the users, competitors, technology, intermediaries and regulation, # Blackwell Publishers Ltd 2001. 108 Cowley Road, Oxford OX4 1JF and 350 Main St, Malden, MA 02148, USA.
and therefore fail to reach their objectives (Perks, 2000). Other inter-organizational projects fail to focus on clear objectives and project planning, which will lengthen or even exhaust the collaboration (Bruce et al., 1995). These problems support the notion that successful product development is a dynamic process of mutual adaptive activities between the different participants in a development project and environmental groups, such as users, external technological researchers and competitors (Leonard-Barton, 1987). The participants need to be flexible to adapt the project to insights gained from the environmental groups, while remaining clear in goal setting and project planning. In our view, improved understanding of the dynamic interplay between the various groups is required to improve the manageability of inter-organizational projects. Stage-wise models of new product development, separating a project in the stages strategic planning, concept generation, pretechnical evaluation, technical development, and commercialisation (Crawford, 1997), help to understand the product development processes over the course of a project. Nevertheless, these stages do not explicitly focus on the processes of mutual adaptation between the diverse groups involved. Furthermore,
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
226
the models do not capture the dynamics of the overlapping, and iterative product development activities in dynamical work environments (Sonnenwald, 1995). Gersick’s empirically derived model incorporates iterative processes that are essential to inter-organizational product development in dynamical environments. It views product development as being punctuated by periods of rapid change (Gersick, 1989). The iterative periods of change are initiated by fresh ideas and information from environmental groups, and consist of matching the objectives and project planning with market opportunities and available resources. In our view, this model is valuable to studying inter-organizational product development. This model, derived from short-term task groups, has not yet been validated in the empirical context of interorganizational product development. Our study aims to validate a new model of interorganizational product development derived from Gersick’s findings. In our view, this model can be useful to support management practices in inter-organizational product development.
A Model for Types and Timing of Inter-group Communication in New Product Development A huge body of literature deals with the communication within research and development teams (e.g. Allen, 1993), between marketing and research and development, or users and research and development (e.g. Souder et al., 1998), and to a lesser extent between marketing, research and development, and production (e.g. Zirger & Maidique, 1990). Ancona & Caldwell (1992) divide inter-group communication into ambassadorial, task coordinating, and scouting communication. Ambassadorial goals include persuading others to support the team, and lobbying for resources. Task co-ordinating communication is aimed at planning communication including co-ordinating activities, reviewing product design, and development communication aimed at resolving design problems. Scouting communication includes scanning and collecting external ideas and information on the competition, market or technology. Ambassadorial and task co-ordinating communication are related to meeting budgets and schedules and realising innovative products. Scouting communication has adverse effects on performance if conducted in later stages of new product development (Ancona & Caldwell, 1992). Even though literature suggests that the timing of the types of inter-
Volume 10
Number 4 December 2001
group communication is relevant (Rochford & Rudelius, 1992), no longitudinal research has been conducted in this respect (Liker et al., 1999). Product development research emphasises that task co-ordinating communication focusing on setting objectives and project planning is essential during the start-up or fuzzy front stage of product development (Cooper, 1985). Contextual factors that influence the definition of product objectives for collaborative innovate product development include resource availability, external collaboration options, and information on competition and technology (Veryzer, 1998). Accordingly, ambassadorial communication and scouting communication appear to be essential in the fuzzy front stage as well. Defining successful objectives for collaborative product development is complex due to market and technological uncertainties. A period in which communication about external ideas, design objectives and resources is felt to be urgent is the period just after an explorative prototype has been built (Veryzer, 1998). This prototype is important to determine, in interaction with users, the suitability of technologies for certain product applications and to further investigate new or improved ideas and concepts. The results clarify product objectives and enlighten the need for complementary knowledge, skills and resources. We therefore hypothesise that partnerships with key users and development partners are likely to be initiated. So, in the explorative prototype stage, task co-ordinating communication focusing on objectives and project planning, combined with ambassadorial and scouting communication, provide essential opportunities to rouse the developmental activities. Gersick’s empirical studies on task groups provide guidelines for the appropriate timing of the types of inter-group communication in new product development. She found that projects consist of alternating bursts of radical change and periods of inertial change (Gersick, 1989). In the periods of radical change, including start-up and the temporal midpoint, groups alter their behavioural patterns, product objectives and project planning on the basis of their learning experiences and on fresh ideas from outside. The groups can achieve significant progress in the subsequent development activities when they have matched the objectives and planning to outside requirements and resources. Furthermore, the groups focus on outside requirements and expectations during the completion of the project. The relevance of external ideas, product requirements and resource acquisi-
# Blackwell Publishers Ltd 2001
INTER-ORGANIZATIONAL COMMUNICATION
tion suggests that scouting, task co-ordinating focused on objectives and planning, and ambassadorial communication is particularly appropriate during the bursts of radical change. During the completion of the project, inter-group communication about the objectives and planning appears to be appropriate. Complementary to these findings, Allen found that both successful and unsuccessful product development projects consist of two cycles of task co-ordinating communication focused on development activities (Allen, 1993). These cycles appear to co-occur with Gersick’s periods of inertial change. Namely, the second cycle in which communication about the development activities revived after the projects had entered their temporal final half. In successful teams, the design activities were discussed among technical experts on a rather continuing basis during these two periods, and not simply when serious design problems developed (Allen, 1993). On the basis of Gersick’s and Allen’s findings we hypothesise that successful inter-group communication in the front stage and the explorative prototype stage consists of relatively frequent scouting and ambassadorial communication, and communication about objectives and planning. In the periods alternating these two stages, it comprises relatively frequent communication about the development activities. Finally, before the completion of the second cycle of the development activities, communication about the objectives and planning revives.
Research Methodology Our study aims to validate our model of product development regarding inter-organizational projects. We compared the hypothesised patterns of successful inter-organizational communication with the actual patterns of the communication between industry, universities and a university hospital during their involvement in a joint project to develop a medical product. Furthermore, we analysed the relation of the congruencies and discrepancies between the patterns with their project performance. We counted per month the frequency of occurrence of ambassadorial, task co-ordinating and scouting communication topics we discerned in the written inter-group communication. Among the task co-ordinating topics we distinguished between topics related to the project objectives and project planning, and topics focused on development activities. We adapted the operationalization of the topics as defined by Ancona and Caldwell
# Blackwell Publishers Ltd 2001
227
to the context of inter-organizational collaboration. An example of a task-coordinating topic focused on project planning is: ‘‘Negotiate with others for delivery deadlines’’. To distinguish time-related trends among these communication types, we applied time series analysis. We used a running average over a time span of 6 months to smooth systematic fluctuations caused by face-to-face meetings that took place on average once per six months. We analysed the reliability of these patterns based on the correlation between the time-series of split samples of the written communication. By means of semi-structured interviews with key project members, including the project co-ordinator, a practising cardiologist, two academic developers, an industrial developer and a manufacturer, and the analysis of reports and minutes, we elaborated the extracted patterns of intergroup communication. In this elaboration, we distinguished time periods in which different types of inter-organizational communication dominated. In addition, the aforementioned sources were used to analyse the project performance and its facilitators and barriers. Project performance was derived from the probability that each cycle of development activities increased the likelihood that an innovative product would be realised that fulfilled user needs, and that these cycles were within time schedule and budget.
The Inter-group Communication in the PUCA-pump Project During the past 9 years, industry, universities and the university hospital have developed the new product under study, the PUCA pump. This intra-ventricular blood pump provides temporary support to a failing heart (Mihaylov et al., 1997). The final prototype consists of an extra-corporeally placed membrane pump, which is connected to a valved polyurethane catheter. By means of a guiding catheter, the tip of the catheter is introduced via the peripheral arteries or open thorax into the left ventricle. The pump, which is pneumatically driven, aspirates at least 3L/min blood from the left ventricle and ejects it into the ascending aorta (Verkerke et al., 1993). Figure 1 shows the time series of the written inter-group communication over the course of the PUCA-pump project. The correlation between split samples of the scouting, ambassadorial, planning, and development communication patterns are respectively 0.58, 0.93, 0.95, and 0.83. These correlations are significant for p50.0005. Particularly, the ambassadorial, planning and development
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
228
Figure 1. Time Series of the Inter-organizational Communication communication patterns appear to be a reliable representation of the written correspondence.
Start-up period and first cycle of development activities November 1990–March 1992 Our model presumes that successful interorganizational communication in the front stage consists of relatively frequent scouting and ambassadorial communication, and communication about objectives and planning. Corresponding to our model, scouting, ambassadorial and planning communication did occur relatively frequent in the first period of the PUCA-pump project. Scouting communication occurred more frequent per month than average from February 1991 to July 1991, and was mainly related to attending conferences about mechanical heart support. Triggered by these conferences and recent work experience, the project leader conceived his idea for a new mechanical heart pump, later to be known as the PUCA pump. Ambassadorial communication was conducted relatively frequently from February 1991 to March 1992, and was related to networking and grant applications. Former network contacts were renewed with groups, including the university hospital, a German and a Swiss university, a Dutch industrial developer of biomaterial coatings, and a Dutch industrial designer and manufacturer of medical catheters. EUREKA grants were allocated to the Dutch partners only. Planning communication was relatively frequent from April 1991 to March 1992, and focused on idea screening and the project planning to include in the grant applications.
Volume 10
Number 4 December 2001
The groups expected the catheter pump to be an innovative device to be applied by cardiologists, and decided on the main approaches to determine the feasibility of the catheter pump, and divided the corresponding tasks. Unpredicted by the model was that development communication occurred more frequently through July 1991 to March 1992. This communication was related to the overlap of a preceding heart-pump project and the PUCA-pump project. March 1992–November 1992 The start-up period is to be followed by a period particularly focused on development communication. Corresponding to our model, inter-organizational communication about the development activities occurred relatively more frequently per month from February 1992 to June 1993. As planned, previous animal tests with an alternative heart pump, post-mortem studies on humans, and numerical simulation of the PUCA pump were used to develop an explorative prototype. The valve system could not be integrated in the catheter as designed by the industrial manufacturer. Therefore, the development activities resulted in an explorative prototype of the PUCA pump, consisting of an alternative catheter with a separate inand outflow valve connected to a purchased membrane pump, once tested in an animal experiment. Not accounted for by our model, planning communication was relatively frequent from April 1992 to December 1992, and scouting communication from May 1992 to December 1992. The design objectives were discussed at the inter-organizational meetings, yet did not
# Blackwell Publishers Ltd 2001
INTER-ORGANIZATIONAL COMMUNICATION
change significantly. The project planning was revised due to diverse technical problems with the driving system for the PUCA pump purchased from an industrial supplier. Despite the lack of financial funds, the Swiss group planned to develop a driving system in-house. In addition to the communication at scientific conferences, which was not considered to have a clear impact on the project, scouting communication focused on finding new approaches to study the effects of the PUCA pump on the cardiovascular system. Performance This first cycle of the four categories of interorganizational communication largely followed the hypothesised patterns of successful inter-organizational communication. They evoked development activities that satisfactorily increased the probability that the PUCA pump would fulfil user needs and that were within time schedule and budget. According to the partners, sufficient time and effort was devoted to reduce the main application and technological uncertainties. Particularly the participation of users and developers in animal experiments with prototypes of an alternative heart pump and the PUCA pump satisfactorily contributed to essential learning experiences on the application of the PUCA pump. Accordingly, the early development communication during start-up was considered to be an effective deviation from our model. Some general factors related to the empirical setting were mentioned that influenced the patterns of inter-organizational communication during all periods of the project. The administrative procedures in order to receive grants were considered to have extended the objective-setting and planning stages superfluously. The project members saw the revision of the project planning during the development activities as the inescapable result of a dynamic group composition. Finally, evaluation and reward systems of universities induce academicians to focus on realising scientific publications. One manifestation of this focus is their regular presence at scientific conferences. The accompanying scouting communication was therefore increased repeatedly over the full course of the project.
Explorative prototype stage and second cycle of development activities November 1992 – October 1993 On the basis of our model we hypothesise that successful inter-organizational communication in the explorative prototype stage
# Blackwell Publishers Ltd 2001
229
consists of relatively frequent scouting and ambassadorial communication, and communication about objectives and planning. This period is to be followed by a period particularly focused on development communication. In addition, we hypothesise that before the final burst of development activities relatively frequent communication focuses on the objectives and planning. Corresponding to the model, relatively frequent scouting, planning and ambassadorial communication took place in this period. The relatively frequent scouting communication from January 1993 to August 1993 focused on scientific conferences. Planning communication was conducted more frequent per month than average from November 1992 to September 1993. Post-mortem studies, numerical simulation and an animal test revealed that cardiosurgeons and not cardiologists would be licensed to insert the PUCA pump in a patient, and that surgical insertion techniques needed to be developed. Moreover, the technical design of the pump system needed to be adapted to the medical conditions, and the valve-system to be redesigned to prevent biomedical problems. Ambassadorial communication was more frequent than average from April 1993 to October 1993. The project members actively searched for new industrial partners and financial support. Since the manufacturer focused on the market of cardiologists, this sole industrial partner withdrew itself from the project. Consequently, the project did not comply with the conditions for receiving new EUREKA grants. Furthermore, the contract with the supplier of the membrane pumps was cancelled because its product was considered to be technologically outdated and became insolvent. October 1993–October 1994 This period is to be followed as well by a period particularly focused on development communication. Inter-organizational communication about development activities did, however, not occur more often than average. During this financially unsupported period, the development activities were concentrated within the Dutch university, reducing the inter-organizational communication. They developed a combined in- and outflow valve, which turned out to be the most radical change of the PUCA-pump concept. In addition, the Dutch university developed a numerical simulation model of the cardiovascular system. Coupling this simulation model to the simulation model of the PUCA pump showed the need to trigger the driving system based on heart activities.
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
230
In contrast to our model, scouting communication was conducted more frequent from April 1994 to February 1995. Scouting stemmed from attending two courses about cardiovascular simulation techniques. The continuing ambassadorial communication resulted in including a surgeon full-time in the development group of the Dutch university, and contracting a new industrial partner in the field of heart assist and its supplier of membrane pumps. Consequently, EUREKA grants were assigned to the inter-organizational project to conduct a research study on the PUCA pump. The new industrial company supplied a driving system that could be triggered by heart activity. October 1994–July 2000 Corresponding to our model, inter-organizational communication about the development activities is conducted more often than average in the period May 1995 to November 1995. The Dutch university and the university hospital developed surgical insertion techniques and a heart-failure technique to create more realistic medical conditions in animal tests. The coating of the PUCA-pump catheter was enhanced by incorporating the technological advances as contrived by the Dutch industrial partner in the field of biomaterials. Successive series of students at the Dutch university have been deployed to develop and improve a production technique for the catheter, the catheter itself, and a valve system of satisfactory quality. Since these development activities were strongly concentrated within the Dutch university, interorganizational communication was drastically reduced. In order to steer the completion of the development activities, in 1997 the need was felt to evaluate the PUCA pump. The Dutch University applied for grants for a systematic evaluation of the pump. In 1998 the PUCA pump was assessed, which evoked final modifications to the PUCA pump and animal experiments with the PUCA pump. At present, nearly ten years after the project started, the PUCA pump is transferred to an industrial company founded by the project co-ordinator. Production techniques are being optimised to manufacture the final design of the PUCA pump at a larger scale. In contrast to our model, scouting communication, which was related to attending scientific conferences, was relatively frequent from September 1995 to February 1996. This type of communication was not considered to have an apparent impact on the course of the project. Another remarkable contrast with the hypothesised model is that the lead-time of
Volume 10
Number 4 December 2001
the second cycle of development activities is considerably longer than the lead-time of the first cycle. Performance The patterns of the inter-organizational communication in the explorative prototype stage and the second cycle of development communication resulted in development activities which satisfactorily increased the probability that the PUCA pump would fulfil user needs but which exceeded time schedule and budget. The PUCA pump’s target user-group had been revised, yet the pump was thought to fulfil the needs of cardio-surgeons to a higher extent than current competitors do. The revision of the design objectives in the explorative prototype stage in particular contributed to the attained quality of the PUCA pump. This change in the perspective on the market opportunity was not used as a motive to kill the project, but was incorporated in the project effectively. Factors that explained the outrun on time schedule and costs were related to the available knowledge, skills and resources, project planning, and design objectives. Due to declined financial grants for the German and Swiss partners and the withdrawal by the manufacturer, development activities were delayed and were not conducted by experienced partners with the most appropriate knowledge, skills and resources. In addition, the progress in these development activities was discontinuous due to a changing membership composition. Another factor is that unclear design objectives due to lack of assessment of the product requirements and product quality decreased progress in, for example, the design of the valve system. In addition, the project planning did not emphasise speed. Accordingly, the development of the heart failure model, cardiovascular simulation model, catheter machine, and surgical insertion techniques became time-consuming peripheral projects. Moreover, based on a trade-off between quality and development time, extra development time was spent to enhance the quality of the PUCA pump.
Conclusions and Discussion Corresponding to our model, the communication patterns in the PUCA-pump project show two periods focused on rapid change of design objectives and planning. Initiated by the first idea generation, the first period coincided with the lobby for co-development partners with valuable resources, skills and
# Blackwell Publishers Ltd 2001
INTER-ORGANIZATIONAL COMMUNICATION
knowledge, and financial grants. The newly defined objectives and planning initiated development activities that reduced market and technological uncertainties. The learning experiences of the development and application of an exploratory prototype initiated the second period of change of the objectives and planning. The most radical changes were related to market opportunity: refocusing from cardiologists to surgeons, task division: building a catheter machine in-house, and broadening the product focus: examining its influence on the cardiovascular system. The revised design objectives evoked the most radical change in the technical design: the combination of the in- and outflow valve. As soon as further external resources and knowledge were internalised, the design objectives and planning resulted in further development activities to enhance the prototype. Some differences between the actual and the hypothesised interactions can be attributed to the context of inter-organizational product development. These deviations need not to be interpreted as negative influences on project performance. Past and concurrent events may have an impact on design situations in an organizational setting (Sonnewald, 1995). We found that concurrent events, in our case animal experiments with an alternative heart pump, caused an effective deviation from our model; the additional development communication during the start-up phase. Furthermore, activities within an academic setting, including education and publicising, are interwoven with product development activities affecting the patterns of inter-group communication. Another deviation from our model is related to groups leaving and entering the project. The dynamical membership composition results in renewed discussions of at least the project objectives and planning and additional networking activities. The impact of the academic setting and dynamical group composition on product development performance is an interesting aspect to explore. Corresponding to the model, remaining deviations between the actual and the hypothesised patterns of successful interorganizational communication comply with the lack of development progress. Causes of the high development lead-time are inadequate resources, unclear design objectives and an unrealistic project planning. Unclear objectives and unrealistic project planning remain a common problem in inter-organizational product development (Bruce et al., 1995). Our results show that the different ‘predevelopment’ activities, including scouting for external ideas and information, lobbying for resources and support, and defining
# Blackwell Publishers Ltd 2001
231
objectives and planning, are strongly overlapping. This coincides with the notion that all activities in the fuzzy front stage are mutually inter-related (Khurana & Rosenthal, 1997). Accordingly, management needs to deal with these activities in an integrated way. For example, clearly defined objectives and planning is an essential input to smooth entry by new development partners in a project. In turn, these new partners may initiate valuable revisions of the design objectives. When the need for an appropriate development partners is met late, the fuzzy front stage is likely to be extended over a longer period of time, or even iterated. In the PUCA-pump project, the focus on design objectives and planning was iterated due to the withdrawal of the industrial supplier of driving systems. This emphasises the importance of an early yet critical screening and selection of product development partners. In addition, the team needs to remain flexible to adapt to unpredictable changing circumstances. The predevelopment and development activities are overlapping as well, though the prime focus of the inter-group communication alternated over the two categories. Besides overlapping, all activities turned out to be of an iterative nature. In contrast to the relevant literature our results show a second ‘fuzzy front stage’ at the time the main market and technological uncertainties were solved. Analogue to Veryzer’s (1998) suggestion that technology-driven processes of radically innovative product development processes precede or flow into more incremental product development processes at the time a representative prototype has been developed, we hypothesise that a suchlike transition occurs during the second fuzzy front stage, which coincided with the existence of a prototype appropriate for in-vivo testing. This finding indicates that effective management practices may differ significantly before and after the second fuzzy front stage. The PUCA-pump project shows that strategic planning is essential to reduce market and technological uncertainties. The development of peripheral technologies, such as the cardiovascular simulation model, can be time consuming. This particular simulation model revealed the need to redesign the driving system, or even to acquire another type. Accordingly, this example shows that the team should make well-deliberated choices about strategies to reduce the main uncertainties related to product design. These strategies should be based on a broad view of product quality, besides technical functionality it needs to integrate product performance in
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
232
the setting of application. This may prevent costly and time-consuming design revisions in a later stage. One strategy, which satisfactorily contributed to reciprocal learning experiences on the application of the PUCA pump, was the participation of users, including surgeons and cardiologists, and developers in animal tests of an alternative blood pump and prototypes of the PUCA pump. Early prototyping is an important manner to shorten development lead-time of innovative new product development (Mullins & Sutherland, 1998). Our results suggest that it can be an adequate trigger for the second fuzzy front stage in the new product development process. Even though flexibility in the product concept is important in the second development stage, continuous changes and revisions may delay inter-organizational collaboration (Bruce et al., 1995). Our case shows that clear design objectives are important to prevent or terminate development activities that do not enhance product quality. This may be particularly relevant in an academic setting in which the attendance to scientific conferences may lead to the regular input of new ideas. The project planning needs to be based on a comprehensive view of the design objectives that take into consideration the development activities on peripheral technologies, and the attainable knowledge, skills and resources. Gersick found that task projects conducted in fixed time periods were completed in a final burst of activities. Due to an unrealistic project planning, the PUCA-pump project did not have an absolute development deadline, and subsequently did not exploit this final period of rapid change. Instead, a tradeoff was made between quality and time overrun. These results suggest that deadlines, particularly before market introduction, need to be forceful and well chosen, in order to become an important tool for pacing new product development. Comprehensive design objectives need to steer the final burst of activities, in order to finalise the prototype. This model of product development deviates from the traditional product development models in some important respects. It shows that strategic planning, concept generation and pre-technical evaluation are strongly intertwined activities. Moreover, it stresses that in successful product development teams the project members are prone to reconsider decisions related to strategic planning and concept generation during the technical development of the new product. Creative reconsiderations of the product concept can significantly enhance new product performance (Stevens et al., 1999). An
Volume 10
Number 4 December 2001
appropriate period for this strategic flexibility is when the team members are likely to understand the meaning of contextual requirements on product outcomes and resources, to have enough time to make significant changes in the design of their product, and to feel the need for progress (Gersick, 1989). The latter represents a social consideration. Comparable to the implementation of new products (Richardson & Nigam, 1999), such social considerations can be relevant to the successful implementation of product changes. The mapped dynamics of inter-organizational communication confirmed the essence of our model of new product development. A limitation of our written communication analysis is that the intensity of the activities lagging behind the correspondence may vary strongly over the communication topics. In addition, the data are a subset of the complete sets of inter-group communication. Yet, the oral communication is generally represented in the written communication, such as in a written summary of the conclusions of a meeting. Moreover, interviews and reports did not reveal significant discrepancies between the graphically depicted and the actual communication trends. The graphically depicted communication reveals the interorganizational communication that had a clear impact on the course of the project. Corresponding to the essence of Gersick’s model of new product development, the communication trends reveal guidelines about effective management practices that focus on the timing of the communication types in inter-organizational new product development projects.
References Allen, T.J. (1993) Managing the flow of technology: technology transfer and the dissemination of technological information within the R&D organization. The MIT Press, Cambridge, UK. Ancona, D.G. and Caldwell, D.F. (1992) Bridging the Boundary: external activity and performance in organizational teams. Administrative Science Quarterly, 37, 634–665. Bruce, M., Leverick, F., Littler, D. and Wilson, D. (1995) Success factors for collaborative product development: a study of suppliers of information and communication technology. R&D Management, 25, 1, 33–44. Cooper, R.G. (1985) Selecting winning new product projects: using the NewProd System. Journal of Product Innovation Management, 2, 34–44. Crawford, C.M. (1997) New products management. Irwin, Homewood, IL. Gersick, C.J.G. (1989) Marking time: predictable transitions in task groups. Academy of Management Journal, 32, 2, 274–309.
# Blackwell Publishers Ltd 2001
INTER-ORGANIZATIONAL COMMUNICATION
Khurana, A. and Rosenthal, S.R. (1997) Integrating the fuzzy front end of new product development. Sloan Management Review, 38, 2, 103–131. Leonard-Barton, D. (1987) The case for integrative innovation: An expert system at digital. Sloan Management Review, 29, 1, 7–19. Liker, J.K., Collins, P.D. and Hull, F.M. (1999) Flexibility and standardization, test of a contingency model of product design-Manufacturing integration. Journal of Product Innovation Management, 16, 248–267. Mihaylov, D., Kik, C., Elstrodt, J., Verkerke, G.J., Blanksma, P.K. and Rakhorst, G. (1997) Development of a new introduction technique for the pulsatile catheter pump. Artificial Organs, 21, 5, 425–427. Mullins, J.W. and Sutherland, D.J. (1998) New product development in rapidly changing markets: an exploratory study. Journal of Product Innovation Management, 15, 224–236. Perks, H. (2000) Marketing information exchange mechanisms in collaborative new product development; the influence of resource balance and competitiveness. Industrial Marketing Management, 29, 179–189. Pitta, D. and Franzak, F. (1996) Boundary spanning product development in consumer markets: learning organization insights. Journal of Consumer Marketing, 15, 5, 66–81. Richardson, P. and Nigam, A. New technology introduction and implementation: the case of paging technology in the Ratlam Division of Indian Railways. Creativity and Innovation Management, 8, 4, 233–241. Rochford, L. and Rudelius, W. (1992) How involving more functional areas within a firm affects the new product process. Journal of Product Innovation Management, 9, 287–299. Schilling, M.A. and Hill, C.W.L. (1998) Managing the New Product Development Process. Academy of Management executive, 12, 3, 67–81. Song, X.M. and Montoya-Weiss, M.M. (1998) Critical development activities for really new versus incremental products. Journal of Product Innovation Management, 15, 124–135. Sonnenwald, D.H. (1995) Contested collaboration: a descriptive model of inter-group communication in information system design. Information Processing & Management, 31, 6, 859–877. Souder, W.E., Sherman, J.D. and Davies-Cooper, R. (1998) Environmental uncertainty, Organiz-
# Blackwell Publishers Ltd 2001
233
ational integration, and new product development effectiveness, a test of contingency theory. Journal of Product Innovation Management, 15, 520–533. Stevens, G., James Burley, J. and Divine, R. (1999) Creativity + business discipline = higher profits faster from new product development. Journal of Product Innovation Management, 16, 5, 455–468. Strieter, J.C. and Tankersley, C. (1998) How managers in high technology organizations perceive the usefulness of information shared during new product development. International journal of technology management, 16, 1–3, 144– 151. Verkerke, G.J., Muinck. E.D. de, Rakhorst, G. and Blanksma, P.K. (1993) The PUCA Pump: A Left Ventricular Assist Device. Artificial Organs, 17, 5, 365–368. Veryzer, R.W. (1998) Discontinuous innovation and the new product development process. Journal of Product Innovation Management, 15, 304–321. Zirger, B.J. and Maidique, M.A. (1990) A model of new product development: an empirical test. Management Science, 36, 7, 867–883.
Marjan (J.M.) Hummel is Assistant Professor in Management of Medical Technologies, at the faculty of Technology and Management, University of Twente, Enschede, NL; Wouter van Rossum is Full Professor of Innovation Management, Dean Faculty of Technology and Management, Dean Faculty of Public Administration, University of Twente, Enschede, NL; Onno (S.W.F.) Omta is Chaired Professor in Business Administration, at the Wageningen University and Research Centre, Wageningen, NL; Gijsbertus ( J.) Verkerke is Assistant Professor in Biomedical Engineering, at the Faculty of Medical Sciences, University of Groningen, Groningen, NL; Gerhard Rakhorst is Assistant Professor in Biomedical Engineering, at the Faculty of Medical Sciences, University of Groningen, Groningen, NL.
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
234
Senior Management Support in the New Product Development Process Jorge Gomes, Petra de Weerd-Nederhof, Alan Pearson and Olaf Fisscher This paper studies the relationship between senior management support to new product development activities by means of a quantitative and qualitative analysis of questionnaire and interview data collected in the United Kingdom and the Netherlands. The quantitative analysis showed that there is a small to medium association between senior management support to new product development and project performance in the dimensions of time, cost, and end product quality. The qualitative analysis suggests that these weak links could be explained by separating the influence of senior management support on new product development activities into direct and indirect effects. Direct effects include issues such as the use of multifunctional senior teams and process champions, whereas indirect effects include issues such as organization mission and goals, and learning and knowledge management systems.
Introduction
O
ne factor contributing to successful new product development activities is believed to be the support provided by senior managers to new product development teams (Brown & Eisenhardt, 1995; Maidique & Zirger, 1985). Despite the unanimous recognition that a supportive senior management team positively influences performance, there is a surprisingly lack of knowledge about what are its effects on different performance variables and how it takes place. The current paper addresses two related questions. Firstly it uses questionnaire data in order to examine the relationship between senior management support and various performance measures, namely time, cost, and end product quality. Secondly, the paper explores how senior management support to product development projects is felt by team members, further describing the ways through which such support occurs in innovative organizations. This question is investigated by means of a qualitative analysis to interview data collected with team members and project managers.
Volume 10
Number 4 December 2001
Senior Management Support in a product development context It is generally accepted by the current innovation literature that the organization and management of new product development is a key strategic issue in an increasingly global and fast-changing business environment (Griffin, 1997). Companies in almost all industrial sectors face the challenge of coping with rapid technological change, shorter product life cycles, and higher complexity in the business system, characterized by a move from products developed in isolation by research and development or marketing, towards projects carried out in a seamless global firm, with multifunctional teams involving internal and external (e.g. suppliers, customers and contracted out research and development or manufacturing) parties from several countries. Within this context of kaleidoscopic dynamics (Rogers, 1996), the drive is for truly innovative companies to develop unique- and customized-products delivered at low prices, with quality and before its competitors. This strategic reorientation with emphasis put on # Blackwell Publishers Ltd 2001. 108 Cowley Road, Oxford OX4 1JF and 350 Main St, Malden, MA 02148, USA.
MANAGEMENT SUPPORT IN PRODUCT DEVELOPMENT
uniqueness and innovativeness (Kumpe and Bolwijn, 1994) requires individuals and organizations to constantly create not only new and improved products and services, but also to recreate themselves through learning and knowledge. One of the factors with a strong impact on this process of continuous innovation (Bartezzaghi et al. 2001) is a company commitment to, and involvement with, innovation, which starts at the very highest level in the hierarchy. The key role of senior management in innovation success has been highlighted since the 1970s. In fact, both the SAPPHO (Rothwell et al., 1974) and the NewProd studies (Cooper, 1979; 1980) found that successful innovative firms have a top leadership committed to innovation and product development. Such commitment is defined by Brown and Eisenhardt (1995) as the provision of financial and political resources to the project team. McDonough (2000) suggests that top managers help cross-functional teamwork by a variety of means, such as demonstrating commitment, helping the team to surmount obstacles, making things happen, and providing encouragement to the team. Similarly, Emmanuelides (1993) proposes that development teams depend heavily on top management for acquisition of necessary resources, approval of design proposals, securing of required legitimacy, and delegation of necessary decision-making authority. Further, the author proposes that a product development strategy should state, implicitly or explicitly, the objectives of the development programme and the importance of performance dimensions for any given project. Song et al. (1997) found a positive impact of senior management support (measured as the promotion of team loyalty over functional loyalty) on interfunctional cooperation. Jassawalla and Sashittal (1998) say that the sense of urgency about product innovation and the priority conveyed by the organization to new product development are strong contributors to the financial success of the company. The authors define this high priority in terms of initiation and allocation of resources to product innovation activities. According to Dvir et al. (1998) and Hobday (1998), such support is important for all types of projects or products. The emphasis placed by the company on the pre-stages of the development process also reflects the supportive role of senior management. Gupta and Wilemon (1996) report the results of a survey to 120 technical directors from technology-based companies; they found that one key factor in successful innovation is senior management support
# Blackwell Publishers Ltd 2001
235
to various technical activities. Cooper and Kleinschmidt (1995) speak of a unified, clear and well-communicated new product strategy for the company and particularly for those involved in new product development. Similarly, in a comparison of four companies, Donnellon (1993) found that the most successful one had a conscious intent to develop a more entrepreneurial, collaborative culture, whereas the others focused solely on speed and efficiency goals.
Research Objectives The majority of the studies referred to above assume a positive impact of senior management support to new product development activities on performance. However, there is neither strong empirical evidence for this relationship nor does the existing literature specify how senior managers, which are committed to innovation, affect different performance measures. Also from the above discussion, it can be concluded that knowledge on the means used by senior managers to support product development activities is still very fragmented and disorganized. In fact, studies often seem to forget how the main actors feel and think about the issue; this leads to an incomplete and contextindependent picture of the phenomena under discussion. With the growing importance of innovation in the organization survival and development, a greater involvement of senior managers in the new product development process is likely to occur; as a result, the aforementioned problems suggest some important questions for research. The current study addresses two research objectives. The first objective is to investigate the relationship between senior management support to new product development and different performance measures. At the project level of analysis, the measures considered in this research were time, cost, and end product quality (Brown and Eisenhardt, 1995; Emmanuelides, 1993; Griffin and Hauser, 1996). It is expected that: H: Senior management support to new product development activities is associated to a) product development time; b) product development cost; and c) end product quality The second research objective is to investigate the means used by senior managers to encourage and promote successful new product development activities, as perceived by those involved in the innovation process, namely project managers and project team members. As this research question has
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
236
an exploratory and descriptive nature, no research hypotheses are formulated.
Method The study was carried out in two steps. The first phase aimed at answering the first research question, and involved a survey of 55 project managers working in 40 companies operating in the UK and the Netherlands in the following industries: chemicals, pharmaceuticals, electronics, telecommunications, toiletries and detergents, cosmetics, and domestic appliances. A structured questionnaire was used to collect information regarding the two variables of interest: senior management support and new product development performance. Respondents were asked to select a project that had been recently terminated and involved a multifunctional team (a team constituted by commercial and technical functions) and then respond to a number of sentences in relation to that project. The first variable was created based on four sentences taken from Cooper and Kleinschmidt (1995), to measure the degree of support to the project under analysis; with regards to performance, six items were adapted from works by Song and Parry (1992), and Song et al. (1997), to measure the degree to which the goals of time, cost, and end product quality were attained in the particular project (Appendix 1). All the variables were measured on a 5-point Likert scale, and reviewed by two academic peers for clarity and simplicity. Despite the small sample, both scales were submitted to exploratory factor analysis, which revealed two distinct factors for the senior management support variable and three factors for the performance variable. The performance factors correspond to the three expected dimensions of time, cost, and product quality; the two support factors were labeled Resources and Commitment. The first factor reflects the degree to which senior management makes available enough resources to product development; the second is concerned with the degree of commitment that senior managers have towards product development. A value of 1 in these two scales indicates that senior management did not dedicate enough resources or was not sufficiently committed to the particular project; a value of 1 in the performance scales indicates that the product: a) was launched before time; b) cost less than budgeted; and c) was of lower quality than expected. Appendix 1 shows the results of reliability, factor analysis, and sample comparison.
Volume 10
Number 4 December 2001
The second phase of this study aimed to answer the second research question and was based on 70 interviews conducted in the UK with project team members and project managers from the same type of industries used in the first stage. 25 companies are represented in this second sample. The collection of data was done through interviews focusing on the personal experiences of managers and team members involved in product development. Interviews were structured around the relevant topics for research, but unstructured as regards what interviewees chose to emphasise. The design of the interview protocol and corresponding topics for collecting information was the result of both literature review and two pilot case studies. Interviewees were asked to report their experiences in relation to recently finished or close to the end projects. The main informants were project leaders and team members, but there was also information collected via interviews with people responsible for the product development process, documents, and external sources (e.g. companies databases). The interview data was analysed through content analysis; the goal of the analysis was to explore and describe the actors’ views and feelings with regards to senior management support, and not to develop theory or confirm existing models.
Results The results are reported in two sub-sections, corresponding to the two research objectives.
Relationship between senior management support and new product development performance Multiple regression was the main technique used to analyse the relationship between the two variables. Several regression analyses were performed in search for the models with the best fit. The particular performance variable was the dependent variable, whereas Resources (Reso) and Commitment (Comm) were the two independent variables. Table 1 shows the main results. The table shows some interesting results. Firstly, the percentage of variance of the dependent variable explained by senior management support varies between 25% for product development time and only 1.3% for end product quality. In other words, the support given by senior management to the product development process is responsible for only a small part of the final outcome, with an impact on the time needed
# Blackwell Publishers Ltd 2001
MANAGEMENT SUPPORT IN PRODUCT DEVELOPMENT
237
Table 1. Influence of Senior Management Support on New Product Development Performance: Final Estimated Regression Models
PD Time PD Cost End product quality
Fit
Equation
Partial correl.
Simple correl.
R2 = 0.251 F(2,48) = 8.023** R2 = 0.159 F(2, 48) = 4.489* R2 = 0.013 F(2, 48) = 0.321NS
Time = 5.6970.33(Reso)70.26(Comm) (t =72.41*) (t =71.86NS) Cost = 5.3270.17(Reso)70.30(Comm) (t =71.13NS) (t =72.02*) Qual = 3.15 + 0.09(Reso) + 0.04(Comm) (t = 0.59NS) (t = 0.24NS)
Reso: 70.33 Comm: 70.26 Reso: 70.16 Comm: 70.28 Reso: 0.03 Comm: 0.12
Reso: 70.32* Comm: 70.40* Reso: 70.28* Comm: 70.35* Reso: 0.09 Comm: 0.08
* p50.05 ** p50.001 NS= no statistical differences found
to take a product from idea to commercialisation and on the costs of carrying out the project, and no significant influence with regards to the final quality of the product. These results suggest two implications. Firstly, the relatively small R2 values indicate that other variables not included in the study also account for developing a product within predicted time and cost. Second, the support given by senior management seems to have a stronger effect on those performance measures that are directly related to the new product development process itself, such as time and cost, than to one of its outcomes, such as end product quality. In fact, the analysis did not reveal any significant impact of the variable senior management support on the quality of the final product. Overall, these results support hypotheses a) and b), but fail to support hypothesis c). The simple correlations shown in the table confirm these conclusions: all the correlations between performance and the two dimensions of senior management support are medium and significant for time and cost, but weak and non significant for end product quality. On the other hand, the impact is more important on time (R2 = 0.251) than on cost (R2 = 0.159). When senior management gives the necessary support to a project, it is more likely that the particular project is developed within the predicted time than within the predicted costs. With regards to the differential effects of the two support factors on performance, the stronger impact of Resources (b =70.33) than Commitment (b =70.26) on product development time, means that when the goal is to develop a product within predicted time, senior managers give their support mainly through making available enough resources to a particular project; less important to developing a product within predicted time
# Blackwell Publishers Ltd 2001
seems to be the degree of senior management involvement on the project. This means that more resources pulled into a project in terms of available people is an important condition to deliver a project that meets the original goal of time to market. Conversely, when the goal is to develop a product within the predicted cost, the most important dimension is the level of senior management commitment to new projects and involvement in the Go/Kill and spending decisions for new projects (b =70.30).
Forms used by senior management to support product development projects Figure 1 summarizes the means used by senior management to support new product development projects as perceived by project team members. The figure shows the specific practices identified in the case material. These practices are briefly presented in the main text below with a few examples from the interviews. The specific practices identified in the data can be grouped into direct and indirect influences of senior management on projects. The first category defines those actions directly exerted by senior managers on the project. The second category specifies those actions, which have an impact on the context. The indirect influences can further be divided into organisation and process actions.
Direct influence of senior management on new product development projects The data shows that direct practices can take many forms: . Multifunctional senior teams that act as
gatekeepers and get involved in the project reviews and go/kill decisions at the vari-
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
238
Figure 1. Senior Management Support to New Product Development: Perceived Practices
.
. .
.
ous milestones of the new product development process; Steering committees to decide on idea prioritization, resource allocation, and other strategic decisions directly related to new product development; Joint leadership between senior managers and project managers; Direct communication channels promoted by flat structures which allow quicker and easier communication channels between team members and senior management; Process champions – these are different from project champions insofar that these are senior individuals that are responsible for implementing and maintaining a structured new product development process in place.
Indirect influence of senior management on new product development projects: organization level At an organization level, interviewees feel more support to new product development from senior management through a number of ways: . Mission and organization goals: one of the
strongest ways to emphasize a company’s commitment to innovation is through its mission and goals. Excerpts of the new mission statement of one company involved in the electronics industry reads:
Volume 10
Number 4 December 2001
‘To exceed our customers’ expectations through innovation, quality and total service (. . .) through high quality people and superior teamwork (. . .); to foster teamwork and individual achievement within a culture of empowerment and accountability’. Similarly, in one company manufacturing control systems, one of the goals is to develop products that were not in production two years ago, thus showing the firm’s strong commitment to innovation. . Strategy: increasingly companies are focusing on what the market needs and wants instead of developing technically perfect products for which little market application may exist or responding to sales people without first assessing the potential of a particular request. For example, after spending millions in a project, a company operating in the electronics sector decided to kill the project because a small competitor had arrived first in the market with a similar product. A decision was also taken to buy the company that actually arrived first in the market, therefore bringing in-house the rights for exploring commercially the new product. . Structural solutions: in some other cases, the strategic emphasis on innovation is translated into the company structure which, by its turn, sends a strong message in terms of the need to innovate. For example, the purpose of the newly created
# Blackwell Publishers Ltd 2001
MANAGEMENT SUPPORT IN PRODUCT DEVELOPMENT
Strategic Planning and Corporate Learning Department of a firm making components for the automotive industry is to boost innovation and bring unity to the way the company generates new ideas and develops projects. Similarly, one of the goals for the creation of one of the companies in the sample ( just one year previous to the interviews taking place) was to explore the market applications of a particular technical innovation.
Indirect influence of senior management on new product development projects: process level There are several examples of potential indirect actions carried out by senior managers to influence the way the new product development process itself operates: . Available resources: in virtually all cases,
the availability of resources was said to be an important condition to successful new product development. The more resources – people, money, production facilities, etc. – are pulled into a project, the likely it is to be developed within the initial goals of time and cost. Also in all cases interviewees suggested that this was one of the most difficult issues to tackle, as projects are not developed in isolation in the company; other business systems (e.g. hierarchical organization) and product portfolio management (e.g. existing product platforms) are just two of the constraints to take into consideration during resources allocation. . Brainstorming/idea generation sessions: in an interview published in a trade journal, the international technical director of a domestic appliances manufacturer said ‘What distinguishes [the company’s product development process] from other crossfunctional design efforts is that it involves marketing, design, and engineering before there is a firm product concept, never mind a timetable for the product’s development’. During the interviews conducted in this firm, it became clear that this quote was referring to the brainstorming sessions regularly held by the company. These special meetings are regularly held by people from the abovementioned functions and sometimes customers and senior management; the goal of these meetings is not only to generate ideas for new projects, but also to define the fit of a new project into the company’s new product development and innovation strategy. . Joint visits to customers: in a tandem interview with the Scientific Consultant
# Blackwell Publishers Ltd 2001
239
and a Senior Analyst in a chemicals manufacturer, it was recognized that, although technology-driven, the company – through its senior management team – is increasingly promoting joint visits to customers held by technical and commercial people. This was said to be important to getting technical people closer to the market and to making them more aware of commercial and sales issues. This is also important because, as said by a project manager, it allows technical people to challenge sales and marketing people’s assumptions and hence generate ideas for new projects. . Internal advertising of team achievements: an internal document provided by a gas producer is entitled European Innovation Awards. The document shows pictures of the teams that successfully developed a new idea in one of three areas: technical innovation, innovation in business, and innovation in operations technology. The Director of Innovation explained that this 8-page leaflet is sent to everybody in the company in order to publish the internal scheme of promotion of innovation. A hidden message for encouragement of teamwork can however be seen throughout the whole document, for example in the photographs and identification of the winning teams. . Learning and knowledge management systems: these are usually computer-based systems through which any employee in the company can submit an idea for a new project. At the firm making components for the automotive industry, a new intranet system had just been introduced, with the aim of boosting not only the number of ideas for new projects, but also the level of knowledge sharing in the company. The system architecture had been conceived by an employee who then had all the support of senior managers to develop the project into an Idea-Generation and Learning system.
Discussion and Conclusions Overall, this research has shown that senior management support to new product development projects is an important aspect for project performance, thus confirming the propositions of authors such as Cooper and Kleinschmidt (1995), Dvir et al., (1998), and Emmanuelides (1993). However, this work has also revealed that the magnitude of this relationship varies with the particular goal of the project. For instance, a strong senior leadership contributes more to developing a product within time than within predicted
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
240
costs. Furthermore, the quantitative data indicates that senior management support is not the only predictor of project performance, and is not relevant at all when the quality of the end product is the goal to attain. Other variables not included in this study help explain project performance, such as the use of multifunctional teams and the matrix structure (e.g. Brown and Eisenhardt, 1995; Griffin, 1997). These factors are well described in the literature, however less information exists with regards to how they affect the relationship between senior management support to new product development activities and project performance. The moderate and mediator effects of such factors can be addressed by future research. Notwithstanding these weak to mediumsize links between the two central constructs of this study, the qualitative phase of this research shows that senior management has a much wider influence on projects than the direct actions undertaken during project reviews or Go/Kill decision gates. In fact, the analysis suggests that senior management has an indirect influence on successful innovation and product development by operating upon the context – organization and process elements – on which project and team activities unfold. Some of the mechanisms shown in the previous section have already been addressed by the literature – e.g. available resources –, but others emerged in the analysis – e.g. learning and knowledge management systems – that recommend a further exploration of the relationship between these mechanisms and the concepts used in this study.
If on one hand the findings of this work call attention to future research lines with regards to new product development and support from senior management, on the other, they are also revealing for what they do not show. For example, the results show the absence of direct or indirect actions undertaken by senior management with regards to the development of joint reward systems. This contradicts authors such as Griffin and Hauser (1996), who have proposed that joint reward schemes are an important determinant of project success through its effect on team motivation and other individual and group variables. The majority of the organizations observed in the current research do not use group reward systems; instead they recognize and promote individual- and functional-related performance. This finding needs further research to clarify why senior management does not foster the use of group reward systems within new product development projects. To conclude, this paper has shown that the role and intervention of senior managers on the product development process is more complex than previous research has suggested. The increasingly strategic importance of innovation in the organization growth and renewal call for a deeper and wider understanding of the topics studied in this work. If the findings here reported have provided some answers, they have also raised some questions that need future attention from managers and academics.
Appendix
Table A1. Reliability and Factor Analysis of the Senior Management Support Scale Original item in the scale (based on Cooper and Kleinschmidt, 1995)
Empirical structure a
1. In this organisation, senior management devotes the necessary resources to achieve the firm’s new project objectives 2. New product development is being carried out with the necessary people in place, and they have their time freed up for new projects 3. Senior management is committed to new projects 4. Senior management is involved in the key Go/Kill and spending decisions for new projects
Number 4 December 2001
F2
.76 .94 .73 .94
Eigenvalue: % total variance: % cumulate variance:
Volume 10
F1
2.3 56.7 56.7
1.1 27.3 84.1
# Blackwell Publishers Ltd 2001
MANAGEMENT SUPPORT IN PRODUCT DEVELOPMENT
241
Table A2. Reliability and Factor Analysis of the New Product Development Performance Scale Empirical structure Original item in the scale (based on Song and Parry, 1992, and Song et al., 1997) 1. Compared to the initial goals, this project took a long time to develop 2. Compared to other projects, this project took a long time to go to market 3. Compared to the initial goals, this project had higher costs to develop 4. Compared to other projects, this project had higher costs 5. Compared to the initial goals, this project ended up by being of better quality 6. Compared to other products, this product is of better quality
Theoretic structure
F1
a
F2
F3 .95
Time .62 .86 Cost .88 .88 Quality .90 Eigenvalue: % total variance: % cumulate variance:
2.1 35.6 35.6
1.6 27.4 62.9
0.8 13.9 76.8
Note: item 2 does not load on factor 3, as expected. For this reason, the variable Time is composed of item 1 only, whereas Cost and Quality were calculated as the average of the two correspondent items.
Table A3. Sample Comparison Sample mean and (standard deviation)
UK (N=35)
Netherlands (N=20)
Support to new product development activities Adequate Resources Senior Management Commitment
3.2 (0.88) 4.1 (0.74)
3.1 (0.96) 3.7 (0.85)
Performance Time Cost Quality
2.7 (0.63) 3.1 (0.89) 3.7 (0.53)
3.0 (0.90) 3.1 (1.25) 3.2 (0.71)
t-test
0.76 1.63
71.41 0.20 2.96*
MannWhitney Z
0.73 1.44
1.38 70.21 2.82*
*p50.05
References Bartezzaghi, E.; Boer, H.; Corso, M.; Coughlan, P. and Gieskes, J.F.B. (2001) Managing Innovation as a Continuous Learning Process: New Challenges for Innovation Management. Paper presented at the Founding Conference of the European Academy of Management, Barcelona, April 19–21. Brown, S.L. and Eisenhardt, K.M. (1995) Product development: past research, present findings, and future directions. Academy of Management Review, 20, 2, 343–378.
# Blackwell Publishers Ltd 2001
Cooper, R.G. (1979) Identifying industrial new product success: Project NewProd. Industrial Marketing Management, 8, 124–135. Cooper, R.G. (1980) Project NewProd: factors in new product success. European Journal of Marketing, 14, 5/6, 277–292. 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, 374–391. Donnellon, A. (1993) Crossfunctional teams in product development: Accommodating the
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
242
structure to the process. Journal of Product Innovation Management, 10, 377–392. Dvir, D.; Lipovetsky, S.; Shenhar, A. and Tishler, A. (1998) In search of project classification: a nonuniversal approach to project success factors. Research Policy, 27, 915–935. Emmanuelides, P.A. (1993) Towards an integrative framework of performance in product development projects. Journal of Engineering and Technology Management, 10, 363–392. Griffin, A. (1997) PDMA research on new product development practices: updating trends and benchmarking best practices. Journal of Product Innovation Management, 14, 429–458. Griffin, A. and Hauser, J.R. (1996) Integrating R&D and Marketing: A review and analysis of the literature. Journal of Product Innovation Management, 13, 191–215. Gupta, A.K. and Wilemon, D. (1996) Changing patterns in industrial R&D management. Journal of Product Innovation Management, 13, 479–511. Hobday, M. (1998) Product complexity, innovation and industrial organisation. Research Policy, 26, 689–710. Jassawalla, A.R. and Sashittal, H.C. (1998) An examination of collaboration in high–technology new product development processes, Journal of Product Innovation Management, 15, 3, 237– 254. Kumpe, T. and Bolwijn, P.T. (1994) Toward the innovative firm – challenge for R&D management. Research and Technology Management, 37, 1, 38–44. Maidique, M.A. and Zirger, B.J. (1985) The new product learning cycle. Research Policy, 14, 299– 313. McDonough III, E.F. (2000) Investigation of factors contributing to the success of cross-functional teams. Journal of Product Innovation Management, 17, 221–235.
Volume 10
Number 4 December 2001
Rogers, D.M.A. (1996) The challenge of fifth generation R&D. Research and Technology Management, 39, 4, 33–41. Rothwell, R.; Freeman, C.; Horsley, A.; Jervis, V.T.P.; Robertson, A. and Townsend, J. (1974) SAPPHO updated – Project Sappho phase II. Research Policy, 3, 258–291. Song, X.M. and Parry, M.E. (1992) The R&DMarketing interface in Japanese High-Technology Firms. Journal of Product Innovation Management, 9, 91–112. Song, X.M.; Montoya-Weiss, M.M. and Schmidt, J.B. (1997) Antecedents and consequences of cross-functional cooperation: a comparison of R&D, manufacturing, and marketing perspectives. Journal of Product Innovation Management, 14(1), 35–47.
Jorge Gomes is Assistant professor at the Superior Institute of Applied Psychology (ISPA), Lisbon, Portugal and Training consultant in Human Resources Management, Nova Forum, Lisbon Nova University. Petra C. de Weerd-Nederhof is associate professor New Product Development at the faculty of Technology and Management, University of Twente, The Netherlands. Alan Pearson is Emiritus Professor of R&D Management at Manchester Business School and Visiting Professor at the faculty of Technology and Management, University of Twente, the Netherlands. Olaf A.M. Fisscher is Full Professor of Quality Management and Business Ethics at the faculty of Technolgy and Management, University of Twente, the Netherlands.
# Blackwell Publishers Ltd 2001
FLEXIBLE LABOUR STRATEGY
243
Flexible Labour Strategy in the Dutch Automotive Industry Maarten van Riemsdijk and Jan de Leede Labour flexibility is a major way for companies to become more flexible. Why companies use flexible labour relations varies widely per industry. We assess the development of labour flexibility within the Dutch automotive industry. Four cases, together representing the production chain, are presented. We show how and why these companies arrived at the high level of sophistication in labour relations they currently have and what combinations of internal and external, numerical and functional forms have emerged. The process has been one of trial and error, characterised by emerging rather than deliberate strategy. It has a clear pattern over time. In three out of our four companies, an innovative labour use strategy emerged, finely tuned to market demands, new institutional realities and specific company needs.
Introduction
O
ver the past decade and a half, labour flexibility has spread rapidly throughout Europe in different forms, co-varying with different economic and social systems (Blanpain, 1999; Pollert, 1991; Looise et al., 1998). In all European countries, part-time labour has grown substantially (Brewster et al., 1997). The Netherlands is ahead of all with 29% of its working population working part-time (OECD states even 36%). But in the past 15 years, just like in other countries, the Dutch have experienced a rapid increase of temporary, on call and agency workers as well. In 1996, a total of 16% of Dutch employees had an a-typical contract. Seven percent worked on a temporary contract, six percent were on call and another three percent were agency workers (Bolhuis van, 1996). In 2000, the growth has stabilised, due to a tight labour market. It is conventional wisdom that the need for flexible labour relations arose out of less predictable demand, more volatile markets and fiercer competition. Some put strategic behaviour by management to regain control, a new market oriented ideology or even the effects of post-Fordistic production systems into the equation. These are broad generalisations however and it is clear that there are large differences between industries and at company level (Bolhuis van, 1996; Marginson, # Blackwell Publishers Ltd 2001. 108 Cowley Road, Oxford OX4 1JF and 350 Main St, Malden, MA 02148, USA.
1991). From statistics, we know that especially temporary contracts and agency workers are very popular in manufacturing. We do not know why. It might be that Atkinson (1984) is right with his famous core-periphery model. He predicted that flexible companies will employ a permanent core of highly skilled and functionally flexible employees that they take good care off. Next to this ‘elite’ they will have one or more layers of numerically flexible employees that are offered fewer opportunities and rewards, being less important for core business. It is equally possible that Ackroyd and Procter (1998) are nearer the mark when they state that in manufacturing a new flexible firm emerges, in which core and periphery blend together with a loss of prerogatives for skilled workers. Higher efficiency and financial dictates lead to more technology, leaving simpler tasks for less skilled people. In spite of all macro-oriented research, we do not know the specific configuration of flexible labour that manufacturing companies use, that is, the combination of internal and external, numerical and functional forms of flexibility. Why they do so is equally unclear. It can be (management-) strategy, ideology, institutional arrangements, market forces, customer demand, competition, opportunity or mimicry. Therefore, we performed case studies in four Dutch automotive companies, using this
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
244
industry as a sub-sector of manufacturing. This is a relatively small industry in the Netherlands and as such it is comparably easy to overview. It has low to medium skilled jobs, facilitating all different forms of labour flexibility. The four companies taken together represent all levels in the production chain. Two are OEMs, one is a first-tier main supplier and one a second-tier jobber. We had access to these particular companies and two of them were in the forefront of labour flexibility developments in the Netherlands. These are the reasons why we chose for these particular cases and for Dutch automotive. We pose two research questions: 1. What configurations of flexible labour do Dutch automotive companies use. 2. Why and how did they arrive at these specific configurations. After an overview of the industry and its use of flexible labour, we will present the four cases. We give general characteristics, the configuration of flexible labour that has been developed and the reasons we found per case. We then discuss key issues from the cases and finally confront our results with the two main strategic theories outlined above and some other grand generalisations within human resource management literature.
The Dutch Automotive Industry The total number of people employed in the Dutch automotive industry in 1999 was only 35,000. There are just three main manufacturers of cars and trucks. Annual automotive production in 1999 was 280,000 cars and 55,000 trucks. In addition, there are some producers of coaches. The total number of employees involved in production is around 18,000 (cars 7,200, trucks 8,000 and busses 2,500). The automotive industry has very distinct characteristics. The FIER report (Moerman et al., 1996) discriminates between
different market demands for different positions in the supply chain. Original equipment manufacturers (OEMs) at the top of the chain control the industry by setting the standards of performance and quality. The main (first tier) suppliers and the co-suppliers (second tier) have to meet high criteria on quality, delivery time and price. For quality, a parts per million standard is common, for delivery times, just in time delivery is widespread. Local first and second tier suppliers compete internationally as geographical location seems to be less and less important in selecting suppliers for OEM’s. Jobbers, low in the supply chain, will be selected mainly on price. As a result, for them labour costs are extremely important. Current figures on employees with a fixed, temporary or agency contract are not available for the industry. Nevertheless, based on an older labour market study of the employers’ organisation we know that in 1995, 85% of the companies within the automotive industry used agency workers and 37% used temporary contracts. Temporary contracts amounted to 7% of the total work population, for agency workers, this was 8% (FME, 1996). These are averages; the level of especially the agency workers differs between seasons. The most common internal types of flexible labour are shift working (20% of the employees) and part-time jobs (7%). In Table 1 these figures are presented and compared with the metal industry and the national average.
Four Cases Method The cases reported here were studied in the 1998–2000 period. In all cases semi-structured interviews were used, based on standardised protocols, with specific questions added
Table 1. The Spread of Flexibility in the Metal and Automotive Industry in 1995 (FME, 1996)
Volume 10
Temporary contracts
Agency contracts
Shift work
Part-time
Pct. Of companies in metal industry Pct. Of companies in automotive
58% 37%
85% 85%
40% n.a.
86% n.a.
Pct. Of employees in metal industry Pct. Of employees in automotive
4% 7%
8% 8%
20% 20%
5,5% 7%
Pct. Of employees in the Netherlands
8%
3%
7%
25%
Number 4 December 2001
# Blackwell Publishers Ltd 2001
FLEXIBLE LABOUR STRATEGY
depending on the respondent’s position. We interviewed the human resource manager and the production manager, two team leaders, at least two permanent staff and at least two temporary staff. The works councils were interviewed and in two cases, the manager of the in-house agency as well. In addition we performed document analysis of internal reports, magazines, social reports and statistical data. Here, we present short descriptions of the cases, mentioning only three topics: 1. general company characteristics (size, product, market, strategy, developments), 2. the configuration of flexible labour (types and degrees of flexible labour, and some developments) and 3. the reasons why these configurations developed.
Case A: Original Equipment Manufacturer of Trucks General Characteristics Company A has recently evolved from a full production plant into a final assembly plant of an international truck-company. It is an original equipment manufacturer and an assemble-to-order operation. End of 1998, the company employed about 2200 employees, of which 525 (24%) were flexworkers. The size in terms of units rose from about 7600 in 1993 (a dip, due to the general economic recession) to 19.000 in 1998. The total staff in the same period increased from 1700 to 2200. This seems a small increase, but remember the site changed into an assembly plant. The strategy of the company is largely set by its foreign-based parent company. Two strategic issues are important. The first is that management wanted to react as quickly as possible to market demands. They first opted for a daily planning, but it was soon clear that this could not be sustained and should be replaced by week-to-week planning. The second strategic issue was the corporate decision to centralise the production of subassemblies like axles, motors and cabins. Part by part, these shops were closed down at company A. In addition a sister company in the Netherlands, a producer of cabins for company A, was marked to close down. Many of the 500 workers of this plant had to be relocated over a period of three years to company A, as forced lay offs had been successfully contested by the unions.
245
this proportion is even higher, as it is restricted almost solely to the lowest level of job complexity where training periods are short. The company also uses several internal numerical types of flexible labour (working hours bank, two-shift system). Finally, there are internal functional flexibility forms (teamwork in final assembly). This configuration is the result of a long process of trial and error. Reasons for Flexibility One reason for the relatively large number of external flexworkers is the company’s intent to avoid a new mass lay-off of hundreds of permanent staff. They went through that expensive and emotionally difficult experience during the recession of 1992/1993. A second reason is the closure of the sister plant and the reduction of sub-assembly production within company A. They have decided to replace these ‘threatened’ jobs with flexible labour, in order to relocate permanent staff into more secure positions. The works council happily agreed to this policy. Due to serious production problems as a result of very shortterm temp hiring practices, the company decided to go into long-term temporary contracts and secondments. Short-term contracts led to bad selection and training, high turnover and loss of production quality. It also placed too high a burden on permanent staff. Internal numerical flexibility forms, like the working hours bank, spurred from the wish to adapt to changes in market demand easily, at low costs. In 1994 this time-for-time system replaced the overtime policy of the years before. With increasing market demand, the working hours bank reached its limitations and in 1995 the company introduced a two-shift system to enhance production capacity. The strategic shift towards a more flexible production with high production quality standards was the main reason to enhance functional flexibility of permanent staff by using teamwork. These teams use internal job rotation for all team members, making them able to assemble different variants of trucks while educating and training individual members at the same time to meet quality standards.
Case B: Original Equipment Manufacturer of Trucks General Characteristics
The Configuration of Flexible Labour Temporary workers and secondment contracts (external numerical flex) make up 25% of the total workforce. In direct production
# Blackwell Publishers Ltd 2001
Company B is a full production (both components production and final assembly) subsidiary plant of an international truck-company. In Holland, it builds highly customised trucks
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
246
on a make-to-order basis. About 2000 people are directly employed in production. After severe problems previous to 1994, production was back at 25.700 in 1995 rising to 33.770 units in 1998. As the company is only a small plant in a much larger corporation, overall strategy is set by the foreign owned parent company. Two strategic developments are important. One is that the company virtually had to make a re-start in 1995. The other is its maketo-order strategy. At first it set a standard of a maximum 7 weeks delivery time for all products. It was soon apparent that this standard was way too high. They changed delivery times to 10–12 weeks. The effect of this change was that the enormous fluctuations in demand experienced earlier are now more predictable and can be smoothened out. The Configuration of Labour The company has a very firm policy on the ratio permanent/flexible production staff. It wants 1600 permanent production employees at a production rate of 96 trucks a day. The actual amount of production personnel at that rate should be 2000, which leaves 400 external numerical flexible employees in production (20%). It has internal numerical flexibility in its shift system. The company wants to use internal functional forms of flexible labour, but finds this difficult, as people are unwilling to change workplaces, let alone departments. It does use teamwork however, with dedicated teams working on clear tasks. There was a very rapid increase in production personnel needed in 1997 (700 FTE = 1700 people), which was extremely costly. It was mostly agencies doing the recruitment from Belgium, England and the Netherlands. In 1998 too many flexible personnel (up to 35%) in production led to major problems in quality and turnover; it placed too high a burden on permanent staff. High costs, too high numbers, bad selection and recruitment, high turnover made company B weary of letting agencies run the show. ‘Company B behind the wheel’ is a free translation of the slogan as far as new flex policies are concerned. In 1999, the company has set up its own Job Centre, together with two in-house agencies. It is the company itself however, that decides on policy within this Job Centre, hence the slogan. The Job Centre will employ all flexible personnel. The deal for the future is that the Job Centre can have up to 30% temporary agency workers, 70% of its workforce must be on fixed term contracts or be employed by one of the participating agencies.
Volume 10
Number 4 December 2001
Reasons for Flexibility The current situation has been arrived at by experience, trial and error. Serious problems during the last recession (1992–1993) forced a major reorganisation of company B. As a result, it has firmly decided on a safe level of permanent staff. The main reason is the trauma of having to fire all personnel and the subsequent policy that permanent jobs should really be permanent. That is an image consideration. Flexible personnel are therefore evidently used as a buffer against hard times. Flexible staff will now only be recruited and employed by their job centre. Bad results as a consequence of using too many agencies and too many flexible staff led to the new recruitment strategy.
Case C: Main Supplier General Characteristics Company C is a production plant of an American parent company with products in four markets. Two of them (truck and car industry) together account for more than 80% of its sales. The company is a first tier (main) supplier for company A and to several car manufacturers. The company employs about 350 people (end of 1998), up to 30% are flexworkers. The company experienced a dip, both in sales and in number of employees, during the recession of 1992–1993. After this, it grew steadily towards its current size. The market demands on product quality, delivery time and reliability are very high and strict. For instance, quality standards are up to 53 parts per million in car supplies. Delivery times are sometimes only two days and just-in-time delivery is essential. There is a seasonal cycle (mainly spring and summer) for its car-products. The strategic choice to stay in these markets and to meet these demands marked a major break in the history of the plant. Part of this strategic choice is the decision to upgrade the engineering department to be able to be a co-developer. The engineering department nowadays employs about 50 people of mainly higher technical vocational education. The production system that was re-designed by these engineers is as simple as possible, providing maximum mix and volume flexibility at lower work floor level. Jobs in the production department consist of basic assembly tasks; workers just need a fine locomotion and high precision. The Configuration of Flexible Labour External numerical flexibility accounts for up to 30 % of the workforce. Especially during
# Blackwell Publishers Ltd 2001
FLEXIBLE LABOUR STRATEGY
the seasonal peak the proportion of agency workers is high. Almost all of them perform tasks of low complexity. Company C also has a small labour pool with another company that has a contrary season. Internal numerical flexibility is overtime and a flexible shift system. With the approval of the unions and the works council, the company can switch from a two to a three-shift system on a week’s notice. In addition, some experienced people rotate among the truck and the car department. The company is trying to enhance the quality of the workers to be able to introduce semi-autonomous teams. Reasons for Flexibility The main reasons for external numerical flexibility are the seasonal production peak and the wish to avoid a costly mass lay-off during recessions. Necessary better production engineering has led to simpler tasks and consequently lower production costs and more numerical flexibility opportunities. The reasons for upgrading the flexible labour policy were the initial high turnover and poor quality of temporary workers. Now there is an in-house agency, which is responsible for selection, appraisal and development of agency workers. The recent emphasis on internal flexibility stems from the observation that only qualified and committed workers are able to meet the levels of production quality required of a first tier supplier.
247
overtime and/or stockpiling are expensive solutions. So far, company D has been able to hold this position. The Configuration of Flexible Labour About 15% of the workforce are agency workers. In exceptional cases outsourcing of parts to a selected number of sub-contractors is used. The company also has internal numerical flexibility. It uses a two-shift system that can be extended into a three-shift system in busy times. In addition overtime is used, but sparingly, because of the disadvantages of stress, higher absenteeism and poor quality. At company D, functional flexibility is limited. New agency workers have to work on two machines (of the total of 100). Permanent employees are specialists on a small group of machines. Reasons for Flexibility The reasons for the external types of labour flexibility are the seasonal production peaks, and the need to adapt to just-in-time delivery and just-in-time production. An important reason to use an agency is also to minimise recruitment problems and selection risks. The agency workers have in effect a long probation period. In addition, the agency performs the role of the personnel department (in particular recruitment and selection). Part of the seasonal production peak is addressed by internal flexibility: the flexible shift system in combination with overtime.
Case D: Jobber General Characteristics Company D is a metalworking company and one of the suppliers of company C. About 70% of its sales are generated in the automotive industry. It is a rather small company with little specialisation and no personnel officer. End of 1997, the company employed about 100 people, of which 15 were flexworkers (15%). The company shows a slow but steady growth since the economic slump of 1992. The batch size differs from one unit to large batches of 500.000 units. Company D is a jobber with about 10 big and numerous small customers. The seasonal peak in production is more or less the same as in the automotive in general: the first two quarters of the year. The just-in-time production and delivery requirements of the companies higher in the chain affect company D. It does not give in to the pressure however, but has pointed out that more flexibility is impossible, as qualified flexworkers in the metalwork industry are hard to find, and
# Blackwell Publishers Ltd 2001
Discussion There are some preliminary observations to be made on the use of and the reasons for the configurations of flexible labour that we found so far. After that, we will discuss the results in the context of some of the existing ‘flexibility literature’.
Preliminary Observations A. Flexibility of labour is an important issue within the automotive industry. Every company uses labour flexibility strategies in order to adapt in time to market and customer demands. This is hardly surprising, as the companies clearly operate in an industrial chain structure. Some of the strain is passed down the chain. Clearly, there are historical reasons as well, all experienced mass lay-offs. B. We are not able to describe a typical and distinct configuration of flexible labour for the automotive industry, although there are
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
248
similarities between companies. One observation is that the large original equipment manufacturers rely more heavily on external numerical flexibility than the smaller companies down the supply chain. But then, larger companies tend to employ much more flexible staff than smaller ones (Bolhuis van, 1996), and therefore this cannot be ascribed to sector influence alone. Most of the external flexworkers are employed at jobs with relatively low levels of complexity that require short periods of training. That means that the high numbers of external workers are found almost solely in assembly. In our fourth case, the number of agency workers is limited because of the tight labour market and because of the nature of the job. Metalwork requires specialised skills. C. Though there is not one distinct configuration there are similarities also in the historical development of the configuration in each company. We can observe a particular learning curve (Figure 1) After serious economic problems and mass redundancies the economy picks up. 1. If production grows, working overtime is the first step. 2. If this does not suffice, very short-term external flexible labour is hired. Above 20–25% of external flexibility in production, problems emerge in quality and quantity of people. Also in product quality and delivery reliability and in the
pressures brought to bear on permanent employees. 3. In a next step, the emphasis shifts to longer-term temporary workers, with better selection and appraisal and better work introduction 4. At the same time, internal numerical flexibility options are tried. 5. Sophisticated systems of external flexibility (in-house agency, thorough selection and training of new entrants) in combination with internal numerical (flexible labour time strategies) and internal functional flexibility (teamwork) are developed as it becomes clear, that flexibility of labour is a permanent situation. Matching market demands for quality and reliability, strategic choices that differ per company, prudent staffing policies for fear of economic downturns, and changes in labour market circumstances, the configuration of flexible labour becomes ever more complex. Personnel departments learn along the way and all of them seem to follow the same general pattern in this industry, as illustrated in figure 1.
Discussion with Flexibility Literature To our minds, the configurations of flexible labour that individual companies decide on are difficult to explain from general theory. Even within one, relatively small industrial sector, many company specific circumstances
Figure 1. Typical Development of Labour Flexibility in the Automotive Industry Between 1985 and 1998
Volume 10
Number 4 December 2001
# Blackwell Publishers Ltd 2001
FLEXIBLE LABOUR STRATEGY
yield an influence. Therefore, as a rule, grand generalisations like the Atkinson (1984) model are not very helpful to explain or predict company behaviour. For example, while a core of permanent employees can be found in all of our companies, these manifestly do not correspond to the characteristics Atkinson ascribed to them. The strict division between core and periphery in terms of training, remuneration and job security he proclaimed is nowhere in evidence in our cases. The New Flexible Firm model proposed by Ackroyd & Proctor (1998, limited to manufacturing companies) seems to apply somewhat better. Companies are looking for production flexibility and possibilities to rapidly vary the number of employed personnel, according to demand and general economic circumstances. A core group can be distinguished, it is seen as the heart of the company workforce, but not treated very differently from the flexible employees. They are certainly not better trained or educated, not exempt from the hardships of fluctuating demands, nor from a general increase in performance standards. In many cases, flexible workers can and do get permanent contracts, are trained equally well (or bad) and asked to perform the same jobs at the same pace. When outperforming permanent staff, the permanence of those relations is soon abandoned in favour of former temporary employees. If they exist at all, core and periphery tend to blend together. Looking at the results of our research so far, the rhetoric of much human resource management literature is clear. Many management strategies have been proclaimed, as conscious acts to regain control or to fight costs at the expense of labour. But in as far as deliberate and consistent employment strategies by management can be detected, it truly seems to be of a third order magnitude. Overall strategic choices, market circumstances and a ‘learning curve’ for flexible arrangements dominate the trial and error way in which our companies have arrived at the configurations of flexible labour they now use. In that sense, we agree fully with Hunter et al. They state: ‘. . . the overriding impression was one of ad hoc measures to keep the business viable in the face of greater competitive pressure to meet temporary needs, or (in the case of multi-establishment organizations) to satisfy divisional headoffices . . .’ (Hunter, McGregor, MacInnes & Sproull, 1993, p.396). The experience of our companies in the automotive over the past ten to fifteen years therefore might at best be labelled as resulting into ‘emerging strategy’. They are not well
# Blackwell Publishers Ltd 2001
249
thought out conceptual plans for staffing in view of the business needs of the future, though that is what human resource management is supposed to be all about. Rather they are slowly emerging ‘fixes’ for actual and direct operational problems that in their turn force management to rethink earlier decisions and –because of their recurrent and persistent nature- they eventually evolve into a more or less coherent approach. Slowly we see ‘rules of thump’ and ‘common practices’ emerge, such as a maximum of 20–25% of flexible staff in production, longer term rather than shortterm flexible contracts, secondment constructions, in-house agencies, labour pools and internal numerical forms of flexible employment. The reasons for the successive stages in the development are crystal clear and have many commonalties over the different companies, as our research shows. In that sense, one could say that indeed some form of labour use strategy is emerging, by default. We expect this knowledge to become more widespread, as networks of HR practitioners do exist and intermediaries copy successful configurations to new client organisations. At the end of the day, it seems the only ‘strategic’ task personnel departments have in production companies is to take all different and often contradictory demands and circumstances into consideration and to deliver what production needs: a flexible operating staff of the right quality and in the right quantity, preferably just-in-time. It seems to us that they are getting better at it, aided by changes in the institutional context that, at least in the Netherlands, offer many more opportunities than 20 years ago.
References Atkinson, J.S. (1984) Flexibility, uncertainty and manpower management. IMS Report No. 89, Brighton, Institute of Manpower Studies. Ackroyd, S. and Procter, S. (1998) British manufacturing organization and workplace relations: some attributes of the new flexible firm, British Journal of Industrial Relations, 36, 2, pp. 163–183. Blanpain, R. (1999) The world of work and industrial relations in developed market economies of the XXIst century. The age of the creative portfolio worker. Blanpain, R. and Biagi, M. (eds.) Non-standard work and industrial relations. Bulletin of Comparative Labour Relations nr. 35. Kluwer Law International, Deventer. Bolhuis, M. Van (1996) External flexibility; an investigation in companies, State Department of Social Affairs and Work, The Hague (in Dutch). Brewster, C., Mayne L. & Tregaskis, O. (1997) Flexible working in Europe, paper presented at the 13th EGOS Colloquium, Budapest.
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
250
Creagh, M. and Brewster, C. (1998) Identifying good practice in flexible working. Employee Relations, The international journal, 20, 5, pp. 490–503. FME (1996) Labour market study The Netherlands; a study of labour and flexibility in the metal industry, FME, Zoetermeer (in Dutch). Hunter, L., McGregor, A., MacInnes J. & Sproull, A (1993), The ‘flexible firm’: strategy and segmentation. British Journal of Industrial Relations, 31, 3, pp. 383–407. Looise, J.C., Riemsdijk, M.J. van and Lange, F. de (1998) Company labour flexibility strategies in the Netherlands: an institutional perspective. Employee Relations, The international journal, 20, 5, pp. 461–489. Moerman, P.A. et al. (1996), Strategy and labour, Onderzoeksteam FIER, OSA-werkdocument W141, Den Haag. (in Dutch) Nevat (2000), Fact Sheet Automotive Suppliers in the Netherlands, Nevat/Holland Automotive, January 2000.
Volume 10
Number 4 December 2001
Pollert, A. (1991) The orthodoxy of flexibility. Pollert, A. (ed.) Farewell to Flexibility? Oxford, Basil Blackwell. Snijder, R. (1994) Choice in the Dutch Automotive Industry; A study into the chances of suppliers, FIER-rapport, Economische Zaken, Den Haag (in Dutch).
Maarten van Riemsdijk is associate professor Human Resource Management at the Faculty of Technology and Management, University of Twente, The Netherlands. Jan de Leede is research/consultant at TNO Work & Employment and assistant professor Human Resource Management, Faculty of Technology and Management, University of Twente.
# Blackwell Publishers Ltd 2001
R&D AS A COMPETENCE CREATING BUSINESS
251
Research and Development as a Competence Creating Business in a Business Olaf Fisscher, Klaasjan Visscher, Alan Pearson and Ursula Weisenfeld Research and development departments in industrial firms may not take it for granted anymore that they are the only preferred supplier of research and development to the company of which they are a part. The growing need to be innovative and the increasing availability of innovative competencies on the market result in a pressure on these departments to become more business-like and to reconsider their sources of competitive advantage over other (potential) suppliers of research and development. Traditionally, scientific and technological knowledge and skills concerning the product were the prime source. Nowadays, managerial competencies and the ability to work for and with your clients and suppliers are becoming more important. To become a competitive business in a business, research and development departments should create the competencies that enable them to create value for their clients. This calls for good competence management, comprising management of human resources, information technology, and internal and external interfaces. In this paper we explore what it means and what it takes for research and development departments to implement competence management, elaborating it theoretically and describing a case of competence management in the research and development department of a European car company.
Introduction
F
or industrial firms it is becoming more and more important to create product and process innovations on a regular basis (Bolwijn & Kumpe, 1991; Nonaka & Takeuchi, 1995). This means that the strategic importance of the research and development function of a company will grow, because it is a major contributor to innovation. But this does not automatically mean that the importance of the research and development department will grow as well, since the research and development function is not necessarily restricted to the research and development department (Kerssens-van Drongelen, 1999). To perform the research and development function, a company may go elsewhere than to their own department. Innovative competencies are more and more available on the market, offered by specialised consultancy agencies, commercial research institutes and universities. And the stronger the pressure is # Blackwell Publishers Ltd 2001. 108 Cowley Road, Oxford OX4 1JF and 350 Main St, Malden, MA 02148, USA.
on a company to produce innovations, the stronger may be the inclination to go elsewhere if the own department cannot entirely fulfil their needs. Furthermore, companies that are focussing on their core competencies, may outsource large parts of their non-core research and development. Therefore, there is a growing pressure on research and development departments, resulting from both the new opportunities and the threats. They have to generate a continuous stream of new products and they have to compete with external sources of innovation. They may not take it for granted anymore that they are the only preferred suppliers of the research and development function to the company of which they are a part. In other words, the relation between the department and the ‘mother company’ is developing from a hierarchical relation into a market relation. This is particularly so for many small and middle-sized research and development departments. They used to have a significant
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
252
impact on the company, especially when the parent company was relatively small and focussed on a narrow product portfolio or a specific geographical area, and dealing with a limited number of technologies. Potential risks for the parent organisation are that this may result in them not recognising potential threats from the development and introduction of new technology or in their inability to respond rapidly enough to remain at the forefront of a changing competitive environment. This is not a new phenomenon (see e.g. Cooper & Schendel, 1976), but it is one, which still faces many companies and we would argue that it is a real concern to smaller departments. They might be unable to change their direction quickly and can become obsolete when the new ‘skills’ can be bought elsewhere. The same departments also face threats even if the technology does not change, because of the increasing pressure on costs and the willingness of organisations to satisfy more of their standard business needs through outsourcing. The challenge for many research and development departments in the years to come is therefore to remain the preferred supplier of innovations for their internal clients, and, possibly, to sell their services to external clients as well. To achieve this, a department should be managed as a business, and as long as they are part of a company, as a business in a business. This means that the management of a department should have integral responsibility for their products and their resources. They should manage their human, technological, financial and other resources to fulfil the needs of their internal and external customers. They should be aware of their (potential) competitors and in this light they should exploit their strategic advantages and minimise their strategic disadvantages. In order to do this we argue that they must focus on the management of one of the resources of a research and development department in particular: competence. Competence can be defined as ‘the ability to do things’, comprising both knowledge and skills. Competence has grown more important during the last decades (Prahalad & Hamel, 1990; Wiig, 1993). This is the case for all businesses, but especially for research and development businesses (Kerssens- van Drongelen, et al., 1996), since for them competence is not only a resource, but also the main component of their product. Managing competence is the key success factor to achieve innovations on a regular basis (Nonaka & Takeuchi, 1995), and therefore the basis for success for a business-like research and development department. This means that
Volume 10
Number 4 December 2001
managing research and development is to a large extent managing competencies in order to create new competencies. In this paper we explore what it means and what it takes for the research and development department of an industrial firm to implement competence management. In particular, we describe the case of the research and development department of a European car company. This department experiences the pressures to become more business-like, as described above, and has shown some preliminary interest in the concept of competence management, but it has not yet implemented it, at least not deliberately and under that label. In this sense the department is typical for most research and development departments in industry. The case study has been conducted as part of the 2nd European Summer Workshop on Strategic Technology Management at the University of Twente. In this workshop, students, junior researchers, senior researchers and practitioners from Germany, the United Kingdom and the Netherlands came together to study ‘competence management in research and development’. The participants of the workshop visited the company, interviewed several members of the research and development department and collected relevant documents. This data was taken as input to answer the question ‘what does it mean and what does it take for this department to implement competence management’. In feedback-sessions, the preliminary answers to these questions were discussed and elaborated with the staff of the company. At a later point, extra interviews were conducted to explore the relation between ‘competence management’ and ‘operating as a business in a business’.
Framework of analysis For research and development, three fields of competence are extremely important, viz. ‘science and technology’, the ‘strategy, structure, and culture of clients’ and ‘management of the new product development process’ (see figure 1). In each of these areas, a business-like research and development department should create competencies to establish a competitive advantage over (potential) competitors. Traditionally, research and development departments derived their advantage mainly from their specific scientific and technological knowledge and skills. However, scientific and technological competencies have become, on the one hand, more complex and specialised (too
# Blackwell Publishers Ltd 2001
R&D AS A COMPETENCE CREATING BUSINESS
253
Figure 1. Fields of Competence complex and specialised for one company to handle), and on the other hand, more globalised and more easily available and accessible. It is nowadays more difficult to gain competitive advantage on this point. Research and development departments in companies producing complex products, such as cars, are likely to be overtaken in their specific scientific and technological competencies by specialised suppliers or research institutes. Managerial competencies, such as spotting opportunities in the market, managing complex new product development projects, co-operation and communication with co-developers and suppliers, in order to create new products that meet customer demands, are becoming a more and more important source of competitive advantage. It is less about ‘knowing all the details’ and more about ‘making it all happen.’ The third source of competitive
advantage, knowledge of the strategy, structure and culture of the clients and the skills to handle them, is very important for ‘internal’ research and development departments. They should have knowledge of the whole organisation, in structural as well as in cultural terms, which can only be copied by external suppliers of research and development services, if they work on a very long-term basis with the company. Managing these competencies has an internal dimension and an external dimension. Internal competence management has to do with managing people (humanware) and technology, especially information technology (hardware and software), the ‘carriers’ of competence. External competence management has to do with managing interfaces in the mutually adaptive relation with external stakeholders (see figure 2).
Figure 2. Components of Competence Management
# Blackwell Publishers Ltd 2001
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
254
Often, competence management is restricted to hardware, software and humanware. Competence management then consists of a series of local initiatives, such as building a database with the dos and don’ts of making windscreen wipers, or a workshop on motivational mechanisms. Without the external component of competence management, these internal initiatives stay local and probably without much effect on the competitive advantage of the research and development department or the company as a whole. Competence management requires strategic choices about which competencies are core and which are non-core, which competencies can be outsourced and which should be kept and developed internally. These strategic choices direct to a large extent the choices to be made in human resource management and technology management. Hiring people with strategic competencies, re-educating people with non-core competencies, reorganising multifunctional project-teams, developing data-bases where they can be of use, supporting the creation of relevant networks with information technology, etc.etc.. Both internal and external competence management is necessary to create the competencies that have value for the external stakeholders. We will elaborate these three components of competence management in the following section
Interface management Interfaces occur when people or groups of people have to collaborate to conduct a task. (Brockhoff et al., 1996). Managing competencies as the key task of a research and development department involves creating, managing and ‘selling’ ideas and knowledge within and outside the company. Having ideas and spending a lot of money is not enough for becoming a truly innovative research and development department, let alone an innovative company. Take Procter & Gamble as an example: the company announced in October 1999 ‘‘that it is prepared to give away or license any of its 25,000 patents, including those used in established brands. Chuck Hong, the firm’s director of R&D for corporate innovations, thinks this will ‘force us to continually invent’.[. . .] P&G has even started looking outside for new ideas’’ (The Economist, October 30th 1999). But, due to the nature of this task: managing competencies, specific interface problems arise, within the company and in the relation with other companies. Within a company, different departments may vary in their time horizons, perceptions
Volume 10
Number 4 December 2001
of situations and priorities of targets. The respective cliche´ is the researcher with a longterm vision of a technically perfect radical invention which has to be introduced as the perfect product into the market, meeting the marketing manager with a relatively shortterm view of what is asked for in the market (for research and development – marketing interface see e.g. Brockhoff, 1989) In the relation with other companies, additional problems arise due to the nature of competencies as a product, e.g.: . The risk of selling competencies to poten-
tial (future) competitors has to be taken into account. . The not-invented-here-syndrome might make it difficult for research and development departments to externally sell their products. . Signalling competencies might imply giving potentially sensitive information. . Buyers need competence to evaluate competence. The problems of externally acquiring technology have been dealt with in literature a decade ago (e.g. Sen & Rubenstein, 1989; Hamel, Doz & Prahalad, 1989; Pisano, 1990). More recently, various kinds of barriers to technology transfer between different institutions have been discussed (Gemu¨nden & Walter, 1996). A look at these barriers illustrates some problems involved in selling competencies externally: . Barrier of not knowing: who has which
competencies. . Barrier of not understanding: how to
integrate the competencies. . Barrier of not wanting: which negative
consequences are associated (to whom it may concern). . Barrier of not being allowed to: rules and laws preventing the acquisition of the competencies. Different instruments are suitable for handling interface problems occurring on different levels. Within a company, developing a milestone chart for a project, defining incentives for contributions and disincentives for not meeting milestones, and creating common perceptions and priorities of the project can be used to manage a project across departments. In relation to other companies, identifying competencies, assisting during implementation, and generating trust and commitment are ways to lower the barriers on the buyer’s side. On the seller’s side, those competencies that are or might become at the heart of the company have to be defined clearly and be protected appropriately. For
# Blackwell Publishers Ltd 2001
R&D AS A COMPETENCE CREATING BUSINESS
non-core competencies, a market orientation needs to be developed.
Human resource management In order to manage new product development successfully in the short-term project perspective and in the long-term programme perspective, human resources management is one of the main tasks, and maybe the most difficult one (see e.g. Shapira, 1995). How can one develop, stimulate and exploit knowledge, creativity, involvement and loyalty? First of all, it is important for research and development managers to focus not only on the new product development processes, but also on the processes of competence creation. Both should be regarded as primary processes. Secondly, it is important to define and create competencies on an individual level and on a collective level of teams, platforms and business units as a whole. And finally, it is important to recognise and value the cultural side of competence creation. People do not only learn through the transfer of explicit, cognitive knowledge, but they also learn through socialisation, i.e. becoming familiar with a field of competence, the language games, practices and the norms and values. Research managers should balance the short-term orientation on operationally effective new product development processes and the long-term orientation on the strategically flexible organisation of the new product development function (De Weerd-Nederhof, 1998). In the short-term perspective, human resources management should facilitate success according to the ‘right person, right place, right time’ principle, in which concrete new product development processes are leading. In the long-term perspective, human resources management should organise individual competence creation. In a study on organisational conditions for individual competence creation in industrial research and development, Brugman (1999) observes a lack of attention for this management of competence. In the context of human resources management, competence management should not only be implemented on the individual level, but also on the collective level. In the organisation of new product development, more and more team-based structures are introduced. Multifunctional teams are common practice in new product development projects and programmes (De Weerd-Nederhof, 1998). And a related concept, platform, receives a growing attention. This concept originates from the automotive industry. In the narrow
# Blackwell Publishers Ltd 2001
255
sense, a platform is a part of the structured frame of a car, and in a broader sense it is used to name the part of a product that is shared by different models in a whole family of products. This concept has been stretched and applied to the organisation of multiproduct development (Muffatto & Roveda, 1999). Robertson and Ulrich (1998) define a platform as a collection of assets (components, processes, knowledge, people and relationships) that are shared by a set of products. The concept connects product architecture with organisational architecture, and helps to create a high level of flexibility through standardisation and modularisation. This flexibility is very important for companies that have a multi-product strategy and operate in highly innovative and competitive markets. The definition and creation of competencies on a team or platform level requires collective experience. People should know what they can expect from each other, in their professional contribution and in their commitment to the task of the group (see Weick & Roberts, 1993). Besides, collective experience makes reuse of competencies possible. The importance of this can be illustrated by the following example (from Lindkvist et al., 1996). In 1992, Ericsson got the prestigious contract from Tokyo Digital Phone to develop and install a mobile telephone system in the Tokyo area. A crucial factor in getting this contract was that Ericsson identified and re-used the competencies of several project teams, which were developed in projects in comparable fields. In order to create competencies on an individual and a collective level, an innovative culture (Neuijen, 1992) should be established. This requires a well-chosen and well communicated strategy, facilitated by structures, systems and procedures, and people committed to this strategy. Culture is the value base of collective ambition, concerted action and competence creation. The following issues are especially important to create an innovative culture with a double focus, on the effective execution of projects and on building competencies. (Neuijen, 1992; Paashuis, 1997; De Weerd-Nederhof, 1998; Fisscher & De Weerd-Nederhof, 1999; Lindkvist et al., 1996). . There should be no room for a not-
invented-here-syndrome. . People should be able and willing to
understand other fields, tasks, interests and languages than their own. . Managing new product development should be valued as a core competence and not as a non-core or even superfluous activity, only costing time and money.
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
256
. There should be a well-chosen and well-
maintained balance with respect to the dominance of scientific and technical disciplines and functional areas, such as marketing, research and development, and production. . There should be balance in rewarding individual contributions and collective contributions.
Information management Information is now becoming the lifeblood of most organisations. It is increasingly available on a global basis and recent developments in both software and hardware are making it readily accessible at point of use and at a reasonable cost. And this does not only apply to scientific and technical information, which is the area to which research managers have paid most attention in the past. It also applies to economic, social, political and environmental information that is likely to affect the current and future activities in which an organisation is, or is likely to be involved. Such information, when accessed early, can provide real competitive advantage and it is not difficult to argue the case that research and development should have access to this information, in order to provide inputs to the decision making process and enable them to be in a position to respond rapidly to potential opportunities as well as threats. The same advances in information and communication technology, coupled with a greater acceptance of openness within organisations, enables projects to be better planned, monitoring to be more effective, relevant feedback provided, and corrective action more rapidly taken.
Case study The company we studied is a middle-sized European car company. Until a few years ago, it was an autonomous enterprise, but nowadays it is part of a worldwide operating company. A reason for this merger was that, given its size, the company lacked sufficient resources, while huge investments in new product development and production facilities were necessary to survive in the market. A package of organisational changes accompanied the merger. We will focus on the changes in the research and development department of the company, considering its function, its organisation, and its position within the company and the external environment. Traditionally, the company developed and produced all parts of the car itself. However,
Volume 10
Number 4 December 2001
because of the intensified, globalised competition, asking for higher quality, lower costs and shorter and more reliable lead times, the shortened product life cycles and the growing complexity of technology, new product development requires nowadays more people, money, expertise and effort than before. Therefore, one can detect trends towards more standardisation, modularization and outsourcing in the automotive industry. Suppliers do not only supply single components, but also whole systems, like engines or brakesystems. The focal points of the redesign of the research and development department that took place include the further orientation on its contribution to the whole value chain, the improvement of its interfaces with other departments (especially marketing and production) and with (system) suppliers, and the strengthening of strategy based core competencies. In this redesign, several developments and difficulties were related to ‘becoming a business in a business’ and ‘competence management’. In the case-study, we identified the following key issues. . The company is changing its position in the
.
.
.
.
value chain. The focus shifts from managing the entire production and sales process to ‘defining the product, managing the programme, and marketing the brand’. Besides, the development of product- and customer-related services is growing more important. This asks for different competencies. The research and development department becomes more involved in the concept development, preceding the development and design of the separate parts and systems. Detailed knowledge of all the parts becomes less important. Because of the growing outsourcing on a systems-level, the department develops longer-term relationships with suppliers/ co-developers and works more in multifunctional cross-organisational teams. The task of the department partly shifts from doing research and development to managing the interface between the company and its suppliers. In operational interface management, gaps in culture, language, and interests have to be bridged. This is a difficult but challenging tasks for the project managers. A problem with outsourcing the development of whole systems is not to lose all the competence connected to those systems. The company has still to be able to articulate functional and technical requirements, to control progress, and to judge
# Blackwell Publishers Ltd 2001
R&D AS A COMPETENCE CREATING BUSINESS
.
.
.
.
ideas and products of the supplier. It should maintain and develop knowledge and skills in order to be able to have fruitful discussions with their suppliers. Another problem with outsourcing has to do with the informal, improvising culture of the department. Managing the joint or outsourced development of systems requires a more formal organisation and communication structure and partly a formalisation of ways of working of people. However, formalisation has the risk of losing the flexibility and creativity of the current culture. Another cultural problem is that the department is dominated by people with a background in mechanical engineering, who focus on physical processes and products. It is difficult to shift focus in this culture towards more managerial tasks and to non-mechanical and less tangible, especially electronic, parts and systems. The development of product- and customerrelated service is also problematic. The status of the development of services is low compared to the ‘technical’ new product development. And newly hired employees, with adequate skills to develop services, have difficulties in creating connections with the traditional researchers and developers. In this respect, the culture has to change, on all levels in the hierarchy. Interorganisational co-operation enables access to a much broader network of scientific and technological competencies. Being a part of such a network, the department should balance their contribution to and their exploitation of the network. This requires the (still hardly developed) competence to assess, hand out and acquire competencies in a controlled and balanced way.
Conclusions In the case company, as well as more generally in research and development departments of companies that make complex products, one can observe a shift in the sources of competitive advantage. Traditionally, scientific and technological knowledge and skills concerning all the parts of the product, were the prime source. Nowadays, managerial competencies and the ability to work for and with your clients are becoming more important. A research and development department should be aware of the growing internal and external pressure to become more businesslike, and should manage its competencies to
# Blackwell Publishers Ltd 2001
257
create competitive advantage. They should not rely on their long history with the company or aim at obtaining a research monopoly within the company. They should let their ‘mothercompany’ co-operate with competing suppliers of research and development, but make sure that they are a vital part of this cooperation. They should not aim at being the only supplier; they should aim at being the best. If they do not take up this challenge, they run the risk of becoming obsolete. Especially departments in small and middlesized firms might be reduced to technical support facilities for local needs, or might even disappear, as information technology provides fast access to local information and other information is available on the market. There seems to be no promising future for research and development departments along the lines of the past. In the case company, where they start to recognise these developments, competencecreating initiatives are still developed locally in the organisation and with a very limited scope. A more integral approach, encompassing management of information technology, human resources and external interfaces, based on strategic choices about core-competencies, are necessary to create competitive advantage. Achieving this has not only to do with making plans and implementing them, but also with finding balances, in particular between formal and informal culture, between planned, controlled action and improvisation, and between outsourcing systems (and the competencies connected with them) and keeping the competencies to be able to do so in a productive way.
References Bolwijn, P.T. and Kumpe, T. (1991) Marktgericht ondernemen, Van Gorcum, Assen. [in Dutch] Brockhoff, K. (1989) Schnittstellen-Management. Abstimmungsprobleme zwischen Marketing und Forschung und Entwicklung, Stuttgart. [in German] Brockhoff, K., Chakrabarti, A. K., Hauschildt, J. and Pearson, A. (1996) Managing Interfaces, in: Gaynor, G.H. (ed.), Handbook of Technology Management, New York, Chapter 27. Cooper, A.C. and Schendel, D. (1976) Strategic responses to technological threats, in: Business Horizons, February, 61–69. De Weerd-Nederhof, P.C. (1998) New product development systems; Operational effectiveness and strategic flexibility, Doctoral disseratation, University of Twente. Fisscher, O.A.M. and De Weerd-Nederhof (1999) Social-dynamical aspects of quality management in NPD, in: Proceedings of the Productivity & Quality Management Conference, MCB University Press, 284–301.
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
258
Gemu¨nden, H.G. and Walter, A. (1996) Fo¨rderung des Technologietransfers durch Beziehungspromotoren, in: Zeitschrift Fu¨hrung und Organisation 4, 237–245. [in German] Hamel, G., Doz, Y. L. and Prahalad, C. (1989) Collaborate with your Competitors – and Win, in: Harvard Business Review 67, 133–139. Kerssens–Van Drongelen, I.C. (1999) Systematic design of R&D performance measurement systems, Doctoral dissertation, University of Twente. Kerssens-Van Drongelen, I.C., De Weerd-Nederhof, P.C. and Fisscher, O.A.M. (1996) Describing the issues of knowledge management in R&D: Towards a communication and analysis tool, in: R&D Management, 26,3, 791–808. Lindkvist, L., So¨derland, J. and Tell, F. (1996) Managing knowledge concurrently: Complexity and learning in projects, in: Proceedings of the R&D Management Conference ‘Quality and R&D’, University of Twente, Enschede, 213–227. Muffatto, M. and Roveda, M. (1999) Developing product platforms: Analysis of the development process, in: Proceedings of the Productivity & Quality Management Confference, MCB University Press, 11–32. Neuijen, J.A. (1992) Diagnosing organizational cultures; Patterns of continuance and change, Doctoral dissertation, University of Groningen. Nonaka, I. and Takeuchi, H. (1995) The knowledgecreating company: how Japanese companies create the dynamics of innovation, Oxford University Press, New York. Paashuis, V.J.B.J. (1997) The organisation of integrated product development, Doctoral dissertation, University of Twente. Pisano, G. (1990) The R&D Boundaries of the Firm: An Empirical Analysis, in: Administrative Science Quarterly, 35 (1), 153–176.
Volume 10
Number 4 December 2001
Prahalad, C.K. and Hamel, G. (1990) The core competence of the corporation, in: Harvard Business Review, May–June, 79–91. Robertson, D.and Ulrich, K. (1998) Planning for product platforms, in: Sloan Management Review, Summer, 19–31. Sen, F.and Rubenstein, A.H. (1989) External Technology and In-House R&D’s Facilitative Role, in: Journal of Product Innovation Management, 6, 123– 138. Shapira, Ph. (ed.) (1995) The R&D workers, Quorum Books, London. Wiig, K. (1993) Knowledge management, Schema Press, Arlington.
Olaf A.M. Fisscher is Full Professor of Quality Management and Business Ethics at the Faculty of Technolgy and Management, University of Twente, the Netherlands. Klaasjan Visscher is Assistant Professor at the faculty of Technology and Management, University of Twente, the Netherlands. Alan Pearson is Emiritus Professor of R&D Management at Manchester Business School and Visiting Professor at the faculty of Technology and Management, University of Twente, the Netherlands. Ursula Weisenfeld is Professor in Marketing and Technology Management, University of Lu¨neburg, and Lecturer, Judge Institute of Management, University of Cambridge, UK.
# Blackwell Publishers Ltd 2001
RESEARCH PERFORMANCE MEASUREMENT
259
Design of a Research Performance Measurement System: The Case of NIAB Steven Visser, Inge Kerssens-van Drongelen, Petra de Weerd-Nederhof and James Reeves At NIAB, a UK based company that provides research, services and information to the agricultural and food sector, a system has been designed that helps to assess and manage the growing research programme of the company. Since the company was ‘privatised’ four years ago, research activities have increased and moved away from solely applied research to a mix of applied and more fundamental research. Being a key element of the new developed company strategy, research now plays an increasing important role in broadening the scope of the company and keeping current services competitive by driving innovation. All research at NIAB is externally funded. In this paper we report on the process to design NIAB’s performance measurement system, for which the Performance measurement system Systematic Design Approach was used. The design process was started with an elaborate structured problem analysis of the research process and its inter and extra-organisational context. Based upon this analysis, firstly a conceptual and secondly a detail design of a performance measurement system was made. To maximise the leverage from research, the system has been designed to optimise the value delivered to the funder as well as the value delivered to internal customers in the form of knowledge that drives innovation.
Introduction
T
his paper reports on the process of designing a research Performance Measurement System (PMS) at NIAB, a company based in Cambridge (UK). NIAB supplies consultancy, training, information, contract research and technical services to governments, supra-governmental agencies, agribusiness and farmers. Because the research programme within NIAB is of growing importance and size, NIAB management feels the need for a system to support the management of research. The changing role of government, pressures on the agricultural sector in the UK and the emergence of genetics as a source of knowledge and techniques in the field of agriculture and food have resulted in changes within NIAB. Privatisation (1996), restructuring (1999), and the formulation of a new strategy and mission (‘NIAB will achieve a key worldwide role in the development of plant genetic resources through research, technical services and training’) (2000) led to a renewed role for # Blackwell Publishers Ltd 2001. 108 Cowley Road, Oxford OX4 1JF and 350 Main St, Malden, MA 02148, USA.
research. Research programmes at NIAB traditionally underpinned the statutory and advisory programmes of variety evaluation and seeds testing, enhancing efficiency through the development of new methods and improving the effectiveness of the work and the value of the resulting information (Wellington & Silvey, 1997). In the new strategy, the role of the research programme has broadened. Research must become one of the key activities within the institute, being a business on its own and at the same time driving innovation within the institute through exploring new techniques, creating new knowledge in the field of agriculture and food, thereby creating opportunities for product and process innovation. It is important to mention here that at NIAB, research is funded externally by governmental and nongovernmental research funders. As a part of the implementation of the new strategy, NIAB management felt the need for a system to improve management of the growing and changing research programme by implementing a performance measurement
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
260
system for research. A research project was initiated together with the University of Twente in the Netherlands. The objective of this research project is to design a Performance Measurement System (PMS), which enables NIAB management and staff to improve control of the research process. The design will be based upon a systematic analysis and diagnosis of the research process within its context. This paper describes the process of analysing the organisation, defining the purpose and function(s) of the PMS and designing the PMS. We will start with a description of the used theory. Next, the design process is described. The paper will be completed with a reflection on the used design approach, recommendations for further use of this approach and for the design of a research PMS in general.
Design Approach The first step in the process of designing the PMS for NIAB was to design the process of designing itself. As distinguished by van Aken (1996), a design project not only involves object design (a model of the entity that has to be realised) but also a process design (a model of the design process itself ) and a realisation design (how the object design should be produced or implemented). Especially when designing a system that is dependent on the people working with it, the process of design is important for the success of the implementation and institutionalisation of the design. Interest in performance measurement and management has rocketed during the last few
years (Neely & Adams, 2001) and a lot of different frameworks and processes for designing performance measurement systems have been developed. Most of these existing approaches consist of a design approach to derive performance metrics, a framework to present the metrics and an underlying theory. The frameworks ‘accommodate’ the used metrics. They help to interpret metrics and show interdependence of metrics in different parts of the framework and the relationship between metrics in the same ‘frame’ of the framework. ‘All (existing) frameworks add value because they all provide unique perspectives on performance. The key is to recognise that, despite the claims of some of the proponents of these various frameworks and methodologies, there is no one ‘holy grail’ or best way to view business performance’ (Neely & Adams, 2000). Therefore it is very important to analyse the organisational context, identify reasons for measuring performance and to decompose these into desired functions and subjects of the PMS (see Figure 2), as this enables the design team to determine which ‘unique perspective’ is most appropriate. Unfortunately, most of the common approaches do not include such a problem analysis phase, so it is not made clear what are, and what are not, required functions of the PMS and whose performance will exactly be the subject of the PMS. For the design of the PMS at NIAB, the Performance measurement system Systematic Design Approach (PSDA) by Kerssens-van Drongelen (1999) was used. This approach was chosen because, unlike many other design approaches, it is a very complete approach that applies all principles of design (hierarchical decomposition, abstraction, sys-
DEFINITIONS Metric or performance indicator: a variable, which indicates the effectiveness and/or efficiency of a process, system or part of a system, when compared with a reference value. Measurement method: a method to assign a value to a metric. Measurement methods can be classified along two dimensions: qualitative/quantitative and objective/subjective methods. Performance measurement: the process of acquisition and analysis of information about the attainment of objectives and plans and about factors that may influence plan realization. PMS functions: different ways to use a performance measurement system and its outputs. PMS subject: the people, or systems of people and resources, whose performance is the subject of the PMS.
Figure 1. Definitions (source: Kerssens-van Drongelen, 1999)
Volume 10
Number 4 December 2001
# Blackwell Publishers Ltd 2001
RESEARCH PERFORMANCE MEASUREMENT
261
THE MEASUREMENT SYSTEM FUNCTIONS TAXONOMY 1.
2.
3. 4. 5.
6. 7.
Provide timely insight into deviations from objectives and environmental factors to support diagnosis by management as to whether and if so which steering measures would be necessary. Fuelling learning about process characteristics, and the influence of external factors and steering measures on these characteristics, and in this way improving the predictive model that may support better decision making in the future. Alignment and communication of objectives, agreements, and rules. Supporting decision making on performance based rewards. Provide timely insight into deviations from objectives and environmental factors to support diagnosis by subordinate(s) as to whether and if so which steering measures would be necessary. Justification of existence, decisions and performance. Motivating people through feedback.
Figure 2. The Measurement System Function Taxonomy (Source: Kerssens-van Drongelen, 1999) tematic variation, solution selection on the basis of the satisficing principle (Pahl & Beitz, 1996, p. 54-60)). The main advantage in this case was that the PSDA helps to decompose the measurement problem and gives guidelines for building and evaluating conceptual designs (frameworks) before designing the PMS in detail.
The design process therefore consisted, in line with the PSDA, of the following steps: a problem definition phase, including problem diagnosis and decomposition, a conceptual design phase, including systematic variation of conceptual designs, a detail design phase and an implementation phase (see Figure 3).
problem definition
determination of the organisational configuration, measurement subject(s), and PMS function(s)
conceptual design
Selection of metrics clusters, measurement method types, metrics framework, frequency, and reporting format
detail design
detailed design of metrics, measurement methods, organisational arrangements, database applications and reports
implementation
Figure 3. The PMS Design Process and its Outputs
# Blackwell Publishers Ltd 2001
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
262
Because, at the beginning of the project, the need for a PMS was not exactly defined, the project was started with an elaborate analysis of the research process and it’s inter- and extra organisational context. This analysis enabled us to get a clear overview of the critical function of research, to examine to which extent these critical functions were fulfilled (the current performance) and third to make a link between the current organisational configuration and the current performance. Based on that information, problems were revealed which were the input for the design process. For analysis and diagnosis, the framework for the description of NDP processes by De Weerd-Nederhof (1998) was used. This framework itself is based on the process-based contingency model for organisations by Boer and Krabbendam, (1993). The framework gives a description of primary-support and management processes, people, organisational arrangements, techniques and tools, the inter organisational context and the extra organisational context.
Methodology In this single case study, several methods of data gathering were used. To get an overview of the company, a document study was done and the company’s directors were interviewed. To reveal the critical functions of research and the extent to which these are fulfilled, a questionnaire was sent to all group heads of NIAB and the research director. Next to that, research group heads and heads of three non–research groups were interviewed. To describe the configuration of the research system, the research director, the key account manager for research, two research group heads and four senior researchers were interviewed. The results of these activities were fed back to and approved by the research director and a complete research group. After the description and diagnosis, a design team was formed consisting of the research director, a research group head, a non-research group head, the key account manager for research and two senior researchers. This design team was involved in making design choices by individually presenting them alternatives and asking them for their preferences. The research director approved all choices. The first author of this paper acted as facilitator for the design process.
Volume 10
Number 4 December 2001
Problem Definition Phase The problem definition phase was aimed at finding the desired function(s) and subjects of the PMS, and to create a list of requirements, which the system design must apply to.
Problem Analysis and Decomposition The result of this phase consists of three parts: a list of critical functions of the research process, a list of problems in fulfilling these critical functions and an elaborate description of the current configuration of the research system and it’s inter- and extra organisational context (which will not be included in this paper). The causes of the problems are found by looking at the organisational configuration. Four critical functions were identified: Drive internal innovation: NIAB research must be the driving force of innovation within the company. This means that research must create opportunities for improving products and creating new products in new and existing markets. The other business areas are the most important stakeholders. Give scientific image: NIAB research must improve the image of NIAB as a research based institute by producing scientific results of good quality and quantity and ‘market’ these results in the scientific community. This will increase the networking capabilities of NIAB, which are very important in the specific context in which the institute operates. Next, research staff has an interest in the ability of building a personal scientific reputation. Generate sufficient financial resources: NIAB research needs to meet its financial goals. NIAB as a whole is the most important stakeholder in this perspective. Satisfy funders: It is important that the funders that pay for the research are satisfied and that the relationship with funders is good. For the description of the problems in fulfilling these critical functions, a distinction was made into operational effectiveness and strategic flexibility. The identified problems in fulfilling these critical functions and their main causes are:
Operational Effectiveness 1. The quantity of delivered research outputs must rise in order to underpin current business. 2. The financial efficiency of the research process needs improvement of control. If
# Blackwell Publishers Ltd 2001
RESEARCH PERFORMANCE MEASUREMENT
the contribution margin rises, possibilities for investment increase. 3. The transfer of research to services is not high enough. Alignment of research and services programmes should improve.
Strategic Flexibility 4. The current perception of customers and people within NIAB of NIAB as a research based institute, must improve. 5. The building of new competencies for the future by doing research is not aligned with the future demand of the current customer base and with the core competencies/unique resources of NIAB. 6. Improvement in the effectiveness and efficiency of the research process only takes place through personal experience of researchers. Transfer of knowledge within groups and throughout the organisation is perceived to be low. The next step was to look in more detail to the underlying causes of the problems and to identify which of these causes could be (partly) taken away by a form of performance measurement. Analysing the earlier made elaborate description of the research system and its inter- and extra organisational context, a total of 24 problem causes were found, of which 13 possibly could be solved by performance measurement. To determine whether, and if so, what kind of performance measurement system could be useful to solve the identified control problems, for every problem a combination of a measurement system function and a measurement subject was defined, which was most likely to take one of the problem causes (partly) away. This was done with use of a list of all possible measurement subjects within NIAB and with use of the taxonomy of measurement system functions (see Figure 2). The result was a table with many possible functionsubject combinations. Together with NIAB management, two PMS functions where chosen that would help to tackle the most urgent problems: 1. providing insight into deviations from objectives to support diagnosis by management as to whether and if so which steering measures would be necessary 2. communication and alignment of goals. Both functions are applied on business area level. Other possible functions and subjects were not rejected, but development of performance measures for these functions can be done at a later stage, based upon the PMS that then already exists.
# Blackwell Publishers Ltd 2001
263
Analysis of Organizational Conditions In all interviews that were conducted during the problem definition phase, staff was asked for their opinion about the concept of performance measurement in general and their felt need for performance measurement at NIAB. Using this information and a checklist by Kerssens-van Drongelen (1999), it was checked whether the conditions for designing a performance measurement system were favourable. The use of this checklist revealed some points of attention, which were used to improve the design process. The most important improvement was to make extra effort to get staff familiar with performance measurement in general, and with the purpose of, and approach chosen for, this measurement system design project in particular.
Analysis of Contingency Factors The next step in the process of problem definition was to analyse the contingency factors that influence the design of the PMS. Kerssens van Drongelen (1999) has shown that innovation strategy, company size, the organisational control system, and the technical and commercial uncertainty of research have implications for the design of a research performance measurement system. Analysis of these factors resulted in a set of requirements for the system design.
Formulation of Requirements The last step in the problem definition phase was to create a list of requirements for the PMS. This list consisted of functional requirements imposed by the chosen measurement system functions, constraints imposed by the contingency factors and additional user demands.
Conceptual Design The aim of this phase was to obtain a conceptual design of the PMS: a combination of metrics cluster(s), measurement method type(s), a rough indication of the frequency and timing of measurements, a type of reference value and a type of reporting format. This was done by generating a list of alternatives and choosing the best alternative, using the list of requirements, constraints and demands and the preferences of NIAB staff and management.
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
264
Figure 4. The Metrics Taxonomy
Metrics Clusters To be able to select metrics on a conceptual level, a taxonomy is needed. For selecting the most appropriate metrics clusters for NIAB research, we used the taxonomy by Kerssensvan Drongelen (1999) that is shown in Figure 4. Next, the guidelines for selecting metrics clusters given in the PSDA proved to be a useful starting point for selection. Having in mind the goals of the PMS: to support decision making by management and to communicate goals, it was decided to use measures throughout the whole process, (inputs/infrastructure, activities, outputs and outcomes). Because of the long cycle time of the research process, only measuring output would make the feed back time too long to enable management to make decisions at the right moment. The next decision was the time span that has to be covered by the metrics. It was chosen to use mainly recent past and near future metrics. For outcomes, far future metrics were chosen. This should enable NIAB research management to make decisions that have far future effects, based upon information about that far future. This is useful since at the very beginning of a research project, when the proposal is written, the outcomes are already largely determined. Starting point for the performance aspects that are important was the NIAB strategy.
Volume 10
Number 4 December 2001
Next, additional insights in useful performance aspects were derived from the different goals that stakeholders of the NIAB research process have. Using the reasoning presented by Anderson and Neely (2000) in their article on the performance prism, for every stakeholder the stakeholder satisfaction (their wants and needs), the strategies, processes and capabilities to attain stakeholder satisfaction and the desired stakeholder contribution was determined.
Measurement Methods To choose the measurement method to collect data, the PSDA guideline (given the specific context of NIAB research) of emphasis on subjective – qualitative measurements unless otherwise arranged by company wide requirements was accepted. In the detail design phase, the measurement methods per individual metric were chosen by optimising the measurement effort and the informational value of the metric-measurement combination.
Clustering of Metrics Together with choosing metrics and measurement methods, a framework to present the metrics had to be developed. Four different alternatives were evaluated: the performance prism (Neely & Adams, 2001), the balanced
# Blackwell Publishers Ltd 2001
RESEARCH PERFORMANCE MEASUREMENT
scorecard (Kaplan & Norton, 1996), a clustering directly derived from NIAB strategy and a new constructed clustering. . The performance prism: the performance
prism does give the possibility to address all stakeholders goals. The problem is that it proved difficult to derive measures for every category, and even if that had been possible, the number of metrics would become very high (at least 20 to 25). The underlying theory of the performance prism though, proved very useful to make sure that all stakeholders are considered to be involved in measurement. . The balanced scorecard: the balanced scorecard is somewhat more succinct but did not fit well with NIAB research because there is not one single customer group and there are no shareholders (stakeholders with only a financial interest). The business model of NIAB research is considerably different from that for which the balanced scorecard was developed. . A clustering derived directly from NIAB strategic goals: this clustering was somewhat unstructured because the strategy is a mix of desired inputs, outcomes, etc. . A self developed clustering, using elements of the three other options (shown in Figure 5): the horizontal axis is based on the strategic goals of NIAB research as they were found in the strategic plan of the company, complemented with the outcomes of the reasoning of the performance prism. The vertical axis is based on the clustering by Brown and Svensson (1995) and has a focus on the causal chain leading to the achievement of objectives, as recommended by the PSDA guidelines for the required PMS functions.
265
After evaluation, the last option was chosen, using the list of requirements and feedback from research staff. The most important advantages of the chosen clustering are the combination of the simplicity of the balanced scorecard, the stakeholder based approach of the performance prism, and the link with the strategic goals of NIAB.
Frequency and Timing The frequency of measurement is a trade-off between quality (up-to-dateness) of measurement data and time and money involved in measurement. Alternatives of once, twice or more than twice a year were evaluated. Consultation of research staff learned that twice a year was the maximum staff would accept, given the current high work pressure. For the timing of measurement the two main alternatives were to align with individual projects or align with the yearly financial cycle of the institute. Aligning measurement with individual project makes measurement data more appropriate for decision making about individual projects but would involve continuous measurement as projects are continuously started, finished etc. Measuring all projects at once before the yearly budgeting procedure would give an up to date measurement before the decisions about budgets are made. It was chosen to measure each individual project, before sending the proposal to the funder. The other measurements of projects will be aligned with the measurement of the whole business area. In this way, the number of measurements is minimised and project information will still never be more than 6 months old. The timing of Business Area measurements will be aligned with the
Figure 5. The NIAB Research Performance Measurement Framework
# Blackwell Publishers Ltd 2001
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
266
budgeting procedure such that recent information is available when making the budgets but without putting too much time pressure on research management during the actual budgeting procedure.
Reporting Formats It was chosen to make three different reporting formats: on business area level, on group level and on project level.
Detail Design In this phase, a list of possible metrics and measurement methods has been created, metrics and measurement methods have been selected, and the organisational arrangements needed for the system to operate have been defined. To facilitate the collection and storing of measurement date and the making of reports, a database application has been built. The result of this phase is a PMS that is ready for implementation.
Choosing Metrics In the preceding phase, clusters of metrics were selected. The next step was to generate a set of usable metrics. This was done using existing metrics catalogues from Kennerly et al. (2001) and from Kerssens-van Drongelen (1999), and metrics mentioned in articles by McGrath (2000), Brown & Svenson (1988), Groen et al. (2001) and Tipping et al. (1995). This resulted in a list of 82 possible metrics, categorised into the 16 frames of the framework. For each metric, a short description of the measurement method was made. Using
2 rounds of selection, this set was reduced to 18 metrics, which have been used in the first version of the measurement system. Selection of metrics was done using 2 criteria: measurement effort and informational value. Measurement effort is defined as the time and money involved in the collection of data. Informational value is defined as the extent to which a metric tells something about the attainment of goals: the information richness of the measurement. The metrics and the measurement methods were first ranked (from 1 to n) and categorised (high/medium/low) on their informational value, next they were categorised on measurement effort (little/average/much). For every metric-measurement method, a figure like Figure 6 was made and based upon that, the best metrics were selected. This selection was discussed with several members of research staff and management. Based upon that, the preliminary selection was altered. The altered selection was compared to the list of demands, constraints and requirements again. The metrics shown in Figure 7 were chosen. To illustrate the balance between operational effectiveness (OE) and strategic flexibility (SF), for each metric it is indicated to which of these forms of organisational performance it relates. The now obtained framework consists of a list of measures that enables NIAB research management to ensure at all times that the research Business Area is fulfilling its critical functions. A distinction is made between key performance measures and additional detailed measures. This is done to support the communication and alignment of goals function as well as the signalling to management function. Communication and alignment re-
Figure 6. Choosing Metrics
Volume 10
Number 4 December 2001
# Blackwell Publishers Ltd 2001
RESEARCH PERFORMANCE MEASUREMENT
267
Figure 7. The selected metrics quires a small set of goals since research on management-by-objectives indicates that if people are given more than six to eight objectives to accomplish, they ignore most of the objectives and concentrate on the two or three they believe are important (Brown & Svenson, 1988). On the other hand, signalling to management whether there are deviations from objectives requires that all objectives are covered and that performance along the causal chain leading to these objectives can be tracked. So for this function monitoring also the detailed metrics will be useful to timely signal deviations. At the moment, a database application is being built, in which all measurement data can be stored and which can generate reports that show the most recent values of all metrics. Next, storing the metric data makes it easy to monitor the development of all metrics over time.
Conclusions In this paper we have presented the design process followed so far to develop a Perform-
# Blackwell Publishers Ltd 2001
ance Measurement System suiting the specific situation of NIAB research. NIAB’s research activities have been analysed and some (control) problems have been found. A performance measurement system has been built that, according to theory, is likely to help NIAB to solve some of those problems. To which extent the performance measurement system will really be a success, can only be proven over time in practice. It will be very interesting to see to what extent the system will fulfil its intended functions and to what extent it can fulfil functions besides that, for which the system has not been specifically designed. As for the theories used to develop NIAB’s PMS, our experiences first of all indicate that research performance measurement in a nonprofit organisation where research is externally funded requires a performance measurement framework that is adapted to the businessmodel of the organisation. This business model is likely to be fundamentally different from that from which many existing frameworks for (research) performance measurement are derived. Hence the common design approaches in which such frameworks are
Volume 10
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
268
taken as the basis for the PMS design, cannot be applied unconditionally. Secondly, we conclude that the PSDA has proven a useful method for designing a PMS in such a situation. Using the PSDA, the designed performance measurement system is likely to fulfil the function that it should since the whole design approach is based on that concept. The conceptual design phase is a very useful step but the high level of abstraction in this phase makes it necessary that all involved people understand the used methodology and the used taxonomies. If not, the usability of the PSDA guidelines for this phase is limited and one will probably restrict the generation of alternatives to already existing frameworks like the Balanced Scorecard or the Performance Pyramid. However, even then the conceptual design phase is useful, as at least the framework will be chosen that is most appropriate for the PMS function(s) and subject(s). Finally, the framework for the description of NDP systems and performance appeared to be suitable for use in the problem analysis phase of the design of a PMS, but it is a very time consuming and elaborate method. Depending on the extent to which the PMS functions are already determined, a more focussed method of analysis will probably be suitable as well. We recommend that the analysis should then be focussed on analysis of the primary process, the management processes, organisational arrangements and critical functions of research. The emphasis placed in De Weerd-Nederhof’s (1998) framework on the balance between operational effectiveness and strategic flexibility stimulated the development of a balanced set of critical functions for NIAB research. Since these functions formed one axis of the conceptual design, the set of metrics chosen does also have this balance.
References Aken, J.E. van (1996) Methodologische vraagstukken bij het ontwerpen van bedrijfskundige systemen. Het paradigme van sociaal realisme, Bedrijfskunde, 68, 2, 14–22 Boer, H. and Krabbendam, J.J. (1993) Inleiding Organisatiekunde, Reader, Faculty Technology and Management, University of Twente, Enschede, The Netherlands Brown, M.G. and Svenson, R.A. (1988), Measuring R&D productivity, Research Technology Management, 31, 4, 11–15 Emmanuel, C., Otley, D. and Merchant K. (1990), Accounting for Management Control, Chapman& Hall, London, UK Groen, A.J., de Weerd-Nederhof, P.C., Kerssensvan Drongelen. I.C., Badoux, R.A.J., Olthuis,
Volume 10
Number 4 December 2001
G.P.H. (2001), Leveraging research and development: creating and justifying added value: an empirical assessment, proceedings of the 8th international product development management conference, Enschede, the Netherlands Kaplan R.S. and Norton D.P. (1996), The Balanced Scorecard: Translating strategy into Action, Harvard Business School Press, Boston MA, USA Kennerly, M., Neely, A. and Waggoner, D., The Catalogue of Performance Measures, http:// www.som.cranfield.ac.uk/som/cbp/catperf.htm, visited June 2001 Kerssens-van Drongelen, I.C. (1999), Systematic Design of R&D performance measurement systems, thesis, University of Twente, Enschede, the Netherlands Leeuw, A.C.J. de (1982), Organisaties: management, analyse, ontwerp en verandering: een systeemvisie, Van Gorcum, Assen, the Netherlands Lynn, G.S., Reilly, R.R. (2000) Measuring team performance, Journal of Technology Management, January–February, 48–56 McGrath, R.G. (2000), Assessing technology projects using real options reasoning, Research Technology Management, 43, 4, 35–50 Neely, A. and Adams, C., Perspectives on Performance: The performance Prism, http://www.som. cranfield.ac.uk/som/cbp/prism.htm, visited Monday, 11 June 2001. Pahl, G. and Beitz, W. (1996) Engineering Design. A Systematic Approach, Springer-Verlag, London, UK Tipping, J.W., Zeffren, E., Fusfeld, A.R. (1995) Assessing the value of your technology, Research Technology Management, 38, 5, 22–39 Weerd-Nederhof, P.C. (1998), New Product Development Systems, Operational Effectiveness and Strategic Felxibility, thesis University of Twente, Enschede, the Netherlands Wellington, P.S. and Silvey, V. (1997)Crop and Seed improvement, a history of the National Institute of Agricultural Botany 1919 to 1996, NIAB, Cambridge, UK
Steven Visser has recently obtained his masters degree in Industrial Engineering and Management from the University of Twente, Enschede, the Netherlands. The project described in this article has been his graduation assignment. Inge Kerssens-van Drongelen is management consultant at Cap Gemini Ernst & Young in Utrecht, the Netherlands, and assistant professor at the Faculty of Technology and Management at the University of Twente, Enschede, the Netherlands. Petra C. de Weerd-Nederhof is associate professor New Product Development at the faculty of Technology and Management, University of Twente, The Netherlands. James Reeves is director of Research and Laboratory Services at NIAB, Cambridge, UK.
# Blackwell Publishers Ltd 2001
INDEX TO VOLUME 10
269
Index to Volume 10 (e.g. 2/79 = number 2, page 79)
HERMENS, Antoine. Knowledge Exchange in Strategic Alliances: Learning in Tension
1. Articles BECH, Nils. Open Doors to Leading Projects: Your New Chance to Understand and Perform Project Leadership
2/96
BHAT, Bhaskar and BOWONDER, B. Innovation as an Enhancer of Brand Personality: Globalization Experience of Titan Industries
1/26
CARTER, Chris and SCARBOROUGH, Harry. Regimes of Knowledge, Stories of Power: A Treatise on Knowledge Management 3/210 CLARKE, Thomas and ROLLO, Christine. Capitalising Knowledge: Corporate Knowledge Management Investments
HUMMEL, Marjan , VAN ROSSUM, Wouter, OMTA, Onno, VERKERKE, Gijsbertus and RAKHORST, Gerhard. Types and Timing of Interorganizational Communication in New Product Development 4/225 HSU, Frederick B. Strategic Decisionmaking in a New Millennium: The Contribution from Hybrid Decisionmaking Modes
1/40
JENKINS, Roger. Enterprise Resource Planning as the Trojan Horse for New Rules of Operations Management 3/201 3/177
FAIRBANK, James F. and WILLIAMS, Scott David. Motivating Creativity and Enhancing Innovation through Employee Suggestion System Technology 2/68 FISSCHER, Olaf, VISSCHER, Klaasjan, PEARSON, Alan and WEISENFELD, Ursula. Research and Development as a Competence Creating Business in a Business 4/251 GASSMANN, Oliver. Multicultural Teams: Increasing Creativity and Innovation by Diversity
3/189
2/88
JONES, Elies, MANN, Darrell, HARRISON, David and STANTON, Neville A. An Eco-innovation Case Study of Domestic Dishwashing through the Application of TRIZ Tools
1/3
KODAMA, Mitsuru. Innovation through Strategic Community Management: The Case of NTT DoCoMo and the Mobile Internet Revolution 2/75 LESTER, Michael. Innovation and Knowledge Management: The Long View
3/165
OLIN, Tommy and WICKENBERG, Jan. Rule Breaking in New Product Development – Crime or Necessity?
1/15
GOMES, Jorge, DE WEERD-NEDERHOF, Petra, PEARSON, Alan and FISSCHER, Olaf. Senior Management Support in the New Product Development Process 4/234
RAGSDELL, Gillian. From Creative Thinking to Organisational Learning via Systems Thinking? An Illustration of Critical Creativity 2/102
HAAPASALO, Harri and KESS, Pekka. In Search of Organisational Creativity: The Role of Knowledge Management
RAY, Tim and LITTLE, Steve. Communication and Context: Collective Tacit Knowledge Tools and Practice in Japan’s Workplace ba
2/110
# Blackwell Publishers Ltd 2001. 108 Cowley Road, Oxford OX4 1JF and 350 Main St, Malden, MA 02148, USA.
Volume 10
3/154
Number 4
December 2001
CREATIVITY AND INNOVATION MANAGEMENT
270
SULIMAN, Abubakr Mohyeldin Tahir. Are We Ready to Innovate? Work Climate-Readiness to Innovate Relationship: The Case of Jordan 1/49 VAN RIEMSDIJK, Maarten and DE LEEDE, Jan. Flexible Labour Strategy in the Dutch Automotive Industry VISSER, Steven , KERSSENS-VAN DRONGELEN, Inge , DE WEERDNEDERHOF, Petra and REEVES, James. Design of a Research Performance Measurement System: The Case of NIAB
4/259
KHANDWALLA, Pradip N. Turnaround Excellence: Insights from 120 Cases
ROLLO, Christine and CLARKE, Thomas. International Best Practice – Case Studies in Knowledge Management 4. Book Reviews
KNOTT, David. The Place of TRIZ in a Holistic Design Methodology 2/126
CHIESA, V. R & D Strategy and organisation: Managing technological change in dynamic contexts
MANN, Darrell. An Introduction to TRIZ: The Theory of Inventive Problem Solving
2/123
MANN, Darrell. ‘A or B’ to ‘A and B’
2/134
Number 4 December 2001
2/139
2/119
LEIFER, Richard, MCDERMOTT, Christopher M., O’CONNOR, Gina Colarelli, PETERS, Lois S., RICE, Mark, and VERYZER, Robert W. Radical Innovation: How Mature Companies can Outsmart Upstarts 1/60
2. TRIZ Articles
SLOCUM, Michael S. and LUNDBERG, Catherine O. Technology Forecasting: From Emotional to Empirical
Volume 10
4/243
3. Books of the Quarter
COOPER, CARY L. Classics in management thought LU, QIWEN China’s Leap into the Information Age: Innovation and organization in the computer industry
3/221
1/64 2/122
1/64
THOMSON, A., MABEY, C., STORY, J., GREY, C. and ILES, P. Changing patterns of management development 1/66
# Blackwell Publishers Ltd 2001
