No title

Introduction

Accelerated learning in new product development teams Gary S. Lynn Ali E. Akgu¨n and Halit Keskin

The authors Gary S. Lynn is Associate Professor, Wesley J. Howe School of Technology Management, Stevens Institute of Technology, Hoboken, New Jersey, USA. Ali E. Akgu¨n is Assistant Professor of Science and Technology Policy and Halit Keskin is Assistant Professor of Management Strategies, both at the School of Business Administration, Gebze Institute of Technology, Gebze, Turkey. Keywords Team learning, New products, Development Abstract Learning in new-product development teams is cited as being vital in today’s competitive, uncertain, and turbulent environments. However, studies on accelerated learning in product-development teams are, surprisingly, lacking. This study proposes a model for accelerated team learning in new-product development based on constructs borrowed from accelerated learning models (or “suggestopedy”) in the individual learning scholarship. It is argued that fastlearning teams launch new products more quickly, and with increased probability of success. Moreover, specific mechanisms to help teams learn more quickly are within the control of teams. These include vision clarity, learning from customers and competitors, information coding, top management support, past product review, aggressive deadlines and daily meetings. Electronic access The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1460-1060.htm European Journal of Innovation Management Volume 6 · Number 4 · 2003 · pp. 201-212 q MCB UP Limited · ISSN 1460-1060 DOI 10.1108/14601060310500922

The complexity of today’s business environment is such that a company cannot survive unless it is flexible, adaptable and capable of learning. One-third of Fortune 500 companies disappear every 15 years due to an inability to change and renew themselves, and to do so quickly (De Geus, 1997). Companies that continuously learn and reinvent themselves on a timely basis are better able to take advantage of emerging opportunities in fast-paced competitive markets (Senge, 1990). Maira and Scott-Morgan (1997), for instance, have stated that organizations are now competing on their ability to change more quickly and more effectively than their rivals. De Geus (1997) has asserted that learning and, more importantly, learning more quickly than competitors, is vital for a company’s survival. Schein (1993), who has researched organizational learning for many years, has noted that organizations must learn to adapt quickly, or will be weeded out in the economic evolutionary process. Fast or accelerated learning is important for many functions within an organization, but it is vital in new-product development in which teams must respond quickly to rapidly changing technologies, customer needs, and competitive actions (Meyers and Wilemon, 1989). However, the new-product development literature has yet to produce a framework for accelerated learning in new product teams. In their suggestions for future research, Meyers and Wilemon (1989, p. 87) also asked: “How might conditions that facilitate individual learning accelerate or hinder team learning?” It is of interest that the literature on educational psychology and individual learning contains several models and methods to explain how individuals or students learn more quickly. Thus, this paper explores certain practices and mechanisms of rapid learning based on individual learning scholarship – for example, “suggestopedy” (Lozanov, 1978; Rose and Nicholl, 1997) – and then applies them (with some modifications) to new product teams.

Accelerated learning For the past 20 years, many scholars and practitioners have discussed the importance of learning (Argyris and Scho¨n, 1978; Senge, 201

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

1990) and fast learning (De Geus, 1997) in organizations. Because scientific and technological breakthroughs are occurring on almost a daily basis (Linksman, 1996), and because market and customer needs are changing quickly, organizations are required to learn increasingly quickly if they are to respond to these turbulent and uncertain environmental dynamics. For instance, Prusak (1997) has noted that a firm’s competitive advantage depends on what it knows and how fast it can learn something new. Similarly, Dandrea-O’Brain and Buono (1996) have stated that organizations should speed up their learning processes and adapt more quickly to the changing global business world. Maira and Scott-Morgan (1997) have noted that fast-learning organizations become increasingly responsive to fluctuations in their industries and to signals from their customers and suppliers. Guns (1996, p. 25) has defined a fast-learning organization as one that determines more quickly than the competition what works and what works better. He identified the possible benefits of being a fast-learning organization as: . quickly acquiring knowledge about customers’ values; . using new technology more effectively; . reducing cycle time of innovations; . being flexible; and . supporting changes in the organization.

Before discussing accelerated learning, a review of traditional team learning is a useful basis from which to begin.

Although fast learning is important for all parts of an organization, it is especially vital for technological new product development (NPD) teams because the rapid pace of change – in products, competitors, customer bases and suppliers – forces NPD teams to learn, and to do so quickly. For instance, Dodgson (1993) has argued that the ability to learn quickly regarding technological opportunities and changing patterns of competition is important for uncertain and rapidly developing technological industries, such as biotechnology. In this vein, as Maani and Benton (1999) have noted, teams should derive ideas from inside and outside the organization to produce innovative solutions in a short period of time, and thus accelerate the learning process. Similarly, Meyers and Wilemon (1989) have noted that both fastpaced industries in modern technology and established industries with mature products should learn more quickly for their new product offerings.

Traditional team learning versus accelerated team learning New product development team learning has been receiving great attention in practice as well as in academia in the past ten years (Brooks, 1994; Edmonson, 1999). Nonaka and Takeuchi (1995) and Leonard-Barton (1995) have asserted that organizations learn via cross-functional teams. Because NPD requires the integration of a broad base of knowledge (Grant, 1996; Kogut and Zander, 1992) by acquiring, processing, storing, manipulating and reducing information and knowledge, a new product team is recognized as a useful process for organizational learning (Moorman and Miner, 1997). However, as Kasl et al. (1997) have noted, the organizational literature neither defines nor clearly describes team learning. In addition, Meyers and Wilemon (1989) and Edmonson (1999) have noted that the number of studies about learning in teams is limited. Brooks (1994, p. 215) has defined team learning as the construction of collective new knowledge, involving the following processes: . problem-posing; . sharing knowledge and ideas; . integrating new knowledge; . gathering data; and . disseminating new information. Edmonson (1999) has defined team learning as the processes and outcomes of groupinteraction activities through which individuals acquire, share, and combine knowledge, including the following processes: . asking questions; . seeking feedback; . sharing information; . experimenting; and . discussing errors or the unexpected outcomes of actions. Kasl et al. (1997) have presented a researchbased model of team learning. Through two case studies (in the petrochemical and manufacturing industries), they identified team-learning processes and conditions – including appreciation for teamwork, individual expression, and common goals, values and beliefs. They defined team learning

202

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

as a process through which a group creates knowledge for its members, for itself as a system, and for others. According to these authors, the process includes the following processes: (1) perception of issues based on inputs or past experiences; (2) transforming those perceptions into new understanding; (3) experimenting; and (4) seeking and/or disseminating information with other individuals or units.

teams need to accelerate their learning processes to cope with today’s fast-paced information flow (Maani and Benton, 1999). Maani and Benton (1999) have noted that teams that are operating under the pressures of budget, time, competition, and information explosion, are required to learn rapidly. They have also noted that rapid learning contributes to the success and improved cycle time of teams by drawing an analogy between new-product development teams and the Team New Zealand group of sailors. Similarly, a careful examination of Morone (1993) and Lynn (1993) – in their studies of GE’s development of computed axial tomography (CAT) scanners – reveals that accelerated learning by the development team was critical to the success of this effort. When GE launched its first CAT scanner (a breast scanner) in January 1975, it failed miserably in field trials. The images were poor and the machine could not differentiate between healthy tissue and malignant tumors. In April 1975, GE responded with a head CAT scanner that it acquired from another company named Neuroscan. Unfortunately, this too had severe problems and was shortly withdrawn from the market. Then, in 1976, GE developed a whole-body scanner called the 7800. However, this scanner also failed because the images were unsatisfactory and mysterious rings appeared on the pictures. If there is truth to the saying that “you learn from your mistakes”, GE was one of the smartest companies in the CAT business at that time. GE parlayed its knowledge quickly and, in 1978, launched its fourth major CT initiative – the 8800. The 8800 corrected many of the problems experienced with the 7800, and the market reacted accordingly. GE’s market share jumped from 20 percent to 60 percent, and GE became the dominant CAT supplier. As Morone (1993, p. 61) observed:

Lynn et al. (2000) identified team learning as a process of knowledge generation, dissemination, and implementation. By studying 281 teams, they demonstrated that team-learning processes involve: . recording, filing, and retrieving information; and . developing common, stable, and supported goals for projects. The above empirical team-learning studies demonstrate that team learning is a collective activity of knowledge generation and dissemination involving information gathering, interpreting, and disseminating to achieve the common goal of the team as well as the organization. It follows that accelerated team learning is the faster construction and dissemination of collective new knowledge. The distinction between traditional team learning and accelerated team learning is that accelerated learning considerably speeds-up learning through the process of doing, acting, and sharing information and knowledge quickly. Features or processes of accelerated learning derived from traditional learning for teams are: (1) adapting and responding to environmental changes quickly; (2) acquiring information about customers and competitors quickly; (3) processing and disseminating such information rapidly; and (4) using technology and other methods more efficiently for a successful NPD project.

Consequences of accelerated learning in teams As noted above, since rapid changes and turbulence in market and technologies quickly render obsolete previous information and knowledge (for example, in electronics), NPD 203

Several cases have been written about the history of CT, but they don’t describe anything that I recognize. They tend to project what ought to have been rather than was. There is a tendency to assume that a lot more occurred by planning than what actually occurred . . . In fact, one thing tended to follow from the next. There were a lot of curves on the road that we hadn’t anticipated. We took things as they came. A lot of people think of product development as involving a lot of planning, but I think the key is learning and an organization’s ability to learn.

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

For GE, it was not simply learning, but fast learning from the past failures that helped GE win. In each round of product development, GE moved quickly to fix failures by learning rapidly from past mistakes. Microsoft is another example of a fastlearning organization. The company’s strength does not reside in producing revolutionary “Version 1.0” products. Rather, its strength lies in its team’s rapid development and commercialization of Versions 1.1, 1.5, 2.0, 2.1, and so on. Fast learning from customer feedback and previous versions helps Microsoft’s teams to succeed. Sitkin (1992, p. 245) has also noted that: “Making strategic failure feasible and useful involves insuring that action and feedback happens fast enough that data [are] quickly generated for evaluation and feedback, so that learning can occur expeditiously”. On the basis of the above discussion, the following proposals are advanced: P1. Fast team learning is positively associated with launching a new product rapidly (that is, speed to market). P2. Fast team learning is positively associated with new product success.

how to accelerate new-product development learning processes. However, there is a rich body of knowledge in the individual learning literature on tools and techniques to help individuals learn quickly (see Table I). The accelerated individual learning literature was built on the seminal works of Lozanov who developed techniques collectively called “suggestology” to improve the memory and brain capacity of individuals (Lozanov, 1978; Rose and Nicholl, 1997). Accelerated learning methods emphasize multisensory approaches, music, colorful visual displays, interaction in classroom settings, games, relaxation, and so on to help students to learn more quickly (Rose and Nicholl, 1997). Because of the extensive research that has been completed on accelerated individual learning, perhaps this scholarship can provide a foundation for building an acceleratedlearning new-product development model. However, several organizational theorists have been skeptical about applying individual learning models to an organizational setting. Weick (1991) and Argyris and Scho¨n (1978), for example, have stated that organizational learning is fundamentally different from individual learning. Others have asserted that each requires a different conceptualization (Popper and Lipshitz, 1998). Daft and Weick (1984) have stated that organizations, unlike individuals, do not have brains. Rather, organizations have cognitive systems and memories (norms, culture, filing, and documentation systems). Simon (1991) has emphasized that an individual’s learning ability is more restricted than that of an organization (bounded rationality). He stated that individuals learn within the context of organizations, and that this context affects their learning. Stata (1989, p. 64) explained that organizations can learn only as fast as the slowest link learns (organizational learning builds on past knowledge and experience or memory) and hence is fundamentally different from individual learning. In contrast, several scholars have argued for a link between individual learning and organizational learning. For instance, Kim (1993) developed a conceptual model that links individual and organizational information with mental models to transform individual learning into organizational learning. He demonstrated a conceptual model called OADI-SMM – observe, assess, design, implement, shared mental models –

Antecedents of accelerated learning in teams Over the past several years NPD literature has provided insights into several factors that can help new product professionals innovate and develop new products more quickly. Significant techniques include: . using cross-functional teams (Cooper and Kleinschmidt, 1987; Gupta and Wilemon, 1990); . securing top management support (Karagozoglu and Brown, 1993); . employing concurrent development techniques (Murmann, 1994); . using technological tools (for example, CAD/CAM) (Nijssen et al., 1995); and . cultivating external relationships (for example, licensing, contracting) (Gold, 1987) among others. Although a great deal is known about how to speed up the new-product development processes, surprisingly little is known about

204

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

Table I Literature review of fast learning Author Lozanov (1978)

Study a

Rose (1985)a

Linksman (1996)

Rose and Nicholl (1997)a

Russell (1999)a

Developed new technique called “suggestopedy” to increase the memory, brain capacity, and reasoning ability of individuals. Stated that all instructions and training become pointless if the new knowledge, habits, and skills are not memorized and automated so that they can be used as a basis for future study (p. 6). Suggestopedy started as a psychological experiment designed to increase memory capacities in the educational process (p. 5) by creating learning conditions for speedy and automated (motor) actions. Suggestopedy makes it possible for individuals to use more direct paths than those normally used to penetrate into the mind for information acquisition, and information retrieval by paraconscious psychical activity. Suggestopedy practices were intensively applied to individuals in schools and hospitals. Its application in organizations has been receiving increased attention and many institutions try to apply this concept to individuals within larger groups. However, suggestopedy has not been applied to teams as a whole, and critical factors have not been tested empirically at the team level Definition of fast learning: paraconscious mental activity that can create conditions to automate and use memory, brain, and intellectual reserves of people effectively Important factors: sleep learning, hypnosis, motivation, intonation (music), interaction with environment, communication, non-directive authority, past experiences Described a step-by-step method for individual accelerated learning, and explored the role of memory and the brain on fast learning. Most of Rose’s accelerated learning factors came from the suggestopedy of Lozanov Definition of fast learning: setting up memorable visual and aural associations in the mind as well as subconscious learning (p. 2) Important factors: visual images, music, rhythm, emotions, association and encoding of information, frequent reviews, motivation, imagination, learning by examples Demonstrated how individuals learn faster based on their learning types (visual, auditory, tactile, and kinesthetic) and more active brain hemispheres (left and right) Definition of fast learning: making superlink between left and right brain Important factors: relaxation, motivation, clear objectives, documentation (note-taking), chunking and coding information, visual information, information acquisition, frequent review of past learnings Developed a six-step technique called MASTER to accelerate individual learning. These steps are: motivating an individual, acquiring the information, searching out meaning, triggering memory, exhibiting what individual knows, and reflecting on how individuals learned. In their seminal study, showed how individuals successfully complete each step to accelerate learning. Also emphasized the importance of accelerated learning for business and organizations. However, they concentrated on fast individual learning in organizations, not team learning as one unit Definition of fast learning: the ability to absorb and understand new information quickly and retain that information (p. 18) Important factors: clear goals and objectives, deadlines, fast information acquisition, music, classifying and chunking information to remember easily, frequent review of past information, motivation, support, relaxation, sense of urgency, past experiences, and analogies Explained how individuals learn faster in organizations. She focuses on the training site of accelerated learning in the work environment, and her assertions were based on how instructors can accelerate the learning of trainees. Suggested that instructors should develop different learning techniques for each intelligence type (visual, analytical, etc.) in the workshops. Her suggestions were extensively adapted from the seminal works of Lozanov Definition of fast learning: changing behavior with increasing speed (p. 4) Important factors: clear goals, prioritizing learning objectives, visual information, coding and chunking information (mnemonic), music, relaxed environments, communication

(continued)

205

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

Table I Author

Study

Lawlor and Handley (1996)a

Emphasized importance of learner-centered training by critiquing curriculum-centered training. They suggested tools, techniques and conditions for trainers to accelerate people’s learning. The methods and techniques suggested to corporate trainers were adapted from the accelerated learning work of Lozanov Definition of fast learning: adapting and learning new skills quickly Important factors: goals, motivations, past learning experiences, attractive and relaxing learning environment, interaction with people, knowing and understanding another perspective in the group, visualization, office yoga, note taking, and rehearsing Explored accelerated learning at Bell Atlantic as a new training method. They demonstrated a pilot application of accelerated learning on two customer-service representative training courses and found that using an accelerated learning format reduced the cost of training courses more than 42 percent with improved employee productivity, job satisfaction, shorter training time, and easier update of new courses. Their observations about accelerated learning on people were: higher team spirit, better problem solving ability, greater confidence, more accurate information to customers, and higher sales rate Definition of fast learning: providing effective training in a short period of time (p. 63) Important factors: motivation and positive suggestion, metaphors and mnemonic devices, relaxation exercise, mind maps (information graphs), games, and collaboration

Gill and Meier (1989)a

Note: a We used these studies to build up our theory

to transfer learning among individuals to enhance organizational learning. Hedberg (1981, p. 6) stated that “. . . in fact no theory of organizational learning is based primarily on observations of organizations’ behavior. Instead, experiments with individual humans, mice, and pigeons provide the bases upon which theories of organizational learning are mostly built.” Popper and Lipshitz (1998) postulated several similarities between individual and organization learning. They stated that individuals have cognitive systems (that enable them to think, reflect and act) which are similar to, although not exactly the same as, those of organizations. Laszlo (1972) asserted that there are many operational similarities between the human brain and organizations in their information-processing systems. Lynn and Akgu¨n (2000) modified individual learning constructs – declarative knowledge (knowing what to do) and procedural knowledge (knowing how to do) – and tested these constructs on 137 newproduct teams. They found that declarative and procedural knowledge significantly affected new product success. Garud and Kotha (1994) developed a model based on the workings of the human brain and applied it to flexible manufacturing systems within an organization. Even though the human brain and flexible manufacturing systems are

fundamentally different concepts, Garud and Kotha (1994) demonstrated the theoretical and operational similarities between the two concepts. The present study uses a similar methodology to that of Garud and Kotha (1994) to link individual learning and team learning based on the theoretical similarities between them. Garud and Kotha (1994) used three levels of comparisons (metaphorical level, analogical level, and level of identity) in their conceptual study to link two different domains (source and target). The metaphorical level shows the abstract and insightful similarities of source and target. The analogical level shows the operational similarities between source and target. The level of identity supports the analogical level with a theoretical basis. The present study uses individual learning as a metaphor for team learning, as have other scholars including Morgan (1998), Dodgson (1993), and Hedberg (1981). For the present study, the analogical level demonstrates the operational similarities between accelerated individual and team learning. For instance, individuals have clear goals before performing any task; similarly, teams should have a clear goal before starting a project. Individuals should have support from parents and/or

206

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

teachers for fast learning; similarly, teams need support from top management. The present study first adapts several antecedents of fast team learning from the scholarship on accelerated learning (see important factors in Table I). These are then clustered into groups to formulate a more parsimonious model that includes: . motivating factors; and . information gathering, processing, and transferring factors (see Figure 1).

environment that can be obstacles to rapid team learning. Several scholars (Davenport and Prusak, 1998; Guns, 1996; Meyers and Wilemon, 1989) have also discussed the importance of having a clear vision of the ability of an organization or group to learn. Kessler and Chakrabarti (1996) have argued that teams without a clear vision (having ambiguous project concepts) promote suspicion and conflict in a team about what should be produced, and that this can result in time-consuming readjustments and debates. New product teams thus should strive to establish a clear vision early in the project, and to identify required product futures, target markets, and sales objectives. Although scholars of organizational and group theory have not directly addressed the effect of vision clarity on fast team learning, given the literature in individual learning we propose: P3. Vision clarity is positively associated with fast team learning.

The resulting model of accelerated learning is consistent with the work of Rose and Nicholl (1997).

Motivating factors Motivating factors are the primary conditions that help individuals learn more quickly. These factors are: . having a clear learning goal; . support from parents and teachers; and . an urgent need to learn more quickly.

Vision clarity The literature on individual learning argues that if individuals have clear goals or vision, they can learn their tasks more quickly. Lucas (1998), for example, has stated that a clearly defined vision helps individuals arrange their various priorities and keeps them focused on the task, enabling the individual to learn more quickly. Russell (1999, p. 22) has observed: There is no learning, accelerated or otherwise, without a clear understanding of the scope of the need.

In a similar manner, having a clear team vision should help team members focus better on changes in the market, technology, and Figure 1 Fast team learning model

Management support The individual learning literature asserts that having clear goals or a clear vision is not sufficient for fast learning. Support from parents and teachers also affects the ability of individuals to learn quickly (Berrington, 1997; Rose and Nicholl, 1997). Support from parents and teachers provides motivation for students when they are trying to learn a new task. This finding is consistent with Butty (1997) who stated that each student needs to be motivated to learn. In a team setting, top management assumes many of the same roles as the parent or teacher. Top management is responsible for helping to create a stimulating, nurturing, and supportive environment for fast learning (Guns, 1996). Top management can promote motivation in several ways: . by providing sufficient funds and resources for the teams; . by empowering team members with the necessary authority; . by mentoring team members and team leaders; and . by removing obstacles during the project. In light of the findings from the individual learning scholarship of the positive association of support on rapid learning and the role that top management plays in this support for NPD, we propose that: P4. Top management support is positively associated with fast team learning.

207

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

Aggressive launch date Another motivating factor is urgency. Urgency in the classroom has been shown to stimulate fast learning. For example, Rose and Nicholl (1997) have stated that making problems urgent motivates individuals to learn quickly. A similar sentiment is also emphasized in the management literature. Schein (1993), for example, has asserted that organizations should create a sense of urgency, guilt, or anxiety to speed-up learning. He stated that management should create an atmosphere that the organization is in trouble, that profits are declining, and that competition is becoming more intense. Andy Grove at Intel has tried to instil a sense of paranoia in his employees so that they will learn and act more quickly (Sherman, 1993). His mantra “only the paranoid survive” is well known throughout the company. Similarly, Bill Gates believes his role as chief executive officer of Microsoft is to create a sense of urgency. He has stated that: “As an act of leadership, I created a sense of crisis about the Internet in 1994 and 1995. Not to leave people paralyzed or unhappy, but to excite them into action” (Gates, 1999, p. 181). In NPD teams, a sense of urgency can be created in different ways. One technique to create urgency is by having an aggressive launch date – for example, deadlines for industry shows and international presentations. Jarvenpaa et al. (1998) have stated that a sense of limited and explicit time increases team performance. Meyerson et al. (1996, p. 190) have mentioned that a limited time in the groups reduces jealousy, misunderstanding, and ambiguity in the activities of team members – because there is no time to waste. Moreover, Cooper (1993) has stated that deadlines are critical in new-product development, and that management must be aggressive, causing team members to stretch a little. However, a tight deadline coupled with fast learning can have some negative effects. Because project duration is limited, teams use the information that is quickly gathered. However, such information might be wrong. For instance, simulation studies performed by Herriott et al. (1985) and by Levinthal and March (1981) showed that fast learning can lead to premature information, because the first signal might be wrong. In this sense, if teams want to learn more quickly to be successful, a tight deadline should be determined by team members, rather than by top management. If

top management imposes a deadline, team members merely perform to fulfil the wishes of management, rather than undertaking fast learning for the benefit of the project. Thus, we propose that: P5. Aggressive launch dates are positively associated with fast team learning.

Information gathering, processing, and transferring Learning about customers and competitors As shown in Figure 1, the second cluster of factors is comprised of information gathering, processing, and transferring. Information is an input in the learning equation. Information gathering gives people an opportunity to learn, and therefore an opportunity to act on that information more quickly (Anderson, 1993). For instance, Anderson (1993) has stated that individuals should first gather information (declarative knowledge) to develop interpretive mechanisms (schemata or mental models) (Johnson-Laird, 1983) – which can then be used in different contexts to increase the speed of learning and generate new knowledge. In this vein, information gathering helps people to learn more quickly how to perform new tasks. For instance, Kieras and Bovair (1984) tested whether information gathering can accelerate learning. In their experiment, they divided subjects into two groups. The first group learned the set of operating procedures for the device by rote and another group learned the device model before the identical operating procedures training. Results showed that the second group learned procedures more quickly and more accurately than the rote group, implying that information gathering accelerates learning. For NPD teams, gathering information from customers, competitors, and the environment should enable a team to learn and to act more quickly (Maani and Benton, 1999). Guns (1996) has stated that benchmarking from customers and competitors is essential for fast organizational and team learning. Slater and Narver (1995, p. 67) have stated that the ability to gather information from customers and competitors gives companies an advantage in the speed and effectiveness of their responses to opportunities and threats. Day (1994) has

208

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

argued that when firms learn from customers and competitors, they will have the ability to sense events and trends in their markets ahead of their competitors. Iansiti (1995) has asserted that, during the development stage of a product, continuous acquisition of customer and competitor information and continuous incorporation of this information into prototypes and models help teams to learn more quickly about changing customer needs and competitive reactions. Teams can capture external information by various means – such as by having a rapid and continuous flow of information about customers and competitors by sales people and marketing research personnel, forming a customer council, meeting with users, creating mechanisms for integrating customer requirements into product design (house of quality, quality functional deployment (QFD), and so on), comparing costs and performances of competitive products at every step in the NPD process, and monitoring competitors. In addition, interviews by the present author with team members and managers reveal that top management can foster this by forming a knowledge team to monitor and capture this information. Schein (1993) has suggested a “transition group” and Rothberg (1999) has mentioned a “shadow team” that monitors external environments to capture information about customers and competitors, and that integrates intelligence to create new learning. There can be a person known as a linking pin (gatekeeper) between new-product teams and the knowledge team, transition group, or shadow team. In this way, team members can obtain current external information. This knowledge-management effort should be augmented with training for team members on where information is, and how to access it. Therefore, we propose that: P6. Learning about customers is positively associated with fast team learning. P7. Learning about competitors is positively associated with fast team learning.

product team that reviews past lessons learned should be able to learn more quickly. Team members can do this by reviewing past project files and by meeting with members of other teams both informally and formally. Karagozoglu and Brown (1993, p. 213) have stated that:

Past product review Another technique to aid in gathering information is using past learning experiences. Holyoak and Thagard (1995) have mentioned that individuals who try to match past learning experiences with new conditions are faster learners. Similarly a new-

A commitment toward continuous and frequent improvement driven by a rich repertoire of past experience reduces technical difficulties and NPD delays.

Past product reviews thus provide a cumulative learning base that can help managers and team members to learn more quickly. However, past product review can also be an obstacle for fast team learning under turbulent conditions – due to fastchanging industry dynamics. In this sense, teams should be cautious in their assessment of past projects. Therefore, we propose that: P8. Past product reviews are positively associated with fast team learning.

Information coding After team members have gathered information, the information must be processed before it can be meaningfully shared with others on the team for effective fast team learning to occur. However, information that makes sense to one person might be confusing to another. A situation of student notetaking is a case in point (Hartley, 1983; Kiewra, 1985). If one student takes notes and gives them to another who missed class, the recipient is likely to have difficulty understanding them if he or she simply copies them word for word (Kiewra, 1985). To increase the knowledge transfer, the recipient must re-state and re-code the information so that he or she can internalize the data and turn them into usable information. In a team setting, organizing or coding information into meaningful clusters can affect the effectiveness and the rate at which teams learn. Coding information includes labeling, indexing, sorting, abstracting, and categorizing (Zack, 1999). Davenport and Prusak (1998) explained that codification organizes information and knowledge – making it easier to understand and communicate. Tidd (1997) has noted that the speed and extent of knowledge sharing among members of an organization is a function of knowledge codification. After capturing information about customers and

209

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

competitors, teams should codify or classify this information into meaningful clusters so that people can understand and internalize it. When information is congruent with the mental model or schema, individuals, teams, and organizations make sense of that information more quickly, leading to speedy learning. In this vein, an individual or a small team, such as a knowledge team, can be formed for the purpose of codifying and clustering information into topics, sources, types, or importance levels to create congruency with current understanding (Garvin, 1997). Then, these information clusters should be entered into a central database that everybody can easily access (Garvin, 1997). Training in the use of this information is critical. Otherwise companies might find that they have storehouses of unused information. Therefore: P9. Information codification is positively associated with fast team learning.

members to keep this information active in their short-term memories. However, the frequency of reviews should be appropriate to the different stages of a project. At the beginning of a project, daily reviews might be helpful in formulating the project vision and charter. However, as the project becomes more established, daily reviews might slow down learning because team members know what to do – they have their “orders” and need to be given the time to accomplish them. Interviews conducted by the present author with project managers, and discussions with scholars, show that teams with formal daily meetings tend to be those that were not effectively communicating information on an ongoing, informal basis in the course of their daily work. Taking all of the above into account, it is proposed that: P10. Daily reviews are positively associated with fast team learning.

Daily reviews The final factor to accelerate learning is frequent review of the material that the team has collected. In the individual learning environment, if parents and teachers frequently enquire about what students have learned during the day, individuals can learn new tasks faster, because frequent reviews of past events keep individuals’ short-term (working) memories active (Anderson, 1993). An active short-term memory allows individuals to use less effort in recalling thoughts because having a larger short-term memory makes it easier to establish links between the new information and long-term memory (Anderson, 1993). The analogy is similar to using computer ready-access memory (RAM) as opposed to the use of a computer’s long-term memory in the hard drive. Accessing RAM takes far less time (Rose and Nicholl, 1997). In a similar fashion, individuals can learn new tasks more quickly by having a larger and more active short-term memory. In a new-product team setting, conducting daily reviews can help keep knowledge and information in short-term active memory and thus increase fast learning. For instance, analyzing the previous day’s meeting reports and action items, reviewing information about customers, competitors, suppliers, and available technologies, and checking technical reports can all help team

Conclusion and future directions This paper has emphasized the importance of fast learning on new-product development, and has demonstrated that fast learning is a critical component of such development. Because fierce market and technological environments make team learning a standard requirement for projects, organizations can be successful in their new products only if they learn more quickly than their competitors. The present study has therefore modified some of the “suggestopedy” variables to help teams to learn more quickly in their productdevelopment efforts. However, future studies could expand and empirically test this model. These investigations could include the effects on accelerated learning of: . visual communication (for example, video conferencing); . the simplification of new-product development processes (for example, using standard components); . informal communications; . experimentation; . learning by doing (for example, team improvisation); . team autonomy; . simulation tools; . CAD/CAM; . rapid prototyping; and . team unlearning or belief changes.

210

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

One of the applications of “suggestology” is to improve the individual’s intelligence and to create a “super-human” (Lozanov, 1978). With appropriate modification, methods and models can be developed to help the newproduct development team to increase its intelligence or to create an intelligent newproduct development team. This study has made a modest beginning in this important, but understudied field. Future researchers will find this area of accelerated learning to be rich and fruitful.

Gates, B. (1999), Business at the Speed of Thought, Warner Books, New York, NY. Gill, M.J. and Meier, D. (1989), “Accelerated learning takes off”, Training and Development Journal, pp. 63-5. Gold, B. (1987), “Approaches to accelerating product and process development”, Journal of Product Innovation Management, Vol. 4, pp. 81-8. Grant, R.M. (1996), “Prospering in dynamicallycompetitive environments: organizational capability as knowledge integration”, Organization Science, Vol. 7, pp. 375-87. Guns, B. (1996), The Faster Learning Organization, JosseyBass, San Francisco, CA. Gupta, K. and Wilemon, D.L. (1990), “Accelerating the development of technology-based new products”, California Management Review, Winter, pp. 24-44. Hartley, J. (1983), “Notetaking research: resetting the scoreboard”, Bulletin of the British Psychological Society, Vol. 36, pp. 13-14. Hedberg, B. (1981), “How organizations learn and unlearn”, in Nystrom, P.C. and Starbuck, W.H. (Eds), Handbook of Organizational Design, Vol. I, Oxford University Press, Oxford. Herriott, S.R., Levinthal, D. and March, J.G. (1985), “Learning from experience in organizations”, American Economic Review, Vol. 75, pp. 298-302. Holyoak, K. and Thagard, P. (1995), Mental Leaps: Analogy in Creative Thought, MIT Press, Cambridge, MA. Iansiti, M. (1995), “Shooting the rapids: managing product development in turbulent environments”, California Management Review, Vol. 38 No. 1, pp. 37-58. Jarvenpaa, S., Knoll, K. and Leidner, D.E. (1998), “Is anybody out there? Antecedents of trust in global virtual teams”, Journal of Management Information Systems, Vol. 14, pp. 29-64. Johnson-Laird, P.N. (1983), Mental Models, Harvard University Press, Boston, MA. Karagozoglu, N. and Brown, W. (1993), “Time-based management of the new product development process”, Journal of Product Innovation Management, Vol. 10, pp. 204-15. Kasl, E., Marsick, V.J. and Dechant, K. (1997), “Teams as learners: a research-based model of team learning”, The Journal of Applied Behavioral Science, Vol. 33, pp. 227-46. Kessler, E.H. and Chakrabarti, A.K. (1996), “Innovation speed: a conceptual model of context, antecedents, and outcomes”, Academy of Management Review, Vol. 21, pp. 1143-91. Kieras, D. and Bovair, S. (1984), “The role of a mental model in learning to operate a device”, Cognitive Science, Vol. 8, pp. 255-73. Kiewra, K.A. (1985), “Investigating notetaking and review: a depth of processing alternative”, Educational Psychologist, Vol. 20, pp. 23-32. Kim, D.H. (1993), “The link between individual and organizational learning”, Sloan Management Review, No. Fall, pp. 37-50. Kogut, B. and Zander, U. (1992), “Knowledge of the firm: combinative capabilities, and replication of technology”, Organization Science, Vol. 3, pp. 383-97. Laszlo, E. (1972), Introduction to Systems Philosophy, Gordon and Breach, New York, NY.

References Anderson, J.R. (1993), Rules of the Mind, Erlbaum Associates, Hillsdale, NJ. Argyris, C. and Scho¨n, D.A. (1978), Organizational Learning: A Theory of Action Perspective, AddisonWesley Publishing, Reading, MA. Berrington, L. (1997), “Parents: do I have to be a genius?”, The Guardian, 2 January. Brooks, A.K. (1994), “Power and the production of knowledge: collective team learning in work organizations”, Human Resource Development Quarterly, Vol. 5 No. 3, pp. 213-34. Butty, D.C. (1997), “Education: Sylvan teams up with city schools to improve students’ performance”, Detroit News, p. S-3:1. Cooper, R.G. (1993), Winning at New Products: Accelerating the Process from Idea to Launch, Addison-Wesley Publishing, Reading, MA. Cooper, R.G. and Kleinschmidt, E.J. (1987), “Success factors in product innovation”, Industrial Marketing Management, Vol. 16, pp. 215-23. Daft, R.L. and Weick, K.E. (1984), “Toward a model of organizations as interpretation systems”, Academy of Management Review, Vol. 9, pp. 284-95. Dandrea-O’Brain, C. and Buono, A.F. (1996), “Building effective learning teams: lessons from the field”, Advanced Management Journal, Vol. 61, p. 4. Davenport, T.H. and Prusak, L. (1998), Working Knowledge: How Organizations Manage What They Know, Harvard Business School Press, Boston, MA. Day, G.S. (1994), “The capabilities of market-driven organizations”, Journal of Marketing, Vol. 58, pp. 37-52. De Geus, A.D. (1997), “The living company”, Harvard Business Review, March-April, pp. 51-9. Dodgson, M. (1993), “Organizational learning: a review of some literatures”, Organization Studies, Vol. 14 No. 3, pp. 375-8. Edmonson, A. (1999), “Psychological safety and learning behavior in work teams”, Administrative Science Quarterly, Vol. 44, pp. 350-83. Garud, R. and Kotha, S. (1994), “Using the brain as a metaphor to model flexible production systems”, Academy of Management Review, Vol. 19, pp. 671-98. Garvin, D.A. (1997), Putting the Learning Organization to Work: Case Study of US Army, Harvard Business School Video, Harvard Business School, Boston, MA.

211

Accelerated learning in new product development teams

European Journal of Innovation Management

Gary S. Lynn, Ali E. Akgu¨n and Halit Keskin

Volume 6 · Number 4 · 2003 · 201-212

Lawlor, M. and Handley, P. (1996), The Creative Trainer: Holistic Facilitation Skills for Accelerated Learning, McGraw-Hill, Maidenhead. Leonard-Barton, D. (1995), Wellsprings of Knowledge: Building and Sustaining the Source of Innovation, Harvard University Press, Cambridge, MA. Levinthal, D. and March, J.G. (1981), “A model of adaptive organizational search”, Journal of Economic Behaviour and Organization, Vol. 2, pp. 37-333. Linksman, R. (1996), How to Learn Anything Quickly: An Accelerated Program for Rapid Learning, Citadel Press, Secaucus, NJ. Lozanov, G. (1978), Suggestology and Outlines of Suggestopedy, Gordon and Breach, New York, NY. Lucas, J.R. (1998), “Anatomy of a vision statement”, Management Review, Vol. 87 No. 2, pp. 22-6. Lynn, G.S. (1993), “Understanding products and markets for radical innovations”, unpublished PhD dissertation, Rensselaer Polytechnic Institute, Troy, NY. Lynn, G.S. and Akgu¨n, A.E. (2000), “A new product development learning model: antecedents and consequences of declarative and procedural knowledge”, International Journal of Technology Management, Vol. 20, pp. 490-510. Lynn, G.S., Reilly, R.R. and Akgu¨n, A.E. (2000), “Knowledge management in new product teams: practices and outcomes”, IEEE Transactions on Engineering Management, Vol. 47, pp. 221-31. Maani, K. and Benton, C. (1999), “Rapid team learning: lessons from Team New Zealand America’s Cup campaign”, Organizational Dynamics, Vol. 27, pp. 48-62. Maira, A. and Scott-Morgan, P. (1997), Accelerating Organization: Embracing the Human Face of Change, McGraw-Hill, New York, NY. Meyers, P.W. and Wilemon, D. (1989), “Learning in new technology development teams”, Journal of Product Innovation Management, Vol. 6, pp. 79-88. Meyerson, D., Weick, K.E. and Kramer, R.M. (1996), “Swift trust and temporary group”, in Kramer, R.M. and Tyler, T.R. (Eds), Trust in Organizations: Frontiers of Theory and Research, Sage Publications, Thousand Oaks, CA. Moorman, C. and Miner, A.S. (1997), “The impact of organizational memory on new product performance and creativity”, Journal of Marketing Research, Vol. 34, pp. 9-16. Morgan, G. (1998), Images of Organization: The Executive Edition, Sage Publications, Thousand Oaks, CA. Morone, J. (1993), Winning in High-tech Markets, Harvard Business School Press, Boston, MA. Murmann, P.A. (1994), “Expected development time reduction in the German mechanical engineering industry”, Journal of Product Innovation Management, Vol. 11, pp. 236-52.

Nijssen, E., Arbouw, A.R.L. and Commandeur, H.R. (1995), “Accelerating new product development: a preliminary empirical test of a hierarchy of implementation”, Journal of Product Innovation Management, Vol. 12, pp. 99-109. Nonaka, I. and Takeuchi, H. (1995), The Knowledgecreating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford University Press, New York, NY. Popper, M. and Lipshitz, R. (1998), “Organizational learning mechanisms: a structural and cultural approach to organizational learning”, The Journal of Applied Behavioral Science, Vol. 34, pp. 161-79. Prusak, L. (1997), “Introduction to knowledge in organizations”, in Prusak, L. (Ed.), Knowledge in Organizations, Butterworth-Heinemann, Newton, MA. Rose, C. (1985), Accelerated Learning, Dell Publishing, New York, NY. Rose, C. and Nicholl, M.J. (1997), Accelerated Learning for the 21st Century, Dell Publishing Group, New York, NY. Rothberg, H.N. (1999), “Fortifying strategic decisions with shadow teams: a glance at product development”, Competitive Intelligence Magazine, Vol. 2, pp. 9-11. Russell, L. (1999), The Accelerated Learning Fieldbook, Jossey-Bass Pfeiffer, San Francisco, CA. Schein, E.H. (1993), “How can organizations learn faster? The challenge of entering the green room”, Sloan Management Review, Winter, pp. 55-62. Senge, P. (1990), The Fifth Discipline: The Art and Practice of the Learning Organization, Currency Doubleday, New York, NY. Sherman, S. (1993), “Andy Grove: how Intel makes spending pay offs”, Fortune, 22 February. Simon, H. (1991), “Bounded rationality and organizational learning”, Organization Science, Vol. 2, pp. 125-34. Sitkin, S.B. (1992), “Learning through failure: the strategy of small losses”, Research in Organization Behavior, Vol. 14, pp. 231-66. Slater, S.F. and Narver, J.C. (1995), “Market orientation and the learning organization”, Journal of Marketing, Vol. 59, pp. 63-74. Stata, R. (1989), “Organizational learning – the key to management innovation”, Sloan Management Review, Spring, pp. 63-74. Tidd, J. (1997), “Complexity, networks and learning: integrative themes for research on innovation management”, International Journal of Innovation Management, Vol. 1, pp. 1-21. Weick, K.E. (1991), “The nontraditional quality of organizational learning”, Organization Science, Vol. 2, pp. 116-24. Zack, M.H. (1999), “Managing codified knowledge”, Sloan Management Review, Summer, pp. 45-58.

212

Introduction

The patterns of success in product development: a case study Petri Suomala and Ilkka Jokioinen

The authors Petri Suomala is Senior Researcher, Institute of Industrial Management, Tampere University of Technology, Tampere, Finland. Ilkka Jokioinen is based at Metso Paper Inc. Turku, Turku, Finland. Keywords Research, Product development, Performance measurement (quality), Case studies, Metalworking industry Abstract Success in product development can be considered a general aim for any R&D activity. Unfortunately, success is very multidimensional. Which dimensions of success one should include and how one can measure these dimensions is an essential question that must be resolved within R&D management. The aim of the paper is to analyse the multidimensionality of success at R&D project level. Using the evidence from three industrial cases, the paper shows the versatility of the concept of success in product development. On the basis of the evidence, the paper concludes that there is often only a vague correlation between three aspects of R&D success – financial, technical and project management. The focus of the paper is on the development of investment goods for the paper industry. Electronic access The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1460-1060.htm

European Journal of Innovation Management Volume 6 · Number 4 · 2003 · pp. 213-227 q MCB UP Limited · ISSN 1460-1060 DOI 10.1108/14601060310500931

Industrial research and development (R&D) utilizes science and technology to construct new or improved products or processes for profit-seeking companies (IRI, 2000). Product development, which is an essential part of R&D, can be seen as an activity that is expected to improve a company’s competitive advantage and future success in terms, for example, of profitability and market share. Based on the hope and trust that tangible returns will be greater than expenditure, considerable sums of money are spent on R&D (Batty, 1988). According to IRI, in US companies alone, representing over one-third of the entire world’s allocation in R&D, US$185.9 billion was invested in industrial research and development in 1999 (IRI, 2000). For comparison, US R&D expenditure in 1950 was US$2.5 billion (Jackson and Spurlock, 1966). With the increase in investments, the question of R&D management and measurement has gained in interest. Performance measurement has been considered a relevant aid in the pursuit of success in product development. Therefore, a wide body of literature concerning the measurement of R&D performance and success has been produced. Some writers (McGrath and Romeri, 1994; Chiesa and Masella, 1996; Kerssens-van Drongelen and Bilderbeek, 1999; Szakonyi, 1994a, b) have established holistic approaches for the assessment of R&D effectiveness, while others have concentrated on project level (Ormala, 1986; Rouhiainen, 1997). There are advocates of continuous or in-process monitoring of development as well as end-ofprocess evaluation (Schumann et al., 1995). There is a variety of methods available for R&D project selection (Cooper, 1985; Hollander, 2000), performance evaluation for managerial purposes, customer perspective (Hirons et al., 1998; Nixon, 1998) and benchmarking. Many creditable reports that describe the state-of-the-art of R&D measurement have been published (Brown and Svenson, 1998; EIRMA, 1985, 1995; Werner and Souder, 1997a, b; Griffin, 1997). While several decades of R&D studies have produced a good deal of data with respect to variables associated with the success and failure of new products, the research has not been able to resolve a practical problem: how should R&D actually be managed in order to

213

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

promote high new product success rates. Poolton and Barclay conclude that managers are still relying on gut-feelings regarding “best practice” in new product development (NPD). Analogously, research has tended to be theory-driven instead of being applicationsbased (Poolton and Barclay, 1998). Driva et al. conclude that in most cases companies do not measure R&D activity very well, but that they are striving to determine how to do so effectively (Driva et al., 2000). In this respect, it seems fair to claim that a good deal of work is still needed to improve the efficiency of the interface between industrial R&D management and academic R&D research. Success in product development can be considered a general aim for any R&D activity. Unfortunately, success is very multidimensional. The question of which dimensions of success one should include and how one can measure these dimensions is an essential question that must be resolved within R&D management (Hultink and Robben, 1995). However, relatively little discussion has focused on the resolution of this question. Yet, as Hart (1993) puts it: Clearly, the way in which NPD success is defined influences the findings which describe the factors contributing to NPD success.

Griffin and Page recognize that success is elusive, multifaceted and difficult to measure. Still, companies and academics use over 75 measures of success in product development (Griffin and Page, 1996). Basically, hand in hand with determining and selecting R&D performance measures for a company, one should also consider the concept of success. What is the form of success that is primarily pursued? Are there any other success dimensions that would be important for us? Knowing the type of success pursued would likely be helpful in choosing the appropriate set of R&D metrics.

R&D processes contributing to new product success From the perspective of R&D management, in practice, seeking new product success might be challenging, seeing that prospects of success likely vary from phase to phase in the R&D process. In this study, the model presented in Figure 1 (Matthews, 1991) has been adopted as a basis that describes the fundamental processes of research and

Figure 1 The process of R&D

product development. In Matthews’ model, R&D is a process that consists of three main phases. It starts with technology development and ends with actual product development that results in a marketable product. Two essential dimensions are established: (1) technological uncertainty; and (2) the amount of allocated resources. If the R&D process proceeds, it is assumed that the technological uncertainty will decrease – i.e. the prospects of new product success should be improved. Typically, as the product gets closer to the market more resources are allocated to the development project. According to Matthews, completing the whole R&D process requires answering five questions: (1) Is it possible: (2) Is it attractive? (3) Is it practical? (4) Is it desirable? (5) How do we do it? The challenges of R&D are likely to differ from phase to phase. Projects belonging to category C have to be managed differently than projects belonging, for instance, to category A. This is because research projects (category C) typically have somewhat different goals, objectives and restrictions than specific product (category A) or concept development projects (category B). Research projects may often deal with technological issues that cannot yet be attached to certain industrial and marketable products – they are mainly serving as a “breeding ground for promising tadpoles”. Only time will tell whether these technologies will lead to commercial products – that is, technological uncertainty is rather high. According to

214

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

Matthews (1991), the costs of these types of projects are typically seen as “overheads”, and the projects are allowed to yield results over a rather long time scale, if ever. On the other hand, the technological uncertainty of product development projects (category A) should be significantly lower, but at the same time the objectives of the projects should have more concrete goals, including the financial ones, than research projects. As Matthews (1991) puts it: “short-term development (in a typical company) is seen as investment”. The most problematic area, however, is the gap between “overhead” and “investment”. These category B projects may often fail to show sufficient justification for funding, since they neither represent a pure “breeding ground” any more, nor have they yet reached the status of “investments” that could be assessed using sound financial measures. These projects can be seen as “strategic options” and the challenge is to identify the most promising and practical ones that have the potential to proceed to actual product development. Matthews’ work provides one valuable insight into the classification of R&D projects or activities, but traditional classifications need to be discussed as well. Batty, for example, has provided a traditional definition of R&D by recognizing three elements: (1) basic research; (2) applied research; and (3) product development (Batty, 1988).

phases. The process starts with concept development, as the writers exclude research activities from the actual development process. The four sequential phases are: (1) system level design; (2) detail design; (3) testing and refinement; and (4) production ramp up (Ulrich and Eppinger, 1995, p. 15).

McLeod (1988) also adds a fourth element – design, which starts after development. A slightly different categorization is proposed by Jackson and Spurlock (1966). They have identified fundamental research, applied research and developmental research (Jackson and Spurlock, 1966). R&D projects, as seen in this case study, mostly belong to the development, design or applied research category. Fundamental or basic research is not so typically conducted at company level. However, many of the applied research or development projects are originally triggered by co-operation with a company and a university or research institute in the basic research area. Defining a detailed research and product development process is a complicated task. Sometimes it is difficult to distinguish which activities should be conducted in a certain R&D process step. However, Ulrich and Eppinger (1995) have established a generic development process that consists of five

According to Ulrich and Eppinger (1995), in the concept development phase many functions of a firm are allotted several tasks and responsibilities. The following concept development related tasks are recognized by function (Ulrich and Eppinger, 1995): (1) Marketing: . define market segments; and . identify lead users and competitive products. (2) Design: . investigate feasibility of product concepts; and . develop industrial design concepts. (3) Manufacturing: . estimate manufacturing costs; and . assess production feasibility. (4) Other: . finance, facilitate economic analysis; and . legal, investigate patent issues. Concept development, as seen from the previous list, should put emphasis on economic issues in analyzing and resolving the potential of a product or solution for the company’s business. In this analysis, customers and production are to be taken into account. Issues not explicitly mentioned above, but still worth investigating in the concept development phase, are corporate strategy and business as a whole. Questions such as “Does the product fit well into our strategy?” and “How exactly will the product strengthen our overall business?” should be the concern in concept development. If a development project passes the assessments conducted in the concept development phase, it will proceed to product development. The main input to product development is the outcome of concept development, the main objective of activities being a commercial and marketable (profitable) product. Thus, product development resolves the detailed final design and prepares the whole company for the

215

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

product launch. In real life, organizing an R&D process may not be as straightforward as described above. Phases are likely to be mixed and partly overlap each other. Importantly, new product success is not something that is considered only in the product development phase. Success should be seen rather as a goal that is striven for in all the phases that constitute the R&D process in the company. Each phase may make a contribution to the overall success.

selecting R&D projects, including: probability of success; time to first sales; profitability; and compatibility with the company’s long-term plans (strategy). However, it is argued that there is little point in trying to give the factors any order of priority. According to McLeod (1988, p. 254), all these factors should be considered as a whole. The construction of numerical indices, struggling with figures and scoring the projects is not seen as beneficial. Cooper and Kleinschmidt (1995) point out that new product performance is a multidimensional concept. Therefore, a single measure for NPD performance monitoring may not be enough. However, many companies still use only one or two measures. Having said that multiple measures are better than a single one, Cooper and Kleinschmidt (1995) continues that many of the measures may not be totally independent of each other. A holistic view of new product performance can be reached with a limited number of measures. The body of literature on new product success suggests that there is a rather versatile set of factors affecting this success. To structure the field, different domains of groups of factors related to NPD success are illustrated in Figure 2. The brackets refer to literature that supports a particular perspective. In this synthesis, the success factors are collected into four perspectives that demonstrate not only the elusive nature of success factors but also the different viewpoints of various stakeholders in the NPD process. The top management of the company, for instance, may assess the NPD activities and outputs through a strategic framework: is there a sufficient fit between the NPD and the corporate or technology strategy, or does the NPD produce essential strategic value for the company. At the level of business units the concerns of management

Selecting development projects that will lead to success Often the number of potential research, development and design projects is greater than it is possible to carry out. The limited resources and skills compel the managers to select projects from those proposed. A comprehensive description of different selection methods is presented by Martino (1995). Ellis (1997) argues that without objectively measuring the process of innovation one is not able to determine whether the expenditures on R&D are beneficial or not. Both the desired outcomes as well as the inputs and R&D processes that contribute to these outcomes should be measured (Ellis, 1997, p. 3). Cooper (1985) has developed the NewProd model for separating probable successful projects from probable losers. He remarks that project selection is pivotal to effective risk reduction in product development. A scoring model could be a valuable tool in screening proposals. According to the NewProd model, product superiority/ quality, market need, growth and size and product scope are the factors that have the strongest impact on probability of success (Cooper, 1985). Hollander (2000) has reported the potential of the Genesis model for project assessment. His study is based on Cooper’s NewProd studies. The objective of both of these models is to provide support for the product development team, especially for “go or no go” decisions. The Genesis model is focused on development projects and teams; the question is does the team have the necessary resources and skills and how the product is positioned in respect of markets and competitors’ products (Hollander, 2000). McLeod (1988) suggests several factors that should be taken into account when

Figure 2 Perspectives having a contribution to new product success

216

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

may include the consistency between the present business and potential produced by NPD: what are the anticipated business impacts caused by or the risks posed by the NPD. Furthermore, inside the technology and product development unit there might be pure technological interests in addition to the economic ones. Throughout the company there are – or at least should be – an interest in maintaining and developing the capability of NPD to produce value for the customer through innovative and competitive new products and services. With regard to the practical organizational context, strategy is one of the core issues related to the success of NPD. The management of NPD has to take into account the corporate strategy and its implications. Projects that are consistent with the strategy are likely to receive the most support from top management. On the other hand, strategy should be flexible enough to enable the initiation of projects that seem to be very promising but that may not yet be at the core of corporate strategy. In addition to corporate level strategy, an R&D strategy may provide arguments for internal marketing of a project, thus facilitating, e.g. a sufficient resource allocation for the project. Second, a product’s relative position with respect to the target market seems to be an important determinant of success. The new product’s competitive position – the product’s competitiveness against competitors’ products – should be therefore carefully considered. Timing of the product launch is one of the things that affect the product’s competitive position. The third perspective affecting the product’s success is technology. Is the new technology mature enough to be manufactured and delivered? Does the customer have the ability to adopt the new technology, and what is the additional value of innovation for the customer? The fourth area is company business. A development project may have synergies with other projects, or it may lead to cannibalizing some existing business. Thus, the business impact of the new product should also be considered. Also, in order to be able to extract fully the success potential of the new product, some fit between the new product and the company’s present business is required. Otherwise, the supply chain, manufacturing process and customer base would have to be subjected to a dramatic adjustment for the new product.

It is suggested that there is a relevant link between the ideas presented in Figures 1 and 2. As Matthews (Matthews, 1991) has stated, for the projects that are neither pure research (category C) nor actual product development (category A) it might be challenging to establish sound arguments that either support or oppose further development efforts. As one solution for this managerial problem, a holistic framework – such as the one presented in Figure 2 – including the identified success factors of NPD might be a useful tool for structuring the issue. Especially in the assessment and in the prioritization of the category B development projects, a decision-maker should consider various factors that represent the viewpoints of different stakeholders in the NPD process. If each of the projects at hand is evaluated from the perspectives presented in Figure 2, it should be possible to more rationally – or at least more explicitly – rank the projects based on the anticipated success factors. Of course, this kind of assessment can probably never be totally objective or definite; however, it is supposed to be an attempt to systematize organizational discussion and decisionmaking regarding projects: . whose justification is not self-evident; and . that cannot be (yet) evaluated using direct financial criteria.

Objective and methods The aim of the paper is to analyse the multidimensionality of success at the R&D project level. Success is considered as a variable that is connected with the NPD process and its phases. That is, success as a goal of NPD can be seen very differently, depending on the phase of the R&D process. The paper also addresses the issue that there are practically meaningful patterns of success, which are also very different from each other, that pose challenges for R&D managers who are striving for effective and generally successful NPD. A core problem is related to the correlation between different success domains: good performance in one success domain may or may not be related to success in another. The paper is based on a case study conducted in the metal industry. All three cases studied are product development projects involving investment goods – products of the metal industry companies A

217

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

and B serving customers in the papermaking industry. In the customers’ business the growth rate is typically between 2 and 4 percent. Big investments in production lines affect the market demand for paper grades, which results in some cyclical fluctuations in the market demand for production machinery. Another essential characteristic of the business is the length of product life cycles (PLCs), which typically vary between ten and 15 years. The challenge to PLC management in this industry is underlined by the fact that the introduction of totally new technology can typically take ten years or more, mainly because of the risks of ruining the payoff of the entire production process when technology is introduced too early. The difficulty and complexity of the technology transfer process from phase C to A in this particular industry area is reviewed with several cases in the paper by Crotogino (2000) in IDS 2000. New technological solutions are commonly introduced in the industry first by rebuilds of paper machines and then finally by new production line deliveries. Thus, feedback from markets and learning by experience tend to be slow and time to market long, at least in the case of major components of the production line. The learning process and the process of managing the risks in the development of components for the papermaking line are improved by demonstrating the potential of equipment on pilot machines. All the product development cases in this paper were carried out in the 1990s and the market launch of products took place in the late 1990s. In each case the following components of success – discussed by Rouhiainen in his dissertation focusing on the management of NPD project implementation in the metal industry (Rouhiainen, 1997) – were evaluated: . project management success; . technical success (businessperson’s, engineer’s and client’s point of view); and . commercial success estimate.

Slevin and Pinto have provided a reference score scale for each success factor that enables benchmarking a certain project performance with the success profiles of known successful projects. (For instance regarding the factor of communication, the 0th percentile of 12 points refers to the fact that none of the studied success projects scored less than 12 points, and the 100th percentile of 99 indicates that the full score of 100 was not achieved by any project.) If a project’s performance in the case of any factor is below the 50th percentile, one should – according to Slevin and Pinto (1986) – devote extra attention to that factor to improve the odds of success. As regards the other success areas, technical as well as commercial success has been analysed in this study by interviewing the key development personnel and management of NPD projects as well as by studying the company’s NPD reporting and documentation. Concerning the theoretical representativeness of the study, the cases studied are quite different from each other. Case 1 was basically a new technology development project, the outcome of which was a new product utilizing technologies known in company A. Case 2 was a product development project of company A aimed at introducing a product similar to but better than existing ones in the marketplace. Case 3 involved introducing an incrementally improved product using technologies known in the companies participating in the development project (companies A and B). In case 3, company B is a supplier of company A. In Table I the cases are outlined in terms of main development characteristics. The key parameters utilized in Table I are defined and discussed by EIRMA (1995); it also illustrates the diversity of the cases in compact form. The rationale for using a case study approach is to provide holistic and meaningful characteristics of real-life events (Yin, 1994). According to Yin (1994), a case study is an empirical inquiry that investigates a contemporary phenomenon within its real-life context. Typically, in case study research, the boundaries between phenomenon and context are not clearly evident. In this study, a product development project is the phenomenon to be investigated and the context is a company’s product development organization and its environment. A holistic

Project management success has been estimated utilizing the PIP questionnaire constructed by Slevin and Pinto (1986). Slevin and Pinto (1986) have identified ten success factors, including for instance project mission, top management support, client consultation, communication and troubleshooting, that are related to a project’s success. By studying 82 successful projects,

218

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

Table I Characteristics of the cases as they were seen in the phase B when development decisions were made

Technologies applied Type of R&D Market Feedback from market Time to market Competition Innovation driver

Case 1

Case 2

Case 3

Known by company A Radical Known Low Long Medium Technology push

Known by company A Incremental Less known Medium Medium High Regulations

Known by companies A and B Incremental Known Medium Medium High Market price/Technology push

view is needed when making R&D decisions because the results of development work would greatly influence the whole company and its business potential in the future. Therefore, the interests of various stakeholders should be somehow considered in the research and development process.

Case evidence Case 1 Case 1 was originally a technology development project established to reduce the investments in a paper production line. The result of the technology and product development project was a commercial product making possible up to 30 percent savings in machinery investments when the product is utilized in full scale. The development history (phases C, B, A described by Matthews (1991)) can be summarized as follows. Phase C lasted four years. It can be described as applied research aiming to study if the technology could be utilized in an application area. The research was conducted both by the company and by external research laboratories as well as technical universities. This phase was focused on minimizing the risks of choosing the wrong technology for the concept and development phase. Phase B lasted three years. The objective of the concept development phase was to evaluate the feasibility of the product concept as a strategic option. During this period, pilotscale equipment for this technology was built and tested. This was done to further minimize the risk associated with the new technology and to evaluate its applicability for several customers. The most critical components of the technology were tested and designed in phase B. Phase A lasted two years. It ended with the first customer delivery of the new technology

product. During this stage the product was developed and designed mainly within the delivery project. However, many design features and solutions were based on the detailed studies made in phase B. Generally, passing a development project from concept development to the actual product development phase requires an estimation that there is enough potential in the concept – in respect of risks associated with it – to achieve an acceptable level of success. In this case, concept development for the product included many (technical) risks that were also identified: . the effect of the technological solutions characteristic of the concept on the quality of paper grade; . the implications of the concept for runnability issues; . the effects of the concept especially on the durability of core peripherals that are in interaction with the product. However, despite the risks, concept development was encouraging to the extent that it was decided to proceed to product development. The selection of the project for actual product development was mainly based on the following criteria (estimated benefits of the product): . Possibility for the customer to have lower investment costs. . The quality and runability of the paper grade remains at the very least the same. . Positive impact on the efficiency of the production (customer process). . Simplicity: good product architecture, low number of components. . Lowering energy consumption. Project management success in case 1 was studied by implementing a reduced PIP questionnaire (see Slevin and Pinto, 1986) among the key people of the development personnel from phase B to phase A. The questionnaire was supplemented by

219

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

interviews conducted in the same target group. The PIP questionnaire revealed the perception that the project was fairly well managed. The weakest identified critical success areas were (see also Figure 3): . Client consultation and client acceptance; evidently, this domain should have been studied better and more comprehensively in phase B. One reason for the poor or absent customer consultation was the confidential nature of the project in this phase. . The schedule/plan, which can be partially explained by the length of the project and changing personnel during the project. . The project mission, which was to some extent unclear. This was mostly due to the fact that the project dealt with fairly radical development, which caused difficulties in specifying and operationalizing the exact goals for the project personnel.

The commercial success of the new technology product so far has been poor: no further references have been obtained two years after the first one. However, interest among customers has been fairly high. The lack of commercial success is also due to competing products. Another reason is probably the timing: the release of the new technology product in case 1 took place too early and was not properly synchronized with other product releases. The product as such is an outcome of long-term technology strategic choices of the company as the technology leader in the industry. The company is still considering additional investments in order to attain the strategic targets set many years ago for the product. Observations from the interviews (n ¼ 4) with key development personnel and product development management can be summarized as follows: . Strategy implications: in line with the established strategy, but payoff is still a question, even if the technology project has lasted a very long time. . Customer and competition: price competitiveness of the product is the major concern. . The business impact of the product is rather fuzzy; there are also some competing products within the same company. . The development project leader feels unsure regarding the customers’ ability to derive full benefit from the product.

What can be said about the technical success on the basis of the inquiry and interviews? From the technical point of view, the product developed was a success in the first delivery in almost every respect. The technical specification and process performance targets set by the customer were met. The new product created in the project made it possible to cut the investment costs by 30 percent for the production line when utilized in full scale. From the businessman’s point of view the manufacturing costs of the product could have been lower in order to be able to meet the market price of competing technology. Figure 3 PIP – profiles of the studied cases

220

.

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

Technology: the profitability of the project appears weak although the technology is considered mature enough.

Case 2 Case 2 was a spin-off of an applied research project intended to reduce the environmental impact of the papermaking process. The partners of the research project were: . a machine supplier; . a customer in the paper industry; and . a commercial research centre. During the project the machine supplier got an opportunity to deliver new process equipment tailored to the specific needs of the client. The decision to develop and deliver was, however, not very spontaneous because the product concept was elaborated quite a long time in the concept development phase. At the time, market value and the major competitors were still being evaluated. Some early signals from customers had been received, showing an emerging market demand for the product. The development history of case 2 can be summarized as follows. Phase C lasted two years and it was more or less a process development using simulation tools. At this stage no equipment development was undertaken. During phase C the equipment developer and supplier became familiar with a new kind of production process environment for the company. Phase B lasted two years. At this time, the product concept was planned and tested on the pilot scale. The dimensioning of the equipment was also studied for scale up to industrial scale. The customer-partner worked with the development team to tailor and specify the product according to his needs. Some other customers were also consulted, but their influence on the final product specification was minor. Phase A lasted for one year and during that time the first customer delivery of the case product took place. What was characteristic of the concept development of this product? The development project was very strongly customer-driven: although there seemed to be a general need for this type of product, the concept was developed in close collaboration with a customer. Thus, the product was primarily tailored to fulfil the customer-specific needs. It can also be said that the very good customer collaboration pushed the project forward by giving a sufficient justification for

the work. Other grounds and criteria to continue were not so much needed. A market study was also conducted at the concept development phase and a lot of effort was expended to select the most appropriate materials for the product. Still, the project resulted in a product that was too expensive to manufacture and whose market was inadequately studied. Particularly, the market for the product does not seem today as tempting as when it was assessed and perceived in the concept development phase. Afterwards, R&D management has pointed out that the concept development might have been too rapid, thus preventing a versatile assessment of different construction alternatives. Project management success according to the stakeholders in the case project was reasonably good (see also Figure 3). Generally, the project was thought to be well-managed and organized. According to the project manager, the weakest area was the lack of comprehensive client consultation and marketwide client acceptance, which was misleading for the product concept choice. Also, the project did not have fully satisfactory and specific goals that would have enabled wellfounded project assessment and monitoring. Regarding the technical success of the case, two customer requirements were met in the first delivery both as regards the technical specification and the service provided for the customer within the delivery project. From the businessman’s point of view, the manufacturing costs were too high in the first customer delivery. This was mainly due to the heavy construction with expensive construction materials. Also, there was only minor experience of the type of manufacturing involved. After the first delivery the technical specification of the developed product was benchmarked more carefully than in concept phase B. The conclusion was that some technical improvements had to be made to reach the level of competition in the market segment within the scope of company A. Commercial success The case product was marketed for a one-year period. There were some customer inquiries but no further discussions with the customers because of the selling price, which seemed to be too much above an acceptable market price level. After positioning the new product in respect to the main competitors and studying

221

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

the market segment volume and availability once again, the product launch was stopped. The product concept was not competitive enough and the best option was to withdraw from the market. Catching up with the market price level was not considered to be feasible. The learning process would have been too long and expensive compared to the expected returns. Observations from the interviews (n ¼ 4) with key development personnel and product development management can be summarized as follows: . Strategy: no clear support from strategy in this project. . Customer and competitive position: no clear value added from the business or customer perspective. . Company’s business: Some additional business and synergy contribution seen; however, poor cost efficiency of the product. . Technology: customer’s capability to exploit technology is fairly good. . Reliability of the technology was good. . Overall: one good area in scoring does not make the product and project lucrative from the perspective of business and strategy.

advantage. However, the R&D collaboration between the two companies is “a difficult sport”: different roles and interests of companies, the question of sharing future revenues, product management and marketing principles, and the combination of different organizational cultures set major challenges. In line with this, the two companies, A and B, did not totally agree on the performance (e.g. breeding ratio) of the product. This caused difficulties since it led, for instance, to different sales promotion argumentation. Among the project personnel opinions varied widely regarding the project management success. Although the technical results were satisfactory, there seemed to be tension between the business interests of companies A and B. Both companies admitted that they had learnt a lot from each other and the development group was committed and knowledgeable. Another common conclusion was that the development project was fairly well managed regarding schedule and budget. However, on average, project implementation profile of this project is far from excellent (see Figure 3). The technical specification of the case 3 product was set at the beginning of the project in phase B. A compromise specification was finally accepted by both companies. Company B’s expectations were higher before and during the development project because it was to be the owner of the product concept to be developed and the supplier of the new product. The specification set for the product was not totally met in the development project, but the gap was not expected to be crucial marketing-wise.

Case 3 The case 3 product development was a cooperation project between companies A and B. The outcome of the development project was a merger of two proven technologies in the industry. Company A had the supporting technology know-how and experience, while company B had the main technology regarding the new product. Concerning the development history of case 3, phase C is not applicable. Phases B and A partially overlapped in the development project. During phases A and B the product concept was tested in the laboratory and on a small scale on a pilot machine. This phase lasted about one year and it ended with the first commercial order. With regard to perspectives of future success in this project, it was particularly interesting that the concept development was started in the belief that a potential new killer product could be constructed by combining two technologies. Therefore, company A could not take the risk of totally opting out of the project. In that case, if the product succeeded, it would have provided the competitors with an essential competitive

Commercial success Two years after the end of the development project, company B has sold some 20 units of the new product. Part of them were sold and delivered through main supplier A. Time to market was short and penetration in the rebuild market was successful. The question of whether the product can be exploited fully on the market for new machines is still open but the prospects seem encouraging. Observations from the interviews (n ¼ 5) with key development personnel can be summarized as follows: . Strategy: the link to the strategy is questionable or weak and not discussed properly inside company A, in contrast with the situation in company B.

222

.

.

.

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

Customer and competition: the product is not believed to provide additional value in a proven way in company A and uncertainty also exists about the price competitiveness. Company’s business perspective: there is some synergy but the product does not provide new business for company A. The total economic effect is somewhat unclear. Technology: the customer is ready to implement the technology and it is manufacturing-wise all right. Reliability in wide machines is still unsure after implementation by several customers.

As a result of the development project, no superior product was created compared to the previous product generation from the perspective of company A. Company B has in any case managed to differentiate the new product in competition and it has sold a considerable amount of case 3 products. The solution developed has therefore demonstrated its potential. It seems to have a certain place in the market applications.

Analysis of cases Figure 3 summarizes the project implementation profiles of the three cases studied. The x-axis represents the important parameters or success domains of a project that are evaluated; and the y-axis illustrates the percentage of known success projects (see Slevin and Pinto, 1986) that perform more weakly in that respective domain than projects now studied. According to the PIP method, the projects seem to be clearly different from each other: the profile of case 2 is the most positive, while that of case 3 is the worst. Case 1 lies in the middle. This observation suggests that the PIP method, despite the fact that it has proved to be a valid measure of success in a larger data set, fails to interpret reliably the odds of success concerning an individual project. In an individual project, it seems that a majority of necessary ingredients of success might be present and still the final success remains poor, or vice versa. This would be one reason to advocate holistic and versatile NPD performance measurement, which is discussed, e.g. by Cooper and Kleinschmidt (1995). The summary of all success perspectives studied in three cases is presented in Figure 4.

In each case, the success profile consists of three different success perspectives: R&D project management, technical and commercial success. Case 1 represents a case that is project management-wise rather neutral. Management of the R&D project included some clear weaknesses but also clear strengths. In terms of technical success, case 1 can be considered a positive project; no serious doubts exist, except perhaps concerning price competitiveness. Importantly, however, the commercial success of the product has been – at least so far – too weak. The volume of business has remained small. In case 2 the project management received good grades. There were no severe weaknesses identified regarding the project management. Also the technical success seemed to be more or less all right. Concerns mainly lay with the manufacturing costs. In contrast to this, there was practically no commercial success at all with this product. Due to the reasons discussed in the case section of this paper, the product was finally found to be not competitive enough for the market. Case 3 represented R&D collaboration between two companies. That resulted in a project, whose management proved to be to some extent problematic. Although the technical success of the project was not according to the most optimistic expectations, it was still acceptable. Most importantly, the commercial success of the product has been rather positive, especially from the perspective of company B. What is the picture of success of NPD drawn by these cases? First, the three cases most certainly do not constitute a homogeneous profile. One of the three projects has led to fairly successful business, while the other two have remained somewhat disappointing efforts for the companies. The only commercially successful project (case 3), however, was not a result of a development project that ran smoothly and as planned. Instead, the development work proved to be rather problematic and even the technical result of the project was – to some extent – a compromise. Furthermore, the project’s connection with the strategy was questionable and it was not discussed thoroughly either before or during the active development. In the light of typical success factors discussed in the literature, including product superiority/ quality, market need, product scope, highquality new product process, clear and well-

223

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

Figure 4 Summary of case results

communicated new product strategy for the business unit and positioning in respect of markets and competitors’ products (see Cooper, 1985, 1996; Hollander, 2000), this project is certainly interesting. The product cannot be characterized as superior, but it may be labelled as a fairly good one that has found its place in the market, indicating some success in positioning the product. Also, reasonable commercial success was achieved without high-quality new product process and a strategy connection. In contrast, case 2, which proved to be a commercial failure, was from the perspective of project management an excellent example. Also the technical success seemed to be all right. An important insight that helps to understand these contradictory observations made in case 2 is provided by Ottum and Moore (1997) and Campbell and Cooper (1999). Ottum and Moore (1997) have investigated the role of market information in new product success or failure in their study and have shown that there is a strong relationship between market information processing and new product success. They also stress that effective market information processing requires not only the gathering of good quality information but also the sharing and using of that information (Ottum and Moore, 1997). However, regarding customer involvement in NPD there is also somewhat surprising evidence available. Namely, it is argued that there is no automatic short-term commercial benefit associated with customer partnering compared to in-house development (Campbell and Cooper, 1999). Indeed, one should be very careful in gathering market information and integrating it in the NPD process. Especially in circumstances analogous to case 2, where industrial investment goods were being

developed, on the basis of a very limited number of customer opinions, it is too easy to get the impression that the product and its characteristics are based on market needs even when this is not the case. One or two active customers in the development phase may, as happened in case 2, lead the developing company into areas which are far from real market need. Furthermore, case 1, which clearly succeeded in terms of technological goals and did not fail from the perspective of project management, did not make a clear commercial difference. In this light, it seems that neither project management nor technical success alone are measures that should be used to anticipate commercial – bottom-line – success in this particular environment. Partly this is due to a situation analogous to that in case 2 regarding market or customer information. Instead of being the result of a market study, the product specification was created mostly on the basis of a strategic definition of company policy. Still, the product seems to be able to deliver value for the customer. It is, however, noteworthy that the company has failed in the task of value/cost assessment regarding this product. To be able to compete in the market successfully, product cost and value have to be in balance. So far, this has not been the case. As regards the problem presented by Matthews (1991) according to which category B projects may often fail to receive justification for funding since they neither represent pure “breeding ground” any more nor have reached the status of “investments” that could be assessed using financial measures, the cases in this paper provide one – though a limited – insight. Especially when examining the financial success of the project, it is necessary to assess and to measure the

224

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

project from the perspectives of all relevant stakeholders. Technological success or strategic fit is not enough if there is, for instance, lack of understanding concerning customers’ value creation or competitors’ positions and offerings. Finally, to turn an otherwise successful project or product into good business, one has to consider also issues such as cost efficiency or manufacturability. As Neely et al. (1995) among others, point out, the main sources of inappropriateness of performance measurement, in general, are short-termism, lack of strategic focus and local optimisation. A holistic view seems to be required. In the light of the cases presented in this paper, NPD is no exception to the other application areas of performance measurement: good performance measurement captures a number of dimensions that are associated with success in a particular case. Regarding the perspectives introduced earlier which make a contribution to new product success, the observations of the study can be summarized as follows: . Strategy: without some kind of strategy connection, an R&D project is not likely to receive the resources and support that are an imperative in assuring the proper conduction of a development project. However, on the basis of all this case evidence, a strategic fit or strategic support does not seem to be able to secure any success if the other fundamental issues such as customer acceptance, right timing or feasible costs are neglected. The cases reported in this paper serve as good examples of this. . Company business: the implications of the project for the company’s business and competitive situation should be evaluated well. Even if the primary target is not the cost leadership, cost efficiency has to be at an acceptable level. A technically good product with the wrong price tag on it is not – at the end of the day – a lucrative product. . Technology: the technologies used in the products under development should be well proven both separately and together in the end-user environment. In the development process itself it could be wise strictly to separate different phases from each other. That is, the development of the product should be carried out separately from the development of the technology. It might also be very valuable

.

to explore fully the potential of existing product technology before introducing new technology. Customer and competition: also, it is very important for a better understanding of customer expectations to consult several customers in the selected market segment during the product development phase. Apparently, customer-driven R&D may well be to some extent misleading if it is based only on the identified needs of a single customer.

This paper is not in a position to give definite recommendations regarding the means for implementing performance measurement or evaluation. At least there is no lack of different available methods (see Martino, 1995). Whether the measurement is done by “construction of numerical indices, struggling with figures and scoring the projects” (McLeod, 1988) or by more subjective assessments, the main challenge is not the actual measurement but the process of making the right decision on the basis of the information available.

Discussion The objective of the paper was to analyse the dimensions of success at R&D project level. Success was seen as a variable that is connected with the NPD process and its phases. An assumption behind the paper was that there are different patterns of success for R&D projects, which are also very different from each other. A core dilemma was related to the correlation between different success domains: a good performance in one success domain may or may not be related to success in another one. In line with the literature (see, e.g. Hart, 1993; Griffin and Page, 1996), the cases investigated in the paper showed that success in NPD is indeed a multifaceted and elusive issue. Project management success, technical success and financial success seldom go hand in hand – as unwelcome as this may be from the management point of view. In the process of specifying and interpreting the meaning of NPD success, this fact reinforces the perception that success is both a context- and stakeholder-specific issue. The very same project can – even with sound argumentation – be interpreted either as a failure or as a success story, depending on who is evaluating

225

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

the project, the situation or the timeframe within which evaluation takes place. Also, different product development projects often represent quite different patterns of success: one project may perform very well in financial terms, another may be profitability-wise a failure but in the technical sense – by meeting the state-of-the-art technical requirements – a big success story, while a third project’s outcome may be poor almost in every respect but the project management has been efficient and well organized, thereby providing the company with a good example of outstanding R&D project management also for the future. The existence of different patterns of success challenges R&D management. Striving for success might include seeking compromises between different forms of success. If a project cannot be successful in every domain, it still may be able to contribute in some areas. Taking into account that investments in R&D are expected to produce returns, especially in the long term, an individual project can be considered a success to some extent if it can either produce a direct impact or indirectly produce at least the preconditions for future returns. Therefore, a systematic demand for direct payoffs in each NPD project that refuses to take into account the importance of projects’ indirect impacts on success may even be harmful in terms of the overall success of NPD. Although the three success perspectives studied suffice to demonstrate the elusive and varied nature of NPD success, this analysis has excluded other issues that are also of interest. For example, as seen in the context of knowledge creation, the concept of product development success receives one additional interpretation: R&D activities can be perceived as successful when they facilitate or generate valuable knowledge. This seems to be reasonably well in line with Lewis’ (2001) notion that NPD can be seen as a specific illustration of organizational learning. Lewis (2001) argues that the interaction between resources and processes is not unidimensional: resources create value when they are utilised in processes, which in turn helps to create new resources or extend existing ones. In NPD this means not only that skilled resources are one of the prerequisites of a good NPD process, but a good process by itself also generates valuable information, knowledge or experience – i.e. resources. Drejer and Riis (1999) view competencies as a

system of human beings using technology in an organized way and under the influence of a culture to create output that yields a competitive advantage for the firm. Both of these views are essentially similar: they perceive competencies to some extent as systems – action and interaction are strongly present, and both link competencies closely to the creation of competitive advantage. Competencies are not just any ability to respond to “random” demands, but rather building blocks for a firm’s competitive advantage. Since the concept of success is itself difficult to define, it is a difficult task indeed to try to describe the road leading to success. Therefore, on the basis of this study, it would be questionable to present any standard formula for new product success. However, consistently with the findings of this paper, many variables that have an influence on success have been identified also in prior studies: for instance, product superiority, market need and competitive position (see, e.g. Cooper, 1985; Cooper and Kleinschmidt, 1995; Hollander, 2000; Hultink et al., 2000). Also, concerning the generalization of the results, it is not claimed that all development projects behave similarly to the projects analysed in this paper. That is, a project could possibly succeed in all domains or it could fail in all three. However, it seems to be a common of product development projects that success often has something to do with coincidences or chaos. Perhaps everything cannot be rationalized.

References Batty, J. (1988), Accounting for Research and Development, 2nd ed., Gower Publishing, Aldershot. Brown, M.G. and Svenson, R.A. (1998), “Measuring R&D productivity”, Research Technology Management, Vol. 41 No. 6, pp. 30-6. Campbell, A.J. and Cooper, R.G. (1999), “Do customer partnerships improve new product success rates?”, Industrial Marketing Management, Vol. 28, pp. 507-19. Chiesa, V. and Masella, C. (1996), “Searching for an effective measure of R&D performance”, Management Decision, Vol. 34 No. 7, pp. 49-57. Cooper, R.G. (1985), “Selecting winning new product projects: using the NewProd System”, Journal of Product Innovation Management, No. 2, pp. 34-44. Cooper, R.G. (1996), “Overhauling the new product process”, Industrial Marketing Management, Vol. 25, pp. 465-82. Cooper, R.G. and Kleinschmidt, E.J. (1995), “Benchmarking the firm’s critical success factors in new product

226

The patterns of success in product development: a case study

European Journal of Innovation Management

Petri Suomala and Ilkka Jokioinen

Volume 6 · Number 4 · 2003 · 213-227

development”, Journal of Product Innovation Management, No. 12, pp. 374-91. Crotogino, R.H. (2000), “The mysteries of technology transfer – luck versus good planning”, paper presented at the International Drying Symposium IDS2000, Noordwijkerhout. Drejer, A. and Riis, J.O. (1999), “Competence development and technology. How learning and technology can be meaningfully integrated”, Technovation, No. 19, pp. 631-44. Driva, H., Pawar, K.S. and Menon, U. (2000), “Measuring product development performance in manufacturing organisations”, International Journal of Production Economics, Vol. 63, pp. 147-59. European Industrial Research Management Institute (EIRMA) (1985), Evaluation of R&D Output, Working Group Reports. Report No. 29, EIRMA, Paris. European Industrial Research Management Institute (EIRMA) (1995), Evaluation of R&D Projects, Working Group Reports. Report No. 47, EIRMA, Paris. Ellis, L. (1997), Evaluation of R&D Processes: Effectiveness through Measurements, Artech House, Boston, MA. Griffin, A. (1997), “PDMA research on new product development practices: updating trends and benchmarking best practices”, Journal of Product Innovation Management, No. 14, pp. 429-58. Griffin, A. and Page, A.L. (1996), “PDMA success measurement project: recommended measures for product development success and failure”, Journal of Product Innovation Management, No. 13, pp. 478-96. Hart, S. (1993), “Dimensions of success in new product development: an exploratory investigation”, Journal of Marketing Management, No. 9, pp. 23-41. Hirons, E., Simon, A. and Simon, C. (1998), “External customer satisfaction as a performance measure of the management of a research and development department”, International Journal of Quality & Reliability Management, Vol. 15 No. 8/9, pp. 969-87. Hollander, J. (2000), “Genesis, a product assessment instrument used during the product development process”, paper presented at the 7th International Product Development Management Conference, EIASM, Leuven. Hultink, E.J. and Robben, H.S.J. (1995), “Measuring new product success: the difference that time perspective makes”, Journal of Product Innovation Management, No. 12. Hultink, E.J., Atuahene-Gima, K. and Lebbink, I. (2000), “Determinants of new product selling performance: an empirical examination in The Netherlands”, European Journal of Innovation Management, Vol. 3 No. 1, pp. 27-34. Industrial Research Institute (IRI) (2000), R&D Facts 2000, Industrial Research Institute, Arlington, VA. Jackson, T.W. and Spurlock, J.M. (1966), Research and Development Management, 2nd ed., Dow JonesIrwin, Homewood, IL. Kerssens-van Drongelen, I.C. and Bilderbeek, J. (1999), “R&D performance measurement: more than choosing a set of metrics”, R&D Management, Vol. 29 No. 1, pp. 35-46. Kulvik, H. (1977), Uusien tuotteiden onnistumiseen tai epa¨onnistumiseen vaikuttavat tekija¨t, Helsinki University of Technology, Helsinki (in Finnish), 93pp.. Lewis, M.A. (2001), “Success, failure and organisational competence: a case study of the new product

development process”, Journal of Engineering Technology Management, No. 18, pp. 185-206. Lindman, M. (1977), “Managing industrial new products in the long run”, Acta Wasaensia, University of Vaasa, Vaasa, 336 pp.. Ma¨kinen, S. (1999), “A strategic framework for business impact analysis and its usage in new product development”, Acta polytechnica Scandinavica, Tampere University of Technology, Espoo, 213 pp.. Martino, J.P. (1995), R&D Project Selection, John Wiley & Sons, New York, NY. Matthews, W.H. (1991), “Kissing technological frogs: managing technology as a strategic resource”, European Management Journal, Vol. 9 No. 2, pp. 145-8. McGrath, M.M. and Romeri, M.N. (1994), “The R&D effectiveness index: a metric for product development performance”, Journal of Product Innovation Management, Vol. 11 No. 3, pp. 213-20. McLeod, T. (1988), The Management of Research, Development and Design in Industry, 2nd ed., Gower Technical Press, Aldershot. Neely, A., Gregory, M. and Platts, K. (1995), “Performance measurement system design: a literature review and research agenda”, International Journal of Operations & Production Management, Vol. 15 No. 4, pp. 80-116. Nixon, B. (1998), “Research and development performance measurement: a case study”, Management Accounting Research, Vol. 9, pp. 329-55. Ormala, E. (1986), Analysing and Supporting R&D Project Evaluation: An Applied Systems Analytic Approach, Helsinki University of Technology, Espoo. Ottum, B.D. and Moore, W.L. (1997), “The role of market information in new product success/failure”, Journal of Product Innovation Management, No. 14, pp. 258-73. Poolton, J. and Barclay, I. (1998), “New product development from past research to future applications”, Industrial Marketing Management, No. 27, pp. 197-212. Rouhiainen, P. (1997), Managing New Product Development Project Implementation in Metal Industry, Tampere University of Technology, Tampere. Schumann, P.A., Ransley, D.L. and Prestwood, D.C. (1995), “Measuring R&D performance”, Research Technology Management, Vol. 38 No. 3, pp. 45-55. Slevin, D.P. and Pinto, J.K. (1986), “The project implementation profile: new tool for project managers”, Project Management Journal, September, pp. 57-71. Szakonyi, R. (1994a), “Measuring R&D effectiveness – I”, Research Technology Management, March-April, pp. 27-32. Szakonyi, R. (1994b), “Measuring R&D effectiveness – II”, Research Technology Management, May-June, pp. 44-55. Ulrich, K.T. and Eppinger, S.D. (1995), Product Design and Development, McGraw-Hill, New York, NY. Werner, B.M. and Souder, W.E. (1997a), “Measuring R&D performance – state of the art”, Research Technology Management, Vol. 40 No. 2, pp. 34-42. Werner, B.M. and Souder, W.E. (1997b), “Measuring R&D performance – US and German practices”, Research Technology Management, Vol. 40 No. 3, pp. 28-32. Yin, R.K. (1994), Case Study Research: Design and Methods, 2nd ed., Sage Publications, Newbury Park, CA.

227

The necessity of mobile service innovation

Benefits of involving users in service innovation Peter R. Magnusson

The author Peter R. Magnusson is Executive PhD Candidate, The FENIX Research Program, Stockholm School of Economics, Stockholm, Sweden. Keywords User involvement, Services, Innovation, Telecommunication Abstract Past research has demonstrated that industrial customers can, in effect, bring about product innovation among their suppliers. However, little seems to be known as to whether consumers are also potential inventors of new services. Presents results from an empirical study with the objective of exploring whether ordinary users can contribute novel service ideas regarding mobile telephony. An experimental approach was used to compare the characteristics of new services suggested by ordinary users with services suggested by professional developers. It was found that the service innovations suggested by the users were more creative and useful than those suggested by the professionals. On the other hand, the suggestions of the professionals were deemed easier to produce. Concludes with a discussion on the contributions and limitations of user involvement, wherein the organisational role of the users involved is discussed. Also makes a proposal regarding how to further investigate the potential of the user as a co-worker in the innovation process. Electronic access The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1460-1060.htm European Journal of Innovation Management Volume 6 · Number 4 · 2003 · pp. 228-238 q MCB UP Limited · ISSN 1460-1060 DOI 10.1108/14601060310500940

The first pioneering endeavours in mobile telephony started in the 1920s. It was not, however, until the mid-1980s that mobile telephony became commercially available to a broader market. The growth of the market has been enormous during the last decade, but it is now starting to decline, resulting in a downward pressure on mobile telephony rates. Mobile operators are thus seeking ways of becoming differentiated, for instance by offering new non-voice services, i.e. mobile data communications services. An enabler of this is the convergence between different networks (infrastructures) – such as telephony, television, GPS (positioning system), the Internet (Kupfer, 1993). A variety of information can be combined and accessed through the mobile phone, which has become an access device for new kinds of services. An illustrating example of this is the successful Japanese system called I-mode (Curtis, 2001). For instance, in order to avoid traffic jams, car drivers can request and obtain information directly to their mobile phones regarding the best route. The development of these new services introduces new complexities, compared with traditional voice telephony. Previously, the operator did not have to worry about either the content or context of a call, revenues only being related to the duration of the call. These novel services will demand a thorough understanding of users’ actions. Telephony operators must understand how value is created for the user; thus this is a totally new situation. The dilemma is that the operators’ traditional service development process is not designed to capture these new wants and needs of users. This raises the question; can users themselves suggest novel ideas that can be developed into new useful mobile communications services? The importance of involving customers or users[1] in product innovation is stressed in the literature. Collaboration between suppliers and users can lead to a mutual understanding of the users’ needs and wishes, as well as an understanding of the technological opportunities (Anderson and Crocca, 1993; Sinkula, 1994; Veryzer, 1998; Hennestad, 1999). It has also been recognised that development times can be shortened if continuous acceptance testing is used during

228

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

development (Gupta and Wilemon, 1990; Iansiti and MacCormack, 1997). Users can come up with suggestions for improvements during development (Norling, 1993; Prahalad and Ramaswamy, 2000). However, users are not normally trusted to play a part in the initial generation of new product ideas. This endeavour is reserved for the company’s professional designers. Users are contacted after the company has developed a new concept for a product or service in order to evaluate these, e.g. focus groups (McQuarrie and McIntyre, 1986). Arguments are also raised against user involvement; that it might not provide any positive effects that justify the extra cost (Gales and Mansour-Cole, 1995; Campbell and Cooper, 1999). Others argue that users do not have sufficient technical knowledge to produce innovations (Christensen and Bower, 1996), or that they cannot articulate their needs (Leonard and Rayport, 1997). Indications to the opposite do, however, exist. Wikstro¨m (1995) is of the opinion that intensive interaction with potential customers is a likely source of generating new ideas and new ways of doing business. Bitner et al. (2000) recommend the close involvement of customers in the design process of technology-based services. Von Hippel has illustrated that certain types of users, so-called lead users, have invented the majority of products in certain industries (von Hippel, 1977, 1978, 1982, 1986, 1988). Lead users are expected to have unique features, which von Hippel (1988, p. 107) defines thus:

consumers and business markets. This indicates that consumers, or ordinary users, might also be utilised as innovators of services. One criticism of von Hippel’s lead user concept has been the difficulty of identifying lead users beforehand. A point of conjecture is that it might be even harder to find lead users on consumer markets. According to Rogers’ theory of innovation diffusion, von Hippel’s lead users would only represent approximately 2.5 per cent of the total number of customers (Rogers, 1995; von Hippel et al., 1999). These figures imply that lead users might not be a good representation of the remaining 97.5 per cent of the market. It is not certain that lead users have the same frame of reference as “ordinary” users, which might lead to the design of superfluous products and services. Accordingly, involving various categories of users seems like an attractive approach to co-opting different users’ attitudes, wishes and needs. However, can we expect that involving users in suggesting new product or service ideas will provide unique contributions, or is this attempt doomed to failure?

Lead users face needs that will be general in a marketplace, but they face them months or years before the bulk of that marketplace encounters them, and Lead users are positioned to benefit significantly by obtaining a solution to those needs.

Von Hippel’s studies of lead users are from industrial markets and do not include consumers. An as yet unanswered question is whether consumers could be used in the same manner as industrial lead users. In a study within the telecommunications industry, Mullins and Sutherland (1998) investigated best practices for new product and service development. One of the “best practices” included user involvement for idea generation, and the use of mock-ups and prototypes to understand customer usage and benefits. The products had a rather high degree of service content and targeted both

Research approach This paper, will investigate the contribution made by user involvement during the very early phases of inventing new end-user services for mobile telephony. The paper builds on a comparative experimental study where users’ proposals for new ideas were compared with those of professional service developers[2]. The purpose of the paper is to examine whether it is meaningful to involve ordinary users in creating suggestions for new end-user services. This can be formulated in the following research question; “Can users contribute something extra to the service innovation process, compared to professional service developers?” A cornerstone of the research design has been the emulation of a real situation of user involvement. Therefore, the research design was attempted in order to constitute a realistic way of organising user involvement in the innovation process. Two experimental trials were held, one with a group of professional service developers and the other with a group of users. The participants were instructed to create one or more service ideas on the basis of a given assignment. Since the resulting

229

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

ideas were solutions to a common assignment, they could be compared and ranked against each other, thus enabling the determination of the users’ contributions when involving them in the idea creation process. The actual experiment consisted of four stages: initiation, idea creation, delivery and evaluation. The outline is illustrated in Figure 1. The experimental outline will be described in more detail further on in the article.

the trial, each participant underwent three different tests to determine more specific personality characteristics: (1) FS-test, correlates with a person’s creativity (Holmquist and Ekvall, 1986). (2) Life orientation test (LOT), indicates whether a person has a positive or negative disposition (Scheier and Carver, 1985). (3) Technology readiness (TR) indicates a person’s willingness to adopt new technology (Parasuraman, 2000).

Context and sample The first group consisted of 12 professional service developers; these were randomly recruited from Telia Mobile, a Swedish mobile telephony operator. All of them came from an R&D unit responsible for developing new non-voice mobile services, i.e. services based on SMS, WAP, GPRS and suchlike. Their professional experience in the field varied between one and ten years. The participants in the second group consisted of 19 “users”, represented by university students in non-technical study programmes, e.g. social science, teacher training, business administration, etc. The students volunteered to participate. The main reason for choosing students was that they represent one of the most frequent SMS user-groups, thus being well representative of users. To obtain individual background data, the participants filled in a form asking for personal data such as age, experience of mobile telephony and SMS usage. On entry to

The groups’ scores, including some other descriptive data, are shown in Table I. In the following sections, the experimental procedure will be described in more detail. The two groups (professionals and users) followed essentially the same procedure, with some minor deviations, which are indicated in the text. Initiation At the initiation meeting, each trial group was gathered and the study introduced. A common, yet individual, assignment was given to each participant. The assignment for the user group was to create service ideas that would provide themselves value. The experts, on the other hand, were instructed to come up with ideas for services that they thought would add value to the “user group”. Through these different formulations, both groups would actually be trying to satisfy the same target group. This way, the service ideas were

Figure 1 Outline of the experiment

230

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

Table I Personal characteristics of the participants

FS-test (creativity) LOT (positive life orientation) TR (technology readiness) Age Sex Mobile telephony experience (years)

Mean SD Mean SD Mean SD Mean SD Females Males Mean SD

comparable. The services were supposed to utilise an application platform, Unified Services (US). US is essentially a converter between SMS[3] messages in GSM and http calls on the Internet, as illustrated in Figure 2. From the user’s point of view, US enables access to information on the Internet by sending and receiving SMS messages. One example of an end-user service was the ability to access a bus timetable on the Internet via SMS calls. No special technical competence was required just to use the services. The instructions and assignment for the experiment were given in both verbal and written form. A diary was handed out to the participants in which they were instructed to document the ideas that came up, as well as the activities that triggered the idea creation. Below is an excerpt of the assignment given to the users (translated from the Swedish): Your assignment is to create – using the knowledge you have attained of Unified Services, along with a mobile phone – services that produce an added value for you. Put simply, you

Professionals (n 5 12)

Ordinary users (n 5 19)

5.92 1.56 23.08 3.70 8.00 4.11 36.50 8.13 2 (17%) 10 (83%) 10.42 6.04

6.37 1.77 24.53 4.34 5.79 5.92 23.79 2.18 4 (21%) 15 (79%) 3.60 2.33

are going to create one or more services that you perceive as important. It is important that you are able to describe how your solution creates an added value for you. The solution can, for instance, facilitate information retrieval, or create a beneficial experience of some kind. The one who decides the value adding is you yourself, accordingly, there is no right or wrong here.

The written assignment and instructions were put in a blank diary. The participants were instructed to document the idea creation process; an excerpt is shown below (translated from the Swedish): It is important that you make notes in the diary of how your ideas emerged. That is, where did you get the idea? Was it from someone you have worked or been in discussion with? Did the idea arise from an association path when you saw or heard something? The purpose of the narratives is to enable us to understand the originating process of your ideas.

To provide the participants with a sense of how these services might work and to inspire them, they were given access to a service portfolio consisting of about ten implemented

Figure 2 Unified Services, platform for developing mobile Internet services based on SMS

231

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

services. To use and test these, the participants were equipped with a mobile telephone (Ericsson 1018S) and a prepaid card (Telia GSM Refill) topped up to a value of approximately 25 euros[4] and a so-called chat-board. The chat-board is a little keyboard that can be plugged into the mobile phone to simplify the input of text messages. All participants received hands-on training on how to use the phone and chat-board by means of testing some of the services. It was also stated that an idea would remain the intellectual property of the creator of the idea, otherwise many of the ideas would probably never have been submitted. In order to make the experimental setting more stimulating, the participants were informed that an award of 80 euros would be given for the best service idea. It was also stated that one could choose to work alone or collaborate with others. If one collaborated, this was to be noted in the diary. Each individual was expected to produce at least one service idea of his or her own. The professionals underwent the same initiation procedure, with the exception that their ideas would become the intellectual property of the company and that they would not receive any best service reward. This conformed to standard policy for job-related ideas created by the employees of the company. It should, however, be noted that the company had a regulated payment system for ideas falling under the protection of patent laws. If an idea was patented by the company the employee would be rewarded with a sum of around 2,400 euros and upward. There was, thus, an incentive for the professionals to come up with valuable ideas as well.

notebook when an idea came up, i.e. not much of the process was captured. These notes were, however, often richer than the resulting service description. Others used the diary to write small narratives around their idea creation. The following example illustrates how one of the participants came up with the idea that an alarm function would be convenient. She wrote:

Idea creation The idea creation period of the study lasted for 12 days. During this period, communication between the trial management and the participants was kept to a minimum. However, a notification was sent, in the form of an SMS, to all participants on day two to remind them to be active in the trial. Technical support was available via email or telephone. Only questions regarding technical problems as well as practical usage of the telephones and services were attended to. Only four support questions were received from the entire group of 19 users. The diaries were used in quite diverse ways. Some of the participants used them more as a

I’m on my way home from the city, walking into the subway that leads to my home. It’s dark . . . a personal alarm would be a function appreciated by all, but by girls in particular.

Delivery After the 12-day idea creation period, a meeting was held at which the service ideas and diaries were delivered. The participants’ service descriptions were submitted; these were written descriptions in a predefined format, and could thus be seen as a first conceptualisation of the service. The format included; the date of idea creation, the developer’s name, the name of the service idea, a short functional description, and the intended user-value of the service. The two groups produced a total of 178 ideas for new services. A total of 55 came from the 12 professional developers, and 123 from the 19 users. All participants were also interviewed within the next two weeks. The interview was semi-structured in nature. Its purpose was to trace the relevant process data, such as important incidents triggering especially good ideas, and was thus a complement to the diaries. Selection of judges The type of judge evaluating a certain result can affect the evaluation. For example, a company expert might evaluate the perceived user-value differently to an ordinary user. Because of this, three different panels of judges were used to evaluate the ideas from different perspectives. The first panel was a group of six experts experienced in evaluating mobile communications service ideas. They could thus be considered good evaluators of the feasibility of implementation, as well as having a broad frame of reference in order to appraise the novelty of the ideas. To assess the ideas from the end-users’ points of view, “users” constituted the two remaining panels. Panel number two represented technicallyskilled users and was composed of three students of computer engineering. Since the

232

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

services in the study were technically based, it might be necessary to have some technical expertise. For this type of services, this panel could also be assumed to be representative of early adopters (Rogers, 1995). The third panel represented ordinary users who are familiar with mobile telephony. This panel had the same background as the users participating in the experiment.

would make assessment rather complex and tedious. To avoid this, only the factors intrinsic to the product were evaluated. The dependent variable, “quality of service idea”, could thus be divided into three dimensions. For other applications, different dimensions are, however, likely to construct the dependent variable. The ideas were assessed using the Conceptual Assessment Technique (CAT) proposed by Amabile (1996). One dimension was evaluated at a time. In other words, if a judge were to evaluate all three dimensions, the procedure would be repeated three times. The judges made the assessment independently of each other, not knowing who the inventor of the idea was. Each idea was classified on a scale of 1 to 10. For the three dimensions (originality, user-value, producibility), a score of one represented the least original, least valuable, and hardest to produce. Similarly, a score of ten corresponded to the most original, most valuable, and easiest to produce. To ensure that all the judges in the panel had understood how the dimension was to be evaluated and to “calibrate” the judges’ assessments, a test round was completed. Five service descriptions were picked and assessed individually by the judges; the panel then discussed the results. In particular, if some of the five ideas were assessed very differently, this was discussed. Finally, the service ideas were evaluated for the dimension in question. The assessment was made individually, i.e. no discussions were allowed among the judges. At the time of evaluation, ideas from groups other than the two discussed in this paper were assessed simultaneously. Each judge therefore assessed a total of approximately 290 ideas. The expert panel evaluated three dimensions, i.e. 870 assessments. The two user panels assessed two dimensions, i.e. 580 assessments. For practical reasons, the assessments had to be performed in different ways. For the experts, a two-day intensive workshop was held at a boarding house, where all the assessments were concluded. For the two user groups, an introduction meeting was held lasting half a day. Both dimensions to be assessed were presented and calibrated, as previously described. However, the judges had to conclude their assessments at home. They received a stamped envelope to post their assessments as soon as possible, at the very latest within a week.

Evaluation What are the interesting quality aspects of these types of services? To answer this question, a focus group session was arranged. Participants included five experts in the assessment of mobile telephony services. These were asked to define the relevant factors for evaluating an end-user service, this resulted in two broad categories of factors: intrinsic (embedded in the service); and market (external environment). The intrinsic factors agreed on could be summed up in three dimensions: originality, user value, and producibility. Much value in today’s companies is determined by the future potential of their offerings. The ability to produce innovative products and services is often a necessity, at least in the telecommunications business. Originality is a concept that enfolds the innovative dimension. One reason for involving users in the development process is to co-opt their preferences, desires and needs. User value is meant to take the user’s perspective of the solution; what is the value of using the service, is it likely that the target group will use the service? The third dimension producibility, i.e. the ability and ease by which the service can be produced; this dimension takes the producers’ perspective. A concept can be excellent from a user’s perspective, and also extraordinarily innovative, but if it cannot be produced it holds no business value for the company. The second category, market factors, is more problematic. The problem with assessing ideas in respect of market criteria is that idea A can be ranked higher than B in one market context, while the opposite ranking could apply in another context. For instance, in a new emerging market, originality is probably very important for business success, while in a mature market, producibility (simplicity to implement) is more important. Assessing the market factors thus implies that the business context has to be specified, which

233

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

Results

Table III T-test for no difference between professionals and users

In Table II, a summary of the results of the service idea assessment is presented. In order to evaluate producibility, a thorough understanding of mobile telephony systems is needed, accordingly only the expert panel evaluated this dimension. The table is divided into three sections, one per type of panel. The first section (first two rows) shows the results of the assessment by the professional expert judges. The next is the technically skilled users and the last the ordinary user judges. As can be seen, all panels assessed differences in the same direction between the two groups, for all three dimensions. The users’ ideas were more original, but at the same time less producible than the developers’ ideas. Also, user-value was assessed more highly for the ideas proposed by the users. To check for significant differences between the two groups (professionals and users), an independent sample t-test was carried out for the three dimensions. The test was a so-called null hypothesis test where no differences were assumed; the results are shown in Table III. Four out of seven t values were significant at the 5 per cent level. Worth noting is the fact that neither of the two t-values from the ordinary user panel were significant. The third non-significant t-value was just above the 5 per cent level ( p ¼ 0.057). A cross-correlation of the three dimensions gave a significant result between two of them, namely originality and producibility, as seen in the scatter diagrams in Figure 3. The other two cross-correlations were not significant. A strong negative correlation between originality and producibility can thus be established. The more original an idea was assessed to be, the harder it was assumed to be to implement. The strongest negative correlation was found between the scores for the users (2 0.702). That result indicates that

Technically skilled user judges t Sig.

Professional expert judges t Sig.

Dimension Originality Producibility User-value

2 2.14* 2.41* 2 1.91

0.034 0.017 0.057

22.18* – 21.98*

0.031 – 0.049

Ordinary user judges t Sig. 20.53 – 20.68

0.598 – 0.496

Note: * p , 0.05

the judges tended to group user suggestions into only two categories: “original but not producible” or “producible but not original”. For the professionals, the judges classified a slightly higher proportion of the suggestions as both original and producible. An interesting observation was that, of the users’ 123 service ideas, 20 (16 per cent) of them were not considered to be “actual services” by the professional developers. The major part of those “non-service ideas” were suggestions to equip mobile phones with extra tools such as a bottle opener, alcoholmeter, radio, etc. Another group featured solutions where the mobile phone acted as an electronic wallet. This indicates that the user’s comprehension of what a service is differs from that of the expert.

Discussion The reported study has provided some important empirical data for discussion. The first topic concerns the assessment of ideas. We have seen that different types of judges arrive at different assessments, raising the question; who is suitable to assess the type of service ideas in this study? The other topic discussed concerns the actual contribution of user involvement to the service innovation process, i.e. “Can users contribute something extra to the service innovation process, compared to professional service developers?”

Table II Service idea assessment by three-judge panels Types of judges Professional experts (n ¼ 7) Technically skilled users (n ¼ 3) Ordinary users (n ¼ 4)

Group

n

Professional developers (n ¼ 12) Users (n ¼ 19) Professional developers (n ¼ 12) Users (n ¼ 19) Professional developers (n ¼ 12) Users (n ¼ 19)

55 123 55 123 55 123

234

Originality Mean SD 2.99 3.55 4.81 5.50 4.55 4.68

1.07 1.79 1.78 2.01 1.49 1.67

Producibility Mean SD 6.84 6.01 – – – –

1.48 2.36 – – – –

User value Mean SD 4.05 4.54 4.74 5.35 5.01 5.18

1.21 1.71 1.98 1.83 1.40 1.61

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

Figure 3 Cross-correlation (Pearson, two-tailed) between originality and producibility

The results have indicated that users produce more valuable and original ideas than professionals do; is this valid? There seem, however, to be severe drawbacks in the users’ ideas; they are unfeasible (not producible). Does this mean that their contribution to the innovation process is limited? Finally, in the study, there was an unanticipated finding; the users’ conception of what a service is seems to differ from the established view among professionals. What are the possible implications of this finding?

Who should evaluate ideas? In the research design presented, the appraisal of user contribution is dependent on the assessment of the proposed ideas. As described, three different types of judges were used. The study indicated that certain technical skills might be needed in order to become a capable judge. The panel of ordinary users did not establish statistically significant differences between the groups in either of the two dimensions evaluated. It could be assumed that this type of service (technically based) is difficult to assess when one does not have a technical background. Without technical skills, it might be hard to see and understand the actual differences, and for that reason there is a tendency to assess the ideas as being fairly equal in the evaluated dimensions. It could also be that technicallyeducated people are more experienced in evaluating objects than others. Experienced evaluators might be more distinct and polarised in their judgements. This experience conforms to those made by Amabile (1996, p. 72). She has noticed that not everyone, not even experts, can be considered an

appropriate judge in every domain. The tentative conclusion is that the type of service ideas examined in this study need technicallyskilled judges.

What are the benefits of users? The study has shown that users can produce qualitatively different, and from some viewpoints better, service ideas than professional service developers do. The result that users create service ideas perceived as providing greater user value is, perhaps, not so surprising. Users themselves ought to be the best ones to know what creates value for them. More remarkable, however, is the fact that the users produced more original ideas than the professional developers did. This is quite contrary to what has been stated by some previous research (e.g. Souder, 1989; Christensen and Bower, 1996). We can conclude that involving users as innovators can result in more innovative services of greater user-value. It is, however, appropriate to reflect on the validity of the results; can we determine that the differences between the groups are caused by the different types of innovators (experts or users)? Could there be any factors affecting the study that could falsify the causation? A number of alternative explanations will be discussed below. The relatively small groups might introduce an interaction effect, linked to personal characteristics. For instance, the ability to create new ideas (being creative) is to a certain extent due to hereditary talents, i.e. personality (Fesist, 1999). The WilcoxonMann-Whitney test did not show any significant differences between the two groups in the three personality variables measured;

235

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

FS (Z ¼ 2 0.744), LOT (Z ¼ 2 1.021), TR (Z ¼ 2 1.243). Neither was there any significant difference between the groups in the distribution by sex. The average age was, however, significantly higher for the professionals (36.5 years) than for the users (23.8 years). Nevertheless, a Pearson correlation (two-tailed), between age and originality of ideas, gave no significant value (r ¼ 2 0.143, p ¼ 0.451). The award (80 euros), set up for the best user idea, might introduce an error; the better results achieved by the users could be dependent on the award. What speaks against this is the fact that the individual professionals could also earn a lot of money, i.e. more than 2,400 euros if their ideas were patented. In other words, the professionals should be considered at least as motivated as the users with regard to performing well. Of course, it might be the case that the monetary reward did not influence the participants at all. If so, the fact that the students did volunteer for the assignment, while the professionals were “ordered” to participate, might explain some of the differences. The conclusion made is that the validity of the results can be considered sufficiently good. The discussion has pointed to a number of possible factors, over and above the difference between being a professional designer and an ordinary consumer of mobile telephony services, that could have contributed to the results. However, most of these factors can be interpreted as further explanations as to why users are more creative and have more insights into consumer needs than professionals do. With the possible exception of motivation, these other factors do not contradict our conclusion, i.e. that it would be beneficial to let users innovate services. However, for further studies, it would be interesting to investigate whether improved incentive schemes for professional developers would make them come up with more creative and useful service innovations.

producibility is probably mainly due to a lack of knowledge of what is “possible to do” with the technology at hand. Users come up with ideas that are original because they are not implementable. The low producibility scored by users might also be explained in terms of an inability to depict their ideas. Ordinary users do not possess the necessary vocabulary to explain their ideas unambiguously. On the other hand, the score in the producibility assessment can be misleading, as the following anecdote illustrates. A group of professional developers were presented with some of the users’ ideas from the experiment. One of them was one of the most original ideas (score ¼ 10), but at the same time the least producible (score ¼ 1). The creator described this idea as follows: “I had received the wrong Sunday paper for the third consecutive week. It suddenly occurred to me that the phone ought to have a function that would enable me to induce an electric shock in the newspaper-boy, so that he would learn to give me the right newspaper”. The professionals’ first reaction was that the service proposal was a ”joke”, but after reflecting on it for a short while something happened. The group collectively started to redesign the proposal. From being something totally unfeasible, they had, within the space of a few minutes, come up with a new, realisable design that produced approximately the same user-value as the initial proposal. The idea thus triggered a design process, which led to a new idea which was feasible yet original. More research is, however, needed in order to investigate whether original, but not as yet feasible, ideas can more generally work as a catalyst in a design process that produces both original and realisable ideas.

What are the limitations of users? The main drawback in the users’ proposals is that they tend to have a more negative correlation between originality and producibility (2 0.702) than those of the professionals do (2 0.473). In other words, a highly original idea renders it hard to implement. The users’ poor scoring in the relationship between originality and

Business reframed by users Another inducement for involving users in service innovation is the knowledge that can be obtained concerning their needs and wishes, as well as their conception of what a service is. As seen in the study, at least some users did not differentiate between the service itself and the equipment (the mobile telephone) used to access the service. Several of the non-service suggestions by users perceived the mobile phone as a modern “Swiss army knife”, a multipurpose device that you always carry with you. This opens up new perspectives for professional service designers. A practical implication might be

236

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

that mobile service providers ought to reframe their current businesses. They might have to collaborate with mobile telephony equipment manufacturers in order to design better (more integrated) offerings, in order to satisfy user expectations.

using the example of the “newspaper-boy”. There is, accordingly, a need for further research on how the ideas generated should be processed in order to be useful to the company. The issue, thus, is not whether users can contribute to service innovation, but how their beneficial potential can be utilised in the best way. A favourable way of finding this out could be the application of the setting used in the study described. It has not only created a deeper understanding of the users’ contributions, but also a grand total of 123 ideas from 19 users. We can thus conclude that the set-up works quite well, both as an engine for producing new, promising service ideas and deeper understanding of the contribution of user involvement. This could be done in the form of action research. Together, researchers and practitioners explored how user-generated ideas could be utilised in order to bring benefit to the company. The research procedure presented in this paper can be seen as a first step towards a model in which user involvement in service innovation can be performed in a manner that serves both practitioners and researchers.

Conclusion The main research question posed by this article was: “Can users contribute something extra in the service innovation process, compared to professional service developers?” As indicated by the previous discussion, users, or more precisely their ideas, can be beneficial in several ways. As demonstrated, users contribute ideas that, on average, hold greater user-value and are more original. First, these ideas can be the input to a design process in which they are made feasible. Second, these ideas can be used as a learning tool in order to understand users better, e.g. their needs, wishes and expectations. Third, the users’ ideas can be used as a source of inspiration, i.e. a catalyst, for professionals to think “out of the box” and reframe the current businesses. An important contribution made by this paper is the intriguing discovery that users involved in a service innovation process seem to produce more original and valuable proposals than professional developers do. It has been known to happen that professional developers’ superior technical knowledge can act as a limitation to their ability to create more original ideas, i.e. a kind of rigidity in their thinking (see Leonard-Barton, 1992). The findings presented in the paper have illustrated that user involvement can vitalise a company’s innovation process. The problem is that the users’ idea proposals tend to be unfeasible to implement. Thus, should the conclusion be that it is, after all, unsuitable to involve users in service innovation? I argue that users can play an important role during this phase, but it is important to reflect on what their function should be. Users are not to be considered a substitute for professional service developers. It is a mistake to perceive the users’ proposals as mature and ready-toimplement service concepts. Rather, the ideas should be seen as a trigger to a design process in which some of the original features must be modified in order to enable implementation. In this way, users can be regarded as inspirers of new services, as previously demonstrated

Notes

237

1 The terms customer and user are often mixed in the research literature. In this paper, users are considered to be the ones that actually use a product, i.e. the end-users. Customers, on the other hand, are the ones paying for the product. In some cases, customers are also users, but not always. In this paper, however, the focus is on consumer products, which is why we can equate customers with users. 2 The research presented in this paper has been a part of the CuDIT research project, conducted partly in cooperation with doctoral candidate in psychology Per Kristensson and doctoral candidate in business administration Jonas Matthing, both sited at the Service Research Center at the University of Karlstad. 3 SMS is an abbraviation for Short Message Service, a technology for sending and receiving text messages using moble phones. SMS is defined within the GMS specification. GMS is a pan-European standard for mobile telephony. The system was introduced in Europe in 1992, but is today to be found all over the world. 4 The cost of sending an SMS was e0.15.

Benefits of involving users in service innovation

European Journal of Innovation Management

Peter R. Magnusson

Volume 6 · Number 4 · 2003 · 228-238

Amabile, T.M. (1996), Creativity in Context, Westview Press, Boulder, CO. Anderson, W.L. and Crocca, W.T. (1993), “Engineering practice and codevelopment of product prototypes”, Communications of the ACM, Vol. 36 No. 4, pp. 49-56. Bitner, M.J., Brown, S.W. and Meuter, M.L. (2000), “Technology infusion in service encounters”, Journal of the Academy of Marketing Science, Vol. 28 No. 1, pp. 138-49. Campbell, A.J. and Cooper, R.G. (1999), “Do customer partnerships improve new product success rates?”, Industrial Marketing Management, Vol. 28 No. 5, pp. 507-19. Christensen, C.M. and Bower, J. (1996), “Customer power, strategic investment and the failure of leading firms”, Strategic Management Journal, Vol. 17, pp. 197-218. Curtis, H. (2001), “I-m.o.d.e. strikes the right cord”, Electronic News, Vol. 47 No. 21, p. 24. Fesist, G.J. (1999), “The influence of personality on artistic and scientific creativity”, in Sternberg, R.J. (Ed.), Handbook of Creativity, Cambridge University Press, Cambridge. Gales, L. and Mansour-Cole, D. (1995), “User involvement in innovation projects”, Journal of Engineering and Technology Management, Vol. 12 No. 1-2, pp. 77-109. Gupta, A.K. and Wilemon, D.L. (1990), “Accelerating the development of technology-based new products”, California Management Review, Vol. 32 No. 2, pp. 24-44. Hennestad, B.W. (1999), “Infusing the organisation with customer knowledge”, Scandinavian Journal of Management, Vol. 15 No. 1, pp. 17-41. Holmquist, R. and Ekvall, G. (1986), BPE: Bedo¨mning av personliga egenskaper, Psykologifo¨rlaget, Stockholm. Iansiti, M. and MacCormack, A. (1997), “Developing products on Internet time”, Harvard Business Review, Vol. 75 No. 5, p. 108. Kupfer, A. (1993), “The race to rewire America”, Fortune, Vol. 127 No. 8, pp. 42-61. Leonard, D. and Rayport, J.F. (1997), “Spark innovation through empathic design”, Harvard Business Review, Vol. 75 No. 6, pp. 102-13. Leonard-Barton, D. (1992), “Core capabilities and core rigidities – a paradox in managing new product development”, Strategic Management Journal, Vol. 13, pp. 111-25. McQuarrie, E.F. and McIntyre, S.H. (1986), “Focus groups and the development of new products by

technologically driven companies: some guidelines”, Journal of Product Innovation Management, Vol. 3 No. 1, pp. 40-7. Mullins, J.W. and Sutherland, D.J. (1998), “New product development in rapidly changing markets: an exploratory study”, Journal of Product Innovation Management, Vol. 15 No. 3, pp. 224-36. Norling, P. (1993), “Tja¨nstekonstruktion” (“Service design”), PhD thesis, University of Stockholm, Stockholm (in Swedish). Parasuraman, A. (2000), “Technology readiness index (TRI) – a multiple-item scale to measure readiness to embrace new technologies”, Journal of Service Research, Vol. 2 No. 4, pp. 307-20. Prahalad, C.K. and Ramaswamy, V. (2000), “Co-opting customer competence”, Harvard Business Review, Vol. 78 No. 1, pp. 79-87. Rogers, E.M. (1995), Diffusion of Innovations, 4th ed., Free Press, New York, NY. Scheier, M.F. and Carver, C.S. (1985), “Optimism, coping, and health: assessment and implications of generalized outcome expectancies”, Health Psychology, Vol. 4 No. 3, pp. 219-47. Sinkula, J.M. (1994), “Market information process and organizational learning”, Journal of Marketing, Vol. 58 No. 1, pp. 35-45. Souder, W.E. (1989), “Improving productivity through technology push”, Research-Technology Management, Vol. 32 No. 2, pp. 19-24. Veryzer, R.W.J. (1998), “Key factors affecting customer evaulation if discontinuous new products”, Journal of Product Innovation Management, Vol. 15 No. 2, pp. 136-50. von Hippel, E. (1977), “Has a customer already developed your next product”, Sloan Management Review, Vol. 18 No. 2, pp. 63-74. von Hippel, E. (1978), “Successful industrial products from customer ideas”, Journal of Marketing, Vol. 42 No. 1, pp. 39-49. von Hippel, E. (1982), “Get new products from customers”, Harvard Business Review, Vol. 60 No. 2, pp. 117-22. von Hippel, E. (1986), “Lead users: a source of novel product concepts”, Management Science, Vol. 32 No. 7, pp. 791-805. von Hippel, E. (1988), The Sources of Innovation, Oxford UP, New York, NY. von Hippel, E., Thomke, S. and Sonnack, M. (1999), “Creating breakthroughs at 3M”, Harvard Business Review, Vol. 77 No. 5, pp. 3-9. Wikstro¨m, S. (1995), “The customer as a co-producer”, European Journal of Marketing, Vol. 30 No. 4, pp. 6-19.

References

238

The introduction of TQM in research and development

The TQM trajectories in research and development: two Italian cases Giorgio Petroni Alberto Ivo Dormio Anna Nosella and Chiara Verbano The authors Giorgio Petroni is Full Professor of Economics and Innovation and Anna Nosella is a PhD student in Engineering Management and Economics, both at Padua University, Padua, Italy. Alberto Ivo Dormio is Assistant Professor, Department of Economics and Technology, San Marino State University, Italy. Chiara Verbano is Researcher, Department of Industrial Engineering, University of Parma, Parma, Italy. Keywords Total quality management, Research, Development, Organizations, Innovation, Italy Abstract Total quality management programmes applied to research and development often produce great improvements in productivity because they follow trajectories that are lateral and parallel to the principal thrust which is aimed at client satisfaction. Such improvements are obtained through a better definition of the objectives of research activity, and through organisational overhauls which highlight the contribution to the innovation process made by the marketing, engineering, and production functions, through the adoption of more efficient management systems (programming, control, and management of human resources in particular). These changes are met with strong resistance from researchers and technologists who, during the start-up phase of such programmes, often refuse to operate in accordance with the internal supplier-client scheme. Such resistance, which was confirmed by the two Italian case studies presented, can be overcome by the intervention of strong leadership and with the setting up of an intense training programme.

Electronic access The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1460-1060.htm

European Journal of Innovation Management Volume 6 · Number 4 · 2003 · pp. 239-252 q MCB UP Limited · ISSN 1460-1060 DOI 10.1108/14601060310500959

In the mid-1980s the USA started the first total quality management (TQM) programmes applied to research and development (R&D) activities. Among the first companies involved were 3M, Eastman Kodak and Xerox (Davidson and Pruden, 1996; Davidson and Chatter, 2001). The early 1990s saw a considerable increase in the number of companies, also in Europe, developing such programmes (EIRMA, 1993). This was due to a series of reasons that stimulated a significant increase in R&D productivity; for example: (1) The reduction in the overall quantity of resources dedicated (from the mid1980s) to industrial research in all developed nations, with the exception of Japan. (2) The greater intensity of competition between companies due to the spread of “globalisation”. (3) The need to reduce the period of time before new products are introduced to the market (time to market). (4) The significant and progressive increase in the costs of R&D activities seen during the 1990s. In this context “relationship networks” took on greater significance. These networks involved companies in the production and spreading of innovation and, in addition, there was an ever increasing number of external research structures (often of small or medium size, but particularly aggressive) linked to the laboratories of large companies (Marrien, 1994). There was, therefore, an inversion of the trend of the previous 30 years (from the 1950s to the 1970s) which had pushed many “technology intensive” companies towards vertical integration, which was often achieved through expensive acquisitions (EIRMA, 1993). In addition, in an attempt to recover the productivity margins of industrial research, programmes following the principles of quality assurance were introduced. The adoption, in particular, of the TQM paradigm, aimed at the efficacy of the supplier-client relationships, led to an overall review of the modus operandi of the company, introducing important organisational and managerial changes (Krapp, 2001; Wolff, 1991). Right from the very start, in 1986-

239

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

1987, the approach to TQM of some large US companies (mentioned above) involved, for the R&D function, the identification of the other main company activities; engineering, production and marketing as internal suppliers. As detailed below, this was a problematic experience due to both cultural resistance from researchers and technologists and because the supplier-client scheme was not always suitable for organising and regulating the relationships between R&D and the other company functions identified as internal clients.

the company (the part involving the company); the rest of the centre’s budget comes from offering services (also to the “business units”) at market prices, therefore operating in competition with other research structures. A final relationship between the research structure and the market described in the literature refers to those “applications development” programmes which companies propose to their clients and which are performed by the R&D function (Lovett, 1995). This, for example, is the case with company research laboratories involved in the production of polymers that assist their clients in the R&D of new applications for traditional polymers, or in improving the transformation processes of these polymers. In these cases the adoption of TQM was to respond to the primary requirement of guaranteeing that the client’s needs were suitably met, the client having made a request within the dynamics of the market. Such a request could potentially be satisfied by a number of suppliers. Because of their coherence with the typical objectives of quality (customer satisfaction) programmes, the above-mentioned TQM approaches in R&D are described here as “orthodox” approaches. In many other cases (the majority found in the literature), the set up of the TQM programmes involves, as already indicated, the R&D function taking on the role of supplier, where the “internal” clients are the production, marketing, and procurement departments, etc. In order for this “internal supplier-client” scheme to be effectively set up, several factors should be considered (Davidson and Pruden, 1996): . There must be a clear definition of the role, the objectives and the content of the projects assigned to R&D. . The role and the projects must be shared (and often even commissioned) by the internal clients. . Performance “standards” must be established for the R&D activity; these standards must make it possible to make timely evaluations of the performances. . Formal rules and behaviour procedures (preparation of documents, periodic checks etc.) must be set up in order to ensure that the pre-established standards and objectives are reached.

The TQM approaches to R&D activities An analysis of the literature shows that there are basically two types of TQM approach to R&D. The first “orthodox” approach concerns cases where the relationships between the organisation purchasing the industrial research activity (client) and the structure performing the requested research activity (supplier) are regulated by market transactions. This occurs, for example, in the case of research centres which work institutionally for third parties; in other words companies, or public bodies which, through contracts, entrust these centres with the task of establishing research projects. Such an example found in the literature is that of the Finnish VTT Electronics research centre (Auer et al., 1996). The offer of research services also spread in Europe in the early 1990s when many public research centres adopted this approach. These include the Dutch TNO and the German Fraunhofer Institute. As a consequence of the increase in competition, the contract based approach to research projects often led to the centres providing the research services adopting formal procedures and standards. This was clearly due to the necessity to suitably satisfy the needs of the clients and, at the same time, to put into place a more effective competitive attack strategy. Market conditions (or rather those that were very close to such conditions) are indicated in the literature when descriptions are given of the transactions between the research centres operating at the “corporate” level and the “business units” of the same company (Roberts, 1983, 1991). In this case only part of the costs of the “corporate” centre activities are covered by

On the basis of this type of set-up, the quality programmes follow paths (lateral trajectories) which, in part, diverge from the traditional

240

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

routes aimed at ensuring that the goods and services requirements of the end-user are met, operating directly as a player in the market? The development of “quality programmes” aimed at the needs of internal clients leads to managerial and organisational innovations of primary importance.

Figure 1 The R&D matrix structure scheme

The development of “lateral trajectories” with the adoption of the TQM With regard to the managerial and organisational innovations that arise out of the adoption of the TQM, one should immediately note the strict link that is created between the development of the business of the divisions (business units), which are the project clients, and the know-how and skills of the researchers and technologists of the R&D corporate staff. Here there was a significant cultural change in the technology intensive companies that adopted the TQM programmes. Technological values became more important than those of science and, similarly, managerial behavioural elements (characterised by the rigorous adherence to of the pre-established standards, time periods, use of resources, costs, etc.) took on greater importance than attitudes aimed purely at the production of know-how. This cultural and operational re-orientation led to a formal modification of the objectives of the central and divisional research staff (Marrien, 1994). This change had the following repercussions at the level of organisational structures: . the reduction in the number of staff in the corporate research centres; . the development of a matrix scheme (following the model shown in Figure 1); and . the growth of a particular structure (business team) created to give a stable set up to the collaboration between researchers and technologists and staff specialised in technical assistance, marketing, testing and finance. In the case of corporate laboratories, the offer of research projects within the context of competition with external laboratories led to the adoption of benchmarking programmes and often to an increase in their responsibilities in the fields of marketing and engineering. This guaranteed the client (even when dealing with an internal client) adequate

assistance for the entire innovation process, from the production of the prototype to the development of the final product and the industrial production phase. In addition, the development of the TQM programmes has, for both the corporate laboratories and the divisional laboratories, led to a close interaction with the production and marketing functions which, as we have indicated, have often taken on the role of internal clients (operating divisions identified within the organisational structure as business units). At the same time, this greater interaction was regulated by precise performance standards that form the basis of any quality programme. This led to a significant strengthening of the project management system, leading to great improvements in the productivity of innovation projects (Roberts, 1991). Here, for example, precise standards were adopted to evaluate the output of the research and innovation projects. The contribution expected of the R&D staff by the clients, including internal clients, (and measured on the basis of predefined standards) pushed the research laboratories to cultivate more consolidated areas of know-how and competence. Projects with a higher level of uncertainty were eliminated. A research model was then adopted which was clearly aimed at developing and consolidating the know-how already possessed and a coneshaped development was therefore activated (see Figure 2). According to this scheme, the further development of pre-existing know-how often produces innovations in adjacent areas, innovations that represent new applications for products or processes already in existence.

241

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

Figure 2 The conical profile of the new research and development model

spaces available for explorative research. This is a consequence of the greater attention being given to the businesses currently managed by the company.

Difficulties and success factors in the application of the TQM in R&D

In this sense, the route leading to the developing of know-how (vertical dimension) usually also leads to a widening of the innovation spaces (horizontal dimension). Interesting cases are to be found here, for example, with new materials obtained from the inclusion of nanopolymers in cement structures or, in the field of traditional biotechnologies, by the use of milk ferments for the conservation of ensiled products or for animal feed (zootechnical products) (Petroni and Verbano, 2001). Finally, the adoption of TQM programmes applied to R&D activity has also involved important changes in the management of the human resources used in innovation processes. In particular, the definition of performance standards required of the research function has produced a more solid base, also including quantitative elements and performance appraisal procedure for researchers and technologists (Patino, 1997). The close collaboration required between marketing, R&D, engineering and production staff has greatly favoured the development of the dual ladder system where, as occurs in the best situations, the professional development of some researchers and technologists involves growth along routes outside the area of research and planning, which is often the sector where the technical personnel have their first work experiences (see Figure 3). The overall orientation of the R&D activity that took shape as a consequence of the adoption of the TQM, alongside the managerial and organisational innovations, has led to an unavoidable reduction in the

The above-mentioned managerial and organisational innovations (many of which are found in the case studies reported below) arise basically from the efforts made by the companies to overcome the difficulties generated by the adoption of TQM. It may be useful to remind the reader briefly of the following factors. The most important objection, as already indicated, concerns the presumed lack of suitability of the supplierclient scheme (which forms the basis of the TQM) in the representation of innovative processes. The route along which industrial research activity develops requires it to be continuously interacting with marketing and production, which operate directly to create innovative output (Bowman and Hurry, 1993). In a similar interactive process (which is circular, as shown in Figure 4) it is often difficult to make a clear separation between the contribution of the researchers and that of the engineers who plan and manage the production plants. The technical specifications of the plants are set in accordance with the performances expected of the product, which are often suggested by the marketing experts. Therefore, since it is often difficult to define where the contribution to the innovation process of the R&D function terminates and where that of the marketing function and production function begins, it becomes problematic to identify and measure precisely the performances of the R&D as supplier of internal client. The second objection concerns the uncertainty of the output of R&D activities. This makes it, by definition, difficult to evaluate the performances so that the emphasis in the TQM programmes tends to be shifted towards the process (in other words, onto the route followed to reach the future innovation) rather than to the result (Faulkner, 1996; Montana, 1992). However, an accurate formalisation of this process (the third objection that is raised against the introduction of TQM programmes) can reduce the margins of independence of the

242

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

Figure 3 The dual ladder scheme and the professional development routes outside R&D

Figure 4 The process of interaction between R&D, marketing and production

trajectories that were mentioned above. The factors of success of such programmes consist of (Davidson and Pruden, 1996; Roberts, 1983): . the strong leadership shown by the top management in sponsor and promoting the programmes in question; . the creation of a TQM manual which highlights the roles and responsibilities of the various functions participating in the programme, and which also defines the criteria for measuring the expected performances; and . the intense training courses for all those involved in the programme.

researchers and technologists, stifling their creativity. In R&D activity, it is not possible to adopt a “golden rule” which governs the TQM programmes, and this is expressed in the phrase “do things right the first time and every time” (Brenam, 2001; Patino, 1997). Other reasons for resistance to the adoption of TQM programmes emerge from empirical analysis performed on a large number of cases (Davidson and Pruden, 1996; Davidson and Chatter, 2001). Among these is the difficulty, expressed by the researchers and technologists, to accept as important innovations (i.e. endowed with true technicalscientific worth) those that appeal to clients or those that satisfy the needs of production and marketing in the case of quality systems with internal clients. Despite these difficulties, many companies, especially American and Japanese, have successfully started TQM programmes based on the internal supplierclient logic, with the overall objective of increasing the R&D productivity and therefore obtaining important improvements and advantages by following the lateral

The objectives and the method of analysis used in the Italian cases There follows a report of the results of an analysis of the TQM system adopted by two large Italian research laboratories: the central research laboratory of the company STMicroelectronics, one of the largest producers of semiconductors; and the FIAT research centre which, created as the central research laboratory (corporate laboratory) of the Italian manufacturer, has been open to the international market and offering its research services since 1994 (see the Appendices 1 and 2). The reconstruction of the two case histories was performed with the following objectives: . To identify the reasons that led to the initiation and development of the TQM programme. . To detect the effects generated by the programme, with reference to both customer satisfaction and to any

243

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

.

.

.

managerial and organisational innovations produced. Such innovations are, in particular, those concerning changes to the structure and to the organisational processes, and those concerning the introduction of new operating systems (for example project management, human resources management models, etc.) or the improvement in existing systems. These innovations also include those concerning cultural changes (values and behaviour) shown by the researchers and technologists. To reconstruct the cases under examination in such a way as to detect the presence of any TQM programme lateral trajectories which accompany (and often overlap) the main route which leads to the satisfaction of the needs of the clients (who, in this case, are mainly internal clients). To detect any relationships that might exist between the effects of the programme aimed at internal clients and those orientated towards external clients. To analyse the success factors of the programme and to highlight the restraints and obstacles encountered.

The information and data required for the reconstruction and interpretation of the two case studies was collected during a long series of interviews with both the TQM programme managers and with those involved in the programme, in other words, external and internal clients and suppliers.

Structure and effects of the total quality programme developed by the Fiat research centre (CRF) The CRF, which was created in 1973, operated for more than 15 years as a corporate research centre (as detailed in the enclosed chart). In this period, it worked mainly for the Automobile Division and the Industrial Vehicles Division (IVECO), though FIAT was (and still is) a well diversified company, operating also in the sectors of earth moving machines (tractors and scrapers), basic metallic materials (TECSID Division), rail transport, aerospace and engineering. One of the most important innovations produced by the CRF was the injection system for diesel engines that goes by the name of “common

rail”, now used by most automobile manufacturers. In 1989 the new chief executive officer (CEO) introduced a quality system which involved both the internal CRF functions and the FIAT Divisions. The latter had been clients whose budget, until 1994, had also included the costs of the CRF itself. Over a five-year period the quality system developed in three directions: (1) A new process organization was set up which, from the original top down approach, progressively involved the entire management. The system was set in motion with the birth of a four-year quality plan which was created as a typical expression of the vertex of the research centre. (2) Criteria were progressively set up for the evaluation of the performances of the various operating units (competence units and project units) and of the individual researchers and technologists. As an initial step to the definition of such standards, a significant effort was made to identify important performance indicators. To this framework of chemicophysical standards (concerning tests of the functioning of equipment, machines and materials) the professional standards of the researchers and technologists were added. Among these standards one should remember the respect afforded to the most scientifically accepted research methods, the respect of the programmed time scales in the development of the project, and the respect afforded to the budget and the formal evaluation of the project results. From this point of view, the programme evolved to produce a quality manual typical of a research structure that operates in accordance with the internal supplier-client circuit. (3) The adoption of the quality methods and values by the personnel operating at the various levels of responsibility was encouraged in order to create a harmonious process. This result was also achieved as a consequence of an intense training programme developed over time. At the end of this long process, the CRF was certified in accordance with the various editions of the ISO 9000 standards whose content has been progressively updated (9001, 9002, . . . 9000/2000).

244

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

In this first period (1989-1994), when the CRF operated basically for the FIAT Divisions, the quality system operating scheme was that outlined in Figure 5. This is a matrix organisation scheme where the contribution to the development of the projects from the various units of competence of the Centre (performances, time periods, cost limits, etc.) has been regulated by precise standards. The project leader in this scheme represents the purchaser of the research project and also has the responsibility, among other things, to check that the quality standards and rules are observed. Since 1994 the CRF has operated in the international research market. The projects commissioned by the FIAT Divisions are also acquired in competition with other research structures, such as the Dutch TNO, the Stanford Research Institute, the German Fraunhofer Institute, Battelle etc. This new role for the CRF received significant help from the organisational and managerial changes seen in the previous five years. In fact, the development of the quality programme, which went on to be defined as the TQM programme, had a profound influence (even before the approach to the market) on the organisational set-up and culture of the centre. In particular, the workforce has been greatly reduced and the organisational roles and functions have been defined with greater accuracy. Some operating systems have been made significantly more effective. The project

management system has, for example, been cleansed of obsolete practices, and the system for evaluating the performances and potential of the scientific personnel has been made more effective. In addition, a systematic benchmarking process was started which, among other things, was required for a better definition of the research methods and the costs-performances ratios of the projects. Another reason for the overall improvement in the productivity of the CRF was the adoption of a rigorous and systematic evaluation of the technico-scientific and economic results of the projects which was performed after their conclusion. There have also been great cultural improvements in management and among the researchers and technologists, who initially showed strong resistance to accepting the logic of the internal market and competition. Many of these people have yet to overcome the conflict felt within the role, due to the fact that they belong both to the company and to the scientific community. As members of the company, the researchers must accept the patent protection of the know-how produced, but to which the scientific community would wish to have free access since such knowledge is considered pertinent to all researchers operating in the same discipline. In addition, the re-training of the researcher towards another technicoscientific area, which is sometimes required by the company as a consequence of changes in outlook and company programmes, becomes

Figure 5 The CRF quality system operating scheme in the period 1989-1994

245

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

an element which lowers the standing of the researcher within his scientific community. Finally, another reason for the uneasiness among the scientific personnel is generated by the implicit evaluation of their work by the client, who often has very little understanding of the method and scientific and technological know-how used. In brief, one can see that in the CRF-Fiat experience, each of the two phases of the development of the quality programmes have led to distinct organisational and managerial changes. In particular, the emphasis of the first phase (1989-1994) was on the “culture of efficiency”. Here the roles of the corporate research structures, the divisional research structures and the research interface functions in the innovation process (marketing, engineering, production) were clarified. The reference standards for the evaluation of expected performances were defined and the project management was significantly improved by adopting more rigorous methods for the selection of project leaders (who also took on the role of research clients) and developing a more rigorous control of the time periods and costs. A formal method was then adopted for the evaluation of the output of the projects commissioned to the CRF, based on the following elements: . impact of project outcome on the business performance; . level of growth of acquired know-how and competencies; . any reduction in the technological distance from competitors; . the conviction that it is possible to embark on a new route; . acquisition of technical know-how to use as an exchange asset/tool; . level of assigned financial resources; and . the formal evaluation of the performances of the researchers and technologists.

Structure and effects of the TQM in the ST-Microelectronics central research laboratories

In the second phase, which saw the entry of CRF into the free market, the business culture gradually started to dominate. Important changes were made to the organizational structure with the elimination of many traditional staff, and new organizational units were created that were able to supply a wide range of technological and managerial services. Stimulated by international competition, the CRF Management performed periodic benchmarking initiatives.

The ST-Microelectronics R&D quality programme (which, with time, has reached the level of a TQM programme) was started in 1987 and was developed with increasing investments provided by the company management. The importance given to such a programme was a result of the intensity of competition that had developed at the global level in a sector with very few important producers. The tendency towards concentration was also a result of the significant emphasis on research that was required in order to compete successfully. In recent years competition has developed primarily in the technological field. The development of the relevant technologies (silicon cutting, plasma process technology, photocomposition, etc.) is so rapid that it produces a progressive reduction in the life cycle of the semiconductors. In fact, each generation of these products has a market life of no more than one year. The critical nature of this technology not only justifies the large investments in R&D (where about one-sixth of the company’s workforce is employed), but also provides the reason for the intense collaboration between the central staff and the R&D divisional units. The microchips and non-volatile memories that are manufactured are essential elements in the development of electronics and, therefore, of the final items produced by the clients of STMicroelectronics (electronic equipment for cars, cell phones, computer memories, etc.). In particular, the task of the research centre is that of supplying the Divisions with new technological platforms; in other words, nonvolatile memories with microchips that are able to contain a quantity of information which, with the succession of generations of such products, is growing in an exponential manner. The Divisions, in their turn, use these platforms to create the circuits in accordance with the design of the product intended for the final client. It is therefore clear that the structure and reliability of the technological platforms supplied to the Divisions must be subjected to a rigorous evaluation process, following the scheme shown in Figure 6, in order to avoid negative consequences for the reliability of the

246

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

Figure 6 The quality route taken by the ST-Microelectronics central research and development structure

performances of the products put together by the Divisions for their clients. At the end of this course of action, the central R&D staff issue a quality certification for the new technological platforms. This course then continues within the operational Divisions which certify the electronic equipment supplied to the final client. STMicroelectronics, for example, has obtained the QS 9000 certification for the apparatus supplied to the automobile industry. The circuit of gradual innovations which led to the quality certification process required intense collaboration between the STMicroelectronics research and production functions and similar functions of the final clients who are, as already mentioned, large multinational manufacturers of automobiles, telecommunications equipment, computers

etc. Also with this company, the development of the quality programmes involved three stages: first, there was the testing and certification of the products, to then pass on to the setting up of an organic process to govern the quality; this required a thorough revision of the organisational structures, roles and operating systems. The innovations introduced represent the important lateral trajectories along which the effects of the TQM also developed.With reference to the last stage, one should note the development of a continuous benchmarking process that involved principally the R&D and marketing functions. The revision of the organisational roles and structures is a consequence of the need to clarify objectives and the responsibilities of the central staff and divisional units involved in R&D activity. Times, costs, and types of methods adopted in the research process were included as standard in the management of the projects, which then saw significant improvements. The elements required to undertake the role of project leader were indicated, which made it possible to perform a more rigorous selection of the staff for this role. Again in this area, a formal methodology was set up to evaluate the results of the individual projects on their completion. Finally, the researchers and technologists also showed strong resistance to the development of the quality system. When it was introduced, intolerance was shown on a number of occasions towards the quantitative standards expected by the procedure used to evaluate the performance of the divisional research units; in other words, by the clients of the central R&D staff. To eliminate these frictions, suitable formalisation was given to the professional development scheme inspired by the paradigm of the “dual ladder”. To make the reward system more effective, an internal academy (ST-Microelectronics University) was set up which gave visibility and professional standing to the best researchers and technologists. Many of these people were then involved in the training of research personnel for the TQM management. Finally, among the problems posed by the development of this programme, one should note the unavoidable transfer of technicoscientific knowledge which was seen with the tightening of collaboration relationships with the company-clients, who are the final recipients of ST products.

247

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

In brief, one can say that, in the case of STMicroelectronics, the adoption of a TQM programme has led to significant developments in the innovation process and, in particular, has also produced important innovations of economic-organisational nature. The TQM programme is inserted in the innovation process, which directly links the corporate R&D activity with the divisional R&D activity and with the requirements of the clients who buy the semiconductors. The divisional research in particular requires periodic progress in scientific knowledge and requires more advanced non-volatile memory models that are able to satisfy the needs of the clients. All in all, these new types of memory are essential for survival in the intense competition which, over the past few years, has developed between computer manufacturers, between mobile phone manufacturers, and also between manufacturers of electronic equipment for automobiles. Given this fierce competition, quality programmes have progressively moved from ensuring product performances to the quality of the managerial and organisational process. To survive against the competition, it is necessary to continuously compare one’s innovations with those of the other principal semiconductor manufacturers (where the leader in the field is still Intel). For this reason, for some years now, relevant benchmarking programmes have been established. At the same time it is also necessary to interact closely with the R&D, marketing and production structures of the clients (end users of the semiconductors). To make this interaction more effective, it was necessary for the R&D staff of ST-Microelectronics to widen their skills which were originally of the technico-scientific type, whereas now they have taken on know-how that is typically managerial (knowledge of markets, the production system of the clients, costs structures, etc.). There was, therefore, a move from the “closed” innovation system, in other words a system limited to technico-scientific aspects, to an “open” innovation system aimed at meeting the needs of the automobile industry, the manufacturers of computers and companies operating in the telecommunications sector. In this sense the R&D structures increasingly took on the appearance of real business teams which gradually transformed and enriched the

traditionally technical culture of the researchers and technologists.

Conclusions The literature on the TQM systems applied to R&D indicates two main directions along which such systems have developed: (1) In some cases these systems have generated real benefits for the companies by raising the level of satisfaction of their clients. This proved to be the case with laboratories that operate directly in the market offering research services, or when the TQM systems involve the divisional laboratories of “market pull” companies working directly in the development of new applications for products that already exist or in technical assistance to the clients. (2) In the majority of cases, the TQM systems adopted make it possible to reach important lateral effects, which concern the following: the achievment of the most effective organizational set-up, a continuous comparison (Bennowski, 1991) with the research performances of the laboratories of the most respected competitors (benchmarking), the development of new operating systems or the improvement of existing systems (e.g. the project management, the improvement in the management systems of researchers and technologists etc.). This occurred, in particular, when the TQM principles were applied to the central research laboratories or to the divisional laboratories of technology push companies. In general, the overall result is a significant increase in R&D productivity obtained in recent years. This result was greatly desired by companies where critical factors for success are to be found in the development of product and process innovation. This increase in productivity often came with a price; in the form of resistance (and sometimes protest) from the scientific personnel in the application of the TQM regulations. The researchers and technologists often object, for example, that the supplier-client scheme cannot adequately interpret the innovation processes which is really more complex than that which is taken into consideration by the TQM regulations,

248

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

which seem to be more suited to the representation of the dynamics of the production processes. This paper has presented two Italian case studies of the application of the TQM principles to central (corporate) R&D laboratories: the research centre (CRF) of the FIAT group and the central laboratory of STMicroelectronics. The analysis of these two cases confirms: . The importance of the lateral trajectories (significant improvement in the R&D organisational and managerial systems) achieved by the TQM programmes. . The beneficial influence that the said programmes can also create in the divisional research laboratories; this influence operates in favour of the direct development of the innovation into the market. . The strong resistance to the adoption of the TQM programmes by researchers and technologists which can often produce high levels of demotivation.

Faulkner, T.W. (1996), “Applying ‘options thinking‘ to R&D evalutation”, Research Technology Management, May-June, pp. 50-6. Krapp, M. (2001), “Quality assurance in research and development: an insoluble dilemma?”, Research Technology Management, September-October. Lovett, J.R. (1995), “Doing the right things right, all the time”, Research Technology Management, MarchApril, pp. 35-8. Marrien, B.A. (1994), “Putting quality in R&D process”, Research Technology Management, JanuaryFebruary, pp. 16-23. Montana, A.J. (1992), “Quality in R&D: if it isn’t perfect, make it better”, Research Technology Management, July-August, pp. 38-41. Patino, H. (1997), “Applying total quality to R&D at Coors Brewing Company”, Research Technology Management, September-October, p. 36. Petroni, G. and Verbano, C. (2001), “Cultura tecnologica, competenze e fattori di crescita nel settore delle biotecnologie: alcuni casi di piccole e medie imprese italiane”, in Ferrero, G. (Ed.), Network e Rapporti Internazionali, ASPI, INS-EDIT, Genova. Roberts, G.W. (1983), Quality Assurance in R&D, Marcel Dekker, New York, NY. Roberts, G.W. (1991), “Managing research quality”, Research Technology Management, JanuaryFebruary, pp. 28-41. Wolff, M.F. (1991), “Quality in R&D: it starts with you”, Research Technology Management, JanuaryFebruary, pp. 9-11.

Finally, in both of the cases that were examined, the decisive thrust in the TQM development was supplied by the strong leadership of the top management and by the intense training activity involving all those working in the process.

References Auer, A., Karjadainen, J. and Seppainen, V. (1996), “Improving R&D processes by an ISO 9001-based quality management system”, Journal of Systems Architecture, April, pp. 236-44. Bennowski, K. (1991), “The benchmarking bemwagon”, Quality Progress, January, pp. 19-24. Bowman, E.H. and Hurry, D. (1993), “Strategy through the option lens: an integrated view of resource investments and incremental choice process”, Academy of Management Review, Vol. 18 No. 4, pp. 760-82. Brenam, L. (2001), “Total quality management in a research and development environment”, Integrated Maufacturing Systems, No. 2, pp. 94-102. Davidson, J.M. and Chatter, J.I. (2001), “Evaluating TQM’s legacies for R&D”, Research Technology Management, January-February, pp. 10-12. Davidson, J.M. and Pruden, A.L. (1996), “Quality deployment in R&D organizations”, Research Technology Management, January-February, pp. 49-55. European Industrial Research Management Association (ERIMA) (1993), Report on Total Quality in R&D, ERIMA, Paris.

Further reading Marrien, B.A. (1999), “Putting TQM into performance review process”, Research Technology Management, March-April, pp. 39-43. Walsh, W.J. (2000), “Developing the metaquality product”, Researh Technology Management, MayJune, pp. 44-9.

Appendix 1. The Fiat Research Centre (CRF) The CRF was founded in 1963, as an independent company controlled by the Fiat group, to perform research activities for the various “business units” of the automobile group. Its budget for 2002 forecasts a sum of 115 million Euro and it employs 960 personnel made up mainly of engineers, physicists and chemists. To these one adds another 200 scholarship holders and people involved in “work experience”. In addition, there are collaborations with about 200 international research centres and universities. The CRF research activity consists in the elaboration of solutions to technological and managerial problems mainly for the automobile sector. The CRF

249

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

works both for the FIAT group, which is responsible for 50 per cent of total annual sales, and for other private organizations and companies (such as Bosh, Borg Warner, Tetrapak) and public bodies (research projects financed with European funds). Among the most important innovations developed by the CRF, one should remember the “common rail” injection system for diesel engines. In addition, the CRF performs training activities for both the Fiat Group and for those linked to it. The main competitors in the technological field are AVL, TNO, Fraunhofer Institute, Ricardo; whereas for managerial methodologies competition is provided by McKinsey, Battelle Institute, Arthur D. Little, Stanford Research Institute. Its organisational structure comprises two principal nuclei (product managements: Vehicles and Engines) and three smaller nuclei made up of: innovative product technologies, business information technology and innovative process technology (see Figure A1). The quality system in the CRF was started in 1989, following the arrival of the new CEO. His first move was to promote the approach to the problem of quality by increasing the sensitivity of the top and middle management.

This move was also due to a specific Fiat Group policy to tackle the management of quality in an organic manner within the entire group. After this move to increase the sensitivity of the management, a four-year quality plan was set up which, replacing the ex-post statistical production control methods, started out with an analysis of the criticalities. The plan also involved investments in new plants, new organizational structures and intense training programs; all with the aim of obtaining an improvement in efficiency and productivity, together with a process of integration between the internal sections of the CRF. The CRF quality plan, which was aimed primarily at the Automobile Division of the FIAT Group involved, among other things, the development of “robust design” methodologies, “design of experiments” and the Taguchi method. Then the TQM methodology was developed (which, in the specific case, was called d x q) and applied to all the CRF activities. The TQM methodology had the following objectives: . the reduction of the bureaucratic circuits and costs, greater reliability of the experimental tests, better financial planning;

Figure A1 The organisational structure of the CRF

250

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

.

.

.

the involvement of all employees via the internalisation of the customer-supplier processes; the creation of a widespread quality culture; and to focus all organisational levels on, and make them sensitive to, the quality objective.

Table AI The main competitors in the semiconductors sector (millions of dollars)

Following on from these approaches, three important innovations were created: (1) the revision of the organisational structure; (2) the definition of a new evaluation scheme for personnel performances; and (3) the re-planning of training programmes on the basis of real needs that emerged. The customer-supplier scheme has certainly favoured the development of more rational and efficient organisational behaviours; these proved to be particularly useful especially when the CRF started to acquire research projects from external bodies. One can say that the success of such projects is largely due to the experiences matured in internal quality.

Appendix 2. The ST-Microelectronics (STM) laboratories The company STM was founded in 1987, the result of an entrepreneurial experience dating from the 1950s and based on the collaboration between the company Thompson and a number of Italian companies. The company is the 6th biggest producer of semiconductors in the world with total annual sales of 7,813.2 million dollars and a gross margin of 3,306.2 million dollars. The company’s main competitors are Intel, Toshiba, Nec, Samsung and Texas Instruments (see Table AI). The company headquarters is at Geneva and its three main research centres are located at Agrate Brianza (Milan, Italy), Rousset (Gre`noble) and Singapore. At the end of 2000 it had 43,000 employees. The semiconductors produced are used in various sectors: electronic consumer goods (25 per cent of total sales), automobiles (9 per cent of total sales) computers and peripherals (22 per cent of total sales), telecommunications (30 per cent of total sales) and other industrial uses (14 per cent of total sales). There are 6,800 people employed in research, with a budget of 1,026 million dollars, working mainly in the

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Intel Toshiba NEC Samsung Texas Instruments STMicroelectronics Motorola Hitachi Infineon Technologies Micron Hynix (Hyundai) Philips Semiconductors Agere (Lucent Tech.) Mitsubishi AMD Fujitsu IBM Matsushita Sony Sanyo Analog Devices Sharp LSI Logic Agilent Technologies National Semiconductor

Sales (2000)

Market share (per cent)

30.400 11.388 10.900 10.592 9.200 7.910 7.875 7.286 6.853 6.448 6.400 5.837 4.875 4.740 4.644 4.470 4.329 4.150 3.300 3.260 2.571 2.550 2.448 2.400 2.301 164.826

15 6 5 5 5 4 4 4 3 3 3 3 2 2 2 2 2 2 2 2 1 1 1 1 1 82

Source: Cahners In-Stat Group, InSerch Reserch

Milan and Grenoble premises and in the divisional laboratories. The research activity, which has produced more than 19,000 patents, is also based on pre-competitive European research projects and on collaboration with more than 40 universities and research organisations both public and private. One element that has favoured the innovation processes has been the collaboration in research programmes with its own customers, such as Alcatel, Nokia, Novell Networks, Philips. The organisation structure of STM is arranged in divisions organised as profit and loss centres and set up on the basis of the products-markets relationship (see Figure A2). These divisions plan, develop and market the various types of semiconductors. There is a strong interaction between the divisional structures and the central R&D nucleus. The latter, in particular, is involved in the preparing of basic technologies and products (new silicon platforms, non-volatile

251

The TQM trajectories in research and development

European Journal of Innovation Management

Giorgio Petroni, Alberto Ivo Dormio, Anna Nosella and Chiara Verbano

Volume 6 · Number 4 · 2003 · 239-252

Figure A2 The organisational structure of ST-Microelectronics

memories, the use of photocomposition etc.) which are supplied to the Divisions. The divisional R&D nuclei, in their turn, plan and create the semiconductors on the basis of the requests from the final customers. The development of a quality system within the R&D activities involved the following phases: . The preparation of a testing and validation methodology, in terms of performances and reliability, of the products that the R&D centre supplies to the individual divisions and which the latter then customise. The process was, therefore, imposed in the customersupplier context. . The planning and creating of a monitoring system for the quality of the

.

production process for the semiconductors supplied to the individual Divisions. This phase proved to be of considerable importance because of the progressive reduction in the life cycle of these types of products and, therefore, the need continually to innovate the production processes. The re-planning of the organisational structures and the re-definition of roles using new project management methodologies.

The TQM programme, as for all company matters, received strong promotion from the top management which dedicated significant investments, especially for the training of personnel.

252

Technophobia, gender influences and consumer decision-making for technology-related products David Gilbert Liz Lee-Kelley and Maya Barton The authors David Gilbert is Professor of Marketing, Liz Lee-Kelley is Lecturer in E-commerce and Maya Barton is Researcher, all at Surrey European Management School, University of Surrey, Guildford, UK. Keywords Gender, Product technology, New products, Internet, Consumer behaviour Abstract Mobile Internet technology (MIT) is an extension of the Internet beyond the static terminal of the personal computer or television. It has been forecasted that by the end of 2005, there will be almost 500 million users of mobile m-commerce, generating more than $200 billion in revenues. Contributes to the body of knowledge on how to approach the study of MIT products. Proposes that consumer perceptions of MIT products can lead to dichotomous decision making and argues that the challenge for marketers is to harness and fit this dichotomy to the MIT product continuum through an understanding of consumer psychological and attribution factors. The overall findings indicate that technology anxiety correlates with demographic variables such as age, gender and academic qualifications. Therefore, the implications of the study are that technology product engineering and marketing should recognise the importance of: study of the psychosocial needs of technology products, human factors in engineering design which need to fit these needs; and developing product designs facilitating consumers’ psychosocial needs.

Electronic access The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1460-1060.htm

European Journal of Innovation Management Volume 6 · Number 4 · 2003 · pp. 253-263 q MCB UP Limited · ISSN 1460-1060 DOI 10.1108/14601060310500968

Introduction This study draws on a host of empirical works such as Brosnan’s (1998) research into technophobia; Bem’s studies (1974, 1977, 1981a, b) of psychological gender in connection with technophobia, Bandura’s (1977a, b) social learning and self-efficacy theories, the investigation of cognitive/spatial abilities (Turkle, 1984, 1988; Turkle and Papert, 1990) and cognitive style (Witkin et al., 1977), including Davis’ (1986, 1989) original technology acceptance model (TAM) and the psychological factors leading to the expansion of the TAM (Brosnan, 1998). It explores attribution theory (Kelley, 1967) and an attribution process meta-analysis (Mizerski et al., 1979) related to aspects of psychological acceptance of technology. The aim of such a comprehensive evaluation of the literature is to derive an approach to assessing psychological gender influences on consumer behaviour object-perception in which technology acceptance theories and attribution process theories are incorporated. While the terms relating to “technology” are capable of richer and complex technical connotations, their use in this paper will only be in the context of a computer-mediated activity. References made to a mobile Internet device (MID) will be generic in nature, e.g. “pocket computer” for a MID that may resemble a computer and/or a “wireless protocol application (WAP) phone” for a MID that may look like a phone.

Technophobia The first US national survey in which a prevalence of negative attitudes towards new technology was evident was attributed to Lee (1970). Three decades later, a British public survey commissioned by Motorola (1996) reported that 49 per cent of the British general public did not use a computer and 43 per cent did not use any form of new technology. There also appeared to be a distinct lack of interest in the Internet as one quarter (of the 85 per cent knowing the Internet) reported a distinct lack of interest in it. Brosnan (1998) highlighted that the psychological factors such as anxiety or negative attitudes could have prevented interested individuals from utilising the new technology. Computer phobia or technophobia is a complex interplay of behavioural, emotional and attitudinal

253

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

components. Jay (1981, p. 47) defines it as “a resistance to talking about computers or even thinking about computers; fear or anxiety towards computers; hostile or aggressive thoughts about computers”. Rosen and Maguire (1990, p. 276) characterize technophobia as “anxiety about current or future interactions with computers or computer-related technology; negative global attitudes about computers, their operation or their societal impact; and/or specific negative cognitions or self-critical internal dialogues during actual computer interactions or when contemplating future interaction”. Rosen et al. (1993) posit that there are three types of technophobes based on an escalating anxiety of how they would react while using a computer – the “uncomfortable users”, the “cognitive computerphobes” and the “anxious computerphobes”. In specific cases of consumer decision making it is likely to be anxiety aversion that results in a reduction of choice. Computer anxiety is an example of state anxiety, conceptualised as a transitory condition, which fluctuates over time. Many writers such as Dukes et al. (1989) and Moldafsky and Kwon (1994) have given their support to the existence of computer anxiety. Attempts to study and assess computer anxiety have resulted in measurement scales such as the computer anxiety rating scales (CARS) using Likert-type self-endorsements to rank anxiety creating items (Heinssen et al., 1987). Other developments included Raub’s (1981) ATC, CAS by Loyd and Gressard (1984), Maurer’s (1983) CAIN and BELCAT-36 by Erickson (1987). Brosnan (1998) has shown the relationship between computer anxiety and demographic variables such as age, gender and academic qualifications should not be overlooked. However, he indicates females may not simply be less positive about computers as certain new technology items are of interest to them. He surmises that females are likely to use computers only when they have a direct and useful purpose in their lives and that there is a convergence of the notions of attitude and computer anxiety to arrive at an overall concept of technophobia. Rubin and Greene (1991) however, found that it is possible to explain apparent sex differences in personality and intelligence using psychological gender rather than biological sex.

Gender and technology Consensus among numerous studies is that females report higher levels of computer anxiety than males (Igbaria and Chakrabarti, 1990; Okebukola and Woda, 1993; Farina et al., 1991; Brosnan and Davidson, 1996). Furthermore, Brosnan (1998) makes the proposition that apparent sex differences are due to the masculinising of technology. This has implications for the understanding of sex or psychological differences in technophobia. This is reflected by a recent survey by the American Association of University Women which found that girls possess the ability to learn and use computers but do not want to be associated with the “geeky” image of technical careers by “guys” tapping on the keyboard all day long (The Herald Sun-Business, 2000). This masculinising or stereotyping of roles is depicted by Chivers (1987) as common in the physical sciences and the “older” craft and technology subjects. The introduction of Information Technology in schools is yet another dividing bias towards the masculine orientation. Rosen et al. (1987) state that computer anxiety is definitely implicated by gender role identity. The critical relevance in the causal factor influencing the apparent sex differences in technophobia is the concept of psychological gender. Bem’s research on psychological androgyny, in which masculinity and femininity are viewed: . as separate constructs; . as variable phenomena rather than fixed biological entities; and . that androgyny is associated with mental rather than physical well being, lends depth and support to the above argument. The perceived cultural norms of femininity and masculinity, not physical or biological attributes should be the factors to consider when examining the origins of technophobia. Brosnan (1998) highlights that in contexts where technology is not masculinised, there is no apparent variance between sex and psychological differences in aversion to technology. Some of the characteristics of masculinity (variables such as dominance, adventurousness, boldness and leadership) may be gleaned from Bem’s sex-role inventory scale (BSRI (Bem, 1974, 1981a)) although this scale has attracted its share of criticisms (Eagly et al., 1995; Baldwin et al., 1986; Lubinski et al., 1983; Taylor and Hall, 1982).

254

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

Bem (1981b) has subsequently developed her gender schema theory, which proposes that sex typing is a learned phenomenon mediated by cognitive processing. As such computing has developed a masculine image on par with the traditionally maculinised subjects such as mathematics, physics and engineering.

been suggested as an addition to Davis’ TAM components of usefulness and ease of use. Brosnan (1998) regressed the three major components and surmised that there is a relationship between sex differences, perceived usefulness and task focus concluding that the perceived usefulness of technology will facilitate the achievement and focus of a task. The literature has indicated that although there is an understanding that computers are meant to aid users to work more efficiently (Hirschheim and Newman, 1991; Winter, 1993), the expected benefits and increased productivity may not have been realised. The authors propose that the expected benefits of technology have not been universally realised owing to psychological segmentation, bringing about a disparity in relating to and communicating about the technology’s effective use. The mobile Internet technology (MIT) continuum here is used to depict this variation in consumer’s attitudes towards devices that carry very similar functionalities (Marakas, 1994; Marakas et al., 2000). The continuum (Figure 1) is anchored at two points: (1) At one end is a social setting whose attitude to computers is viewed as being “locally simplex” in nature, i.e. an

Technology psychological models Fishbein and Ajzen’s (1975) theory of reasoned action (TRA) refers to the attitude and intention to act in a certain manner by an individual. From this and the literature reviewed so far, the authors propose that, although there appears to be a correlation between the cognitive style and spatial ability of an individual and his/her reaction to environmental and social stimuli, personal experience and familiarity in what are deemed the so-called more masculine subjects such as mathematics, physics and computers could override the correlation to psychological gender. Cognitive style/spatial ability/ analytical strategies can therefore be developed and are not fixed biological entities. However, gender itself may undermine selfefficacy and hence the gaining of experience in the first place. These inter-related psychological factors affecting individuals’ attitudes towards technology explicate consumer decision making in the high-tech products of the marketing arena. Psychological factors among others will have a strong influence on high-tech purchase intentions (Viardot, 1998). This is supported by Davis’ (1993) TAM and Brosnan’s (1998) research of psychological factors influencing TAM. Davis’ model depicts the causal relationships between system design features, perceived usefulness, perceived ease of use, attitude toward using, and actual usage. Davis hypothesises that a prospective user’s attitude (in terms of perceived usefulness and perceived ease of use) towards a given system is a major determinant of whether he/she would actually use it. Helping the user perform his/her job better would appear to be key to potential usage. Davis recalls Bandura’s (1977a, b) self-efficacy theory and applies that in this context arguing that self-efficacy beliefs which are essentially value judgements, are similar to the perceptions of ease of use and perceived usefulness, influencing the decision making of the potential user. “Fun” has also

Figure 1 Authors’ marketing model – consumer decision making perspective continuum

255

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

individual will see the computer as controllable and alterable by humans, as a tool that enables challenging extensions into otherwise normally and easily attainable realms of accomplishments. (2) At the other end is a social setting in which computers are viewed as being “globally complex” in nature, i.e. an individual in this instance will see the computer as “incomprehensible” and external, as an autonomous entity. From the earlier literature, it is now possible to construct a hypothetical causal relationship between masculine gender appropriateness of computers and computer look-alike and feel products with the perceived cause of anxiety affecting consumer decision making. A further influence is proposed with psychological factors significantly correlating with consumer decision-making along the MIT continuum shown in Figure 1.

Methodology In the case of this study the product choice provides a neat dichotomous dependent variable; hence a “two-group discriminant analysis” was chosen based on the work of Hair et al. (1998). As a result of questionnaire size constraint as well as the existence of extensive research on technophobia, the authors made the decision to accept technophobia as a bona fide phenomenon. The MID in the consumer’s perspective would fall into two categories, a computercentric product and a phone-centric product. The product choice was selected as the dichotomous dependent variable and the psychographic factors represented the independent variables forming the psychological profile of a specific respondent, affecting the consumer decision-making. Other factors such as attitude to computer usage formed the intermediate variables. The operationalisation of the dependent and independent variables are detailed in Figure 2. Conjecturing the above model and utilising the literature, the authors addressed the following research questions that: . technophobia does not correlate significantly to just the non-computer user community; . feminine psychological gender will correlate significantly with technophobia;

Figure 2 Research model

.

.

.

technophobia brought about by gender inappropriateness is moderated by perceptions of self efficacy; computer experience overrides psychological gender in correlating significantly with technophobia; and the influence of psychological factors significantly correlates with the consumer decision making along the MIT product continuum.

The continuum for the purposes of this paper is limited to a dichotomous product choice in the forms of a mobile Internet product that looks and feels like a computer and a mobile Internet product that looks and feels like a phone. In designing the research questionnaire (see Appendix for an example questionnaire) the authors paid careful attention to the structure of the measurement scales, specifying the dependent variable to drive the two-group discriminant analysis. The psychographic independent variables were obtained by using two sets of scales and the demographic information and the product usage information providing the intermediate variables. The authors recognise that one of the main disadvantages of multivariate analysis is that the impact of measurement error and poor reliability cannot be directly seen because they are embedded in the observed variables. Therefore, rather than designing an original scale, standard scales with proven reliability and validity were employed. For example, as a basis for obtaining information from respondents on psychological gender and computer selfefficacy in consumer decision-making, statistical measurement would be in the form of multiple discriminant analysis utilising two scales, the “androgyny scales” and the “attitude to computer usage scales” (Popovich et al., 1987). To preserve reliability value and to encourage participation by

256

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

potential respondents, scales with large battery of items were also avoided. The measurement instrument was structured in five sections: (1) General demographics. (2) Androgyny scales. This section’s aim is to verify if sex role typing has an influence on various aspects of behaviour. This is an important independent variable to test. The conception of masculinity and femininity as two separate orthogonal dimensions rather than as two end points of a bipolar dimension is used to test the relation between sex role typing and consumer decision making. Bem’s sex-role inventory scales (BSRI (Bem, 1974) – 27 questions) was used as a guide to determine the moderation of psychological gender on the gender appropriateness behaviour of individuals in consumer decision making of technology products. Based on the BSRI, this study’s androgyny scale consisted of 25 items – nine in each of the masculine, feminine and neutral categories – measured as a five-point Likert scale of strongly agree to strongly disagree. Following Pedhazur and Tetenbaum’s (1979) critique of the BSRI and taking on board, Bem’s (1979) reply, items considered as not relevant to this study such as “aggressive” and “yielding” were excluded. (3) Attitude to computer scales. The attitudes to computer usage were adapted from Popovich et al. (1987) (20 questions). The scale for this study has 18 items, measured as a five-point Likert scales of strongly agree to strongly disagree. This was included as a moderating variable. (4) Computer, Internet and mobile consumption questions. The aim of this section is to obtain product consumption patterns. Categorical data would be collected in particular, a measure of the respondent’s computer experience; the main purpose of using computers; experience with the Internet; the main purpose of using the Internet; experience with mobile phones and the main purpose of using mobile phones. (5) MIT product choice questions. This last section seeks to obtain for each of the product types (Product A ¼ pocketcomputer; product B ¼ mobile phone),

the respondent’s opinion of that product type before he/she made his/her choice. The choice is for a product type and not for a particular brand. Product choice, between a MID that looks and feels like a computer or one that looks and feels like a phone forms a dichotomous dependent variable. A picture was provided for each one of the product types; the respondent was asked to describe the opinion of what the picture looked like, hence they formed an opinion of the product type before they made their choice of the product. The pictures of the MIDs are not a representation of a particular brand. The sample consisted of 161 individuals who were selected on the basis of a convenience sample with the filter of having a minimum of some computer literacy and being from the ABC1 (middle class) groups. The product choice distribution in this sample population was 82.33 (82 choosing computer-look-alike product A and 33 choosing phone-look-alike product B) with the rest categorised under missing data, choosing neither of the choices for the MID. The data collected were first tested for reliability using Cronbach’s alpha for the following (cases with missing data were not included): feminine psychological gender (nine items): Cronbach alpha of 0.6261 representing a just acceptable reliability rating. The masculine psychological gender (nine items): Cronbach alpha of 0.6883 representing an acceptable reliability rating. A positive attitude to computer usage scale (ten items): Cronbach alpha of 0.7451 representing a robust reliability rating. A negative attitude to computer usage scale (eight items): Cronbach alpha of 0.5699 representing a weaker than normal reliability rating. Three out of the four scales showed acceptable reliability (. 0.60) with the negative attitude to technology scale as just acceptable. The authors note that there were some negative attitude scoring within the inter-items characteristics and that these items might have affected the analysis of product choice prediction in the multiple discriminant analysis. However, given the overall reliability of the data, it was decided to execute a two-tailed Pearson correlation matrix across the items on the androgyny scale, all the items on the ATCUS scale as well

257

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

as product choice. Correlations between the “psych” items and “tech” items revealed an association between the psychological gender and technophobia variables, thus confirming a general tendency towards psychological gender and technophobia. The authors selected Wilks” Lambda (L) as the statistic of choice for weighing up the addition or removal of variables from the analysis. The F test provides the significance of the L when a variable is entered or removed. The variable with the largest F (F to enter) is included. This process is repeated until there are no further variables with an F value greater than the critical minimum threshold value. At the same time, any variable which had been added earlier but which no longer contributes to maximizing the assignment of cases to the correct groups because other variables in concert have taken over its role, is removed when its F (F to remove) value drops below the critical maximum threshold value. These critical values are listed in the analysis conducted below. An initial multiple discriminant analysis (MDA) was executed with the product choice as the dichotomous dependent variable and the “psych” and the “tech” scales as the independent variables. To eliminate any preconceived perceptions of prices affecting the choice a free coupon was offered. Information about the data and the number of cases in each category of the grouping variable is shown in the Tables I-IV. The number of cases for each independent variable within each level of the grouping has been edited, since in this analysis every independent variable within a particular level has the same number of cases. The edited group statistics are shown in Table II. The number of cases in each of the groups is not equal; hence the authors have chosen to run the analysis using the discriminant Table I Multiple discriminant analysis, case processing summary Unweighted cases Valid Excluded

Total Total

Missing or out-of-range group codes At least one missing discriminating variable Both missing or out-of-range group codes and at least one missing discriminating variable

n

Per cent

115 8 33

71.0 4.9 20.4

6 47 162

3.7 29.0 100.0

Table II MDA, edited group statistics table Choice if free

Valid n

Product A Product B Total

82 33 115

Table III Multiple discriminant analysis, prior probabilities for groups Choice if free

Prior

Product A Product B Total

0.713 0.287 1.000

Cases used in analysis Unweighted Weighted 82 33 115

82.000 33.000 115.000

Table IV Multiple discriminant, classification results

Choice if free Original count

Per cent

Product A Product B Ungrouped cases Product A Product B Ungrouped cases

Predicted group membership Product A Product B 105 30 10 96.3 76.9 71.4

analysis classifications, prior probabilities to be computed from group sizes. A number of uses were made of MDA. Finally the real value of MDA surfaces, with the classification are shown in Tables III and IV.

Results and discussion It was found that over 70 per cent of original grouped cases were correctly classified. This high percentage is very encouraging and indicates the research’s relevance for praxis marketing applications. The MDA showed a clear success in discriminating between the two groups into profiles choosing Product A or Product B. It supports the authors’ proposition that consumer perceptions of MIT products are dichotomous and that technology phobia correlates with demographic and psychological contexts. As such, the chosen methodology and the instrument utilising discriminant analysis lend themselves particularly to predicting consumers’ MIT product preferences. The robust reliability of the study’s androgyny scales and attitude to computer scales

258

4 9 4 3.7 23.1 28.6

Total 109 39 14 100.0 100.0 100.0

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

augments the literature’s reference to sex and psychological differences in the stereotyping of technology adoption. Moreover, the results of the addition or removal of an independent variable in the stepwise discriminant analysis were monitored by a statistical test and the conclusion was used as a basis for the inclusion of that independent variable in the final analysis. There being only two groups in this analysis, there was only one discriminant function. It was noted that no variables appeared to have been removed in the stepwise discriminant analysis. The value for F to remove was 2.71 and for F to enter was 3.84 but none of the variables were less than the criteria set. There is therefore a clear indication that technophobia not only correlates with the non-computer users community, but there is also a distinct association between the consumer’s situated psychological gender and his/her product preference. This is further reflected by the study’s masculine scale items dominant in choosing the computer look alike Product A in which characteristics such as: “has leadership qualities”, “masculine self-labelling”, and “makes decisions easily” yielded significance levels of 0.047, 0.045 and 0.030 respectively. Furthermore, the technophobia brought about by gender inappropriateness may be moderated by perceptions of self-efficacy and that previous exposure can mediate one’s psychological fear of new technology. Univariate ANOVA was used to indicate whether there is a statistically significant difference among the grouping variable means for each independent variable. The study’s overall findings support Brosnan’s (1998) proposition that technology anxiety correlates with demographic variables such as age, gender and academic qualifications. Indeed, the significant relationship ( p , 0.05) between feminine psychological gender and technophobia illustrates the general agreement (Brosnan, 1998; Campbell and McCabe, 1984) that females exhibit higher levels of computer anxiety and computer negativity than males. However, it is worth noting that individuals such as females exhibiting androgynous characteristics moderate this sex difference in attitude to technology.

Managerial implications The literature review has focused on “technophobia” and existing research on “psychological gender” and “cognitive theory” in developing the construct of this paper. From the extensive existing research and the findings of this study, the authors conclude that technophobia does not correlate only to the non-computer user community, but rather, the main factors contributing to technophobia are found to be anxiety and negative attitude as interceded by the users’ psychological gender and previous success. It has as its basis the study findings of Rubin and Greene (1991) that sex differences can be more psychological rather than biological and that psychological role identity implicates technology adoption as well as consumer decision making. Females exhibiting male androgynous qualities would appear to have a more positive attitude towards computers in general. From the literature, it is clear that technophobia emerges largely from the masculinisation of technology (Brosnan, 1998). Therefore, the importance of appreciating the differences between the psychological male and the psychological female cannot be overlooked in MIT “hardware” and “software” product design and promotion. It could be that MID products will penetrate a wider consumer base if marketed as an incremental innovation to mobile phones with the convenience of access to useful Internet source information. Furthermore, it would appear that there exist different levels of technophobia from being “uncomfortable users” to “cognitive computerphobes” to “anxious computerphobes” (Rosen et al., 1993) albeit that not all technophobes seek to avoid their source of anxiety. Hence, marketers are well advised to be willing to endorse several product designs that will meet consumers’ psychosocial needs along the MIT consumer perspective continuum (MITCPC). The authors present an exploratory approach by applying Bem’s self-perception theory of radical behaviourism and selfefficacy. The implications for companies extend beyond the recognition that when a consumer buys a MID, be it a pocket computer or a WAP phone, the decision is a result of perceived strong environmental causations as well as the individual’s internalised experiences and attitudes. Since computer anxiety and technophobia is a

259

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

transitory condition and therefore may fluctuate with time and treatment (Campbell, 1988), marketers should not ignore the implications of social learning, self-efficacy and cognitive style and spatial ability when seeking to understand the psychological attitude and behaviour of their target customers. Additionally as with the motor car industry in which some cars are deliberately designed to interest the male or female gender, perhaps it is more cost-beneficial to segment the MIT products to target and appeal to the two consumer perspectives. For example, introducing the possibility of “upgrading” along the product simplexcomplex continuum may be an option to keep abreast of the morphing needs of users. After all, the literature has revealed that experience with technology such as computers, provides an antecedent to reducing the correlation between feminine gender and negative attitude to technology. This ability to design economically feasible myriad levels of relational marketing services to complement the different gender demands along the MITCPC will be a distinct competitive edge. This study has confirmed that both human and psychological factors are an essential component when purchasing new products. For the marketer involved in MID, it is necessary to understand the inter-relatedness of the psychological factors affecting individual attitudes and the consumers’ ultimate decision to buy high-tech products. Therefore, the causal relationships between system design features, perceived usefulness, perceived ease of use and availability (e.g. size and shape of the device, access to the Internet, graphics quality and speed of access, etc.) should rest high on the agenda of mobile phones innovation. However, given the cost and pace of technological developments, corporate executives are faced with having to collaborate to compete by entering into alliances to facilitate the integration of MID with electronic businesses. As such, the enforced vertical and often horizontal cooperation within the MID industry is fast becoming the norm and brings with it “new” and additional risks in terms of profit/loss sharing, intellectual property and human capital issues. This brings with it the importance of designing acceptable and economically feasible new technology products. The outcome for this study is the confirmation that psychological factors

implicate consumer decision-making along the MIT product continuum. However, the continuum for the purpose of this paper is limited to a dichotomous product choice in the forms of a mobile-Internet product that looks and feels like a computer and a mobileInternet product that looks and feels like a phone. An extension of the product range to include other digital personal devices will allow wider confirmation of the factors related to this paper which influence decisionmaking. Additional studies of technophile and psychological gender will yield further insights on how to harness and utilise the dichotomous MIT product continuum for greater competitive advantage. Therefore, mapping psychosocial implications to attribution process theory, as applied to consumerdecision-making, technology acceptance, design preferences, may provide a fruitful area of possible exploration.

References Baldwin, A.C., Stevens, L.C., Critelli, J.W. and Russell, S. (1986), “Androgyny and sex role measurement: a personal construct approach”, Journal of Personality and Social Psychology, Vol. 51 No. 5, pp. 1081-8. Bandura, A. (1977a), “Self efficacy: toward a unifying theory of behavioural change”, Psychological Review, Vol. 84, pp. 191-215. Bandura, A. (1977b), Social Learning Theory, Prentice-Hall, Englewood Cliffs, NJ. Bem, S.L. (1974), “The measurement of psychological androgyny”, Journal of Consulting and Clinical Psychology, Vol. 42, pp. 155-62. Bem, S.L. (1977), “On the utility of alternative procedures for assessing psychological androgyny”, Journal of Consulting and Clinical Psychology, Vol. 45, pp. 196-205. Bem, S.L. (1979), “Theory and measurement of androgyny: a reply to the Pedhazur-Tetenbaum and LockS LeyColen critiques”, Journal of Personality and Social Psychology, Vol. 37, pp. 1047-54. Bem, S.L. (1981a), Bem Sex Role Inventory Professional Manual, Consulting Psychologists Press, Palo Alto, CA. Bem, S.L. (1981b), “Gender schema theory: a cognitive account of sex typing”, Psychological Review, Vol. 88, pp. 354-64. Brosnan, M. (1998), Technophobia, The Psychological Impact of Information Technology, Routledge, New York, NY. Brosnan, M. and Davidson, M. (1996), “Psychological gender issues in computing”, Journal of Gender, Work and Organization, Vol. 3 No. 1, pp. 13-25. Campbell, N.J. (1988), “Correlates of computer anxiety of adolescent students”, Journal of Adolescent Research, Vol. 3 No. 1, pp. 107-17.

260

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

Campbell, P. and McCabe, G. (1984), “Predicting the success of freshman in a computer science major”, Communications of the ACM, Vol. 27, pp. 1108-13. Chivers, G. (1987), “Information technology – girls and education: a cross-cultural review”, in Davidson, M.J. and Cooper, C.L. (Eds), Women and Information Technology, John Wiley, London, pp. 13-32. Davis, F.D. (1986), “A technology acceptance model for empirically testing new end-user information systems: theory and results”, doctoral dissertation, Sloan School of Management, MIT, Boston, MA. Davis, F.D. (1989), “Perceived usefulness, perceived ease of use and user acceptance of information technology”, MIS Quarterly, Vol. 13, pp. 318-39. Davis, F.D. (1993), “User acceptance of information technology: system characteristics, user perceptions and behavioral impacts”, International Journal of Man-Machine Studies, Vol. 38 No. 3, pp. 475-87. Dukes, R.L., Discenza, R. and Couger, J.D. (1989), “Convergent validity of four computer anxiety scales”, Educational and Psychological Measurement, Vol. 49, pp. 195-203. Eagly, A.H., Karau, S.J. and Makhijani, M.G. (1995), “Gender and the effectiveness of leaders: a metaanalysis”, Psychological Bulletin, Vol. 11 No. 7, pp. 125-45. Erickson, T.E. (1987), “Sex differences in student attitudes toward computers”, PhD dissertation, University of California, Berkeley, CA. Farina, F., Arce, R., Sobral, J. and Carames, R. (1991), “Predictors of anxiety towards computers”, Computers in Human Behaviour, Vol. 7 No. 4, pp. 263-7. Fishbein, M. and Ajzen, I. (1975), “A Bayesian analysis of attribution processes”, Psychological Bulletin, Vol. 4, pp. 195-277. Hair, J.F. Jr, Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis, 5th ed., PrenticeHall, Englewood Cliffs, NJ. Heinssen, R.K. Jr, Glass, C.R. and Knight, L.A. (1987), “Assessing computer anxiety: development and validation of the computer anxiety rating scales”, Computers in Human Behaviour, Vol. 3, pp. 49-59. (The) Herald Sun-Business (2000), “Girls leaving high-tech jobs to the boys”, The Herald Sun-Business, 4 July. Hirschheim, R. and Newman, M. (1991), “Symbolism and information systems development: myth, metaphor and magic”, Information Systems Research, Vol. 2, pp. 29-62. Igbaria, M. and Chakrabarti, A. (1990), “Computer anxiety and attitudes towards microcomputer use”, Behaviour & Information Technology, Vol. 9 No. 3, pp. 229-41. Jay, T. (1981), “Computerphobia – what to do about it?”, Educational Technology, Vol. 21, pp. 47-8. Kelley, H.H. (1967), “Attribution theory in social psychology”, in Levine, D. (Ed.), Nebraska Symposium on Motivation, University of Nebraska Press, Lincoln, NB. Lee, R.S. (1970), “Social attitudes and computer revolution”, Public Opinion Quarterly, Vol. 34, pp. 53-9. Loyd, B.H. and Gressard, L.P. (1984), “The effects of sex, age and computer experience on computer attitudes”, AEDS Journal, Vol. 18, pp. 67-77.

Lubinski, D., Tellegan, A. and Butcher, J.N. (1983), “Masculinity, femininity and androgyny, viewed and assessed as distinct concepts”, Journal of Personality and Social Psychology, Vol. 44, pp. 428-39. Marakas, G.M. (1994), “Anthropomorphic behavior and information technology: when the metaphor becomes the model”, Academy of Management Conference, August 15-17, Dallas, TX. Marakas, G.M., Johnson, R.D.. and Palmer, J. (2000), “A theoretical model of differential social attributions toward computing technology: when the metaphor becomes the model”, International Journal of Human-Computer Studies, Vol. 52 No. 4, pp. 719-50. Maurer, M.M. (1983), “Development and measurement of a measure of computer anxiety”, unpublished Masters thesis, Iowa State University, Ames, IA. Mizerski, R.W., Golden, L.L. and Kernan, J.B. (1979), “The attribution process in consumer decision making”, Journal of Consumer Research, Vol. 6, September, pp. 123-40. Moldafsky, N.I. and Kwon, I.W. (1994), “Attributes affecting computer aided decision making: a literature survey”, Computers in Human Behaviour, Vol. 10 No. 3, pp. 299-323. Motorola (1996), The British and Technology: Prepared for the Future? The Motorola Report, Information Society Initiative, London. Okebukola, P. and Woda, A. (1993), “The gender factor in computer anxiety and interest among some Australian high school students”, Educational Research, Vol. 35 No. 2, pp. 181-9. Pedhazur, E.J. and Tetenbaum, T.J. (1979), “Bem sex-role inventory: a theoretical and methodological critique”, Journal of Personality and Social Psychology, Vol. 37, pp. 996-1016. Popovich, P.M., Hyde, K.R., Zakrajsek, T. and Blumer, C. (1987), “The development of the attitudes toward computer usage scales”, Educational and Psychological Measurement, Vol. 47, pp. 261-9. Raub, A. (1981), “Correlates of computer anxiety in college students”, University of Pennsylvania Dissertation Abstracts International, Vol. 42 No. 4775A, doctoral dissertation. Rosen, L.D. and Maguire, P. (1990), “Myths and realities of computerphobia: a meta-analysis”, Anxiety Research, Vol. 3, pp. 175-91. Rosen, L.D., Sears, D.C. and Weil, M.M. (1987), “Computerphobia behaviour”, Research Methods, Instruments and Computers, Vol. 19, pp. 167-79. Rosen, L.D., Sears, D.C. and Weil, M.M. (1993), “Treating technophobia: a longitudinal evaluation of the computerphobia reduction program”, Computers in Human Behaviour, Vol. 9, pp. 27-50. Rubin, D. and Greene, K. (1991), “Effects of biological and psychological gender, age cohort, and interviewer gender on attitudes toward gender inclusive language”, Sex Roles, Vol. 24 No. 7-8, pp. 391-412. Taylor, M.C. and Hall, J.A. (1982), “Psychological androgyny: a review and reformulation of theories, methods and conclusions”, Psychological Bulletin, Vol. 92, pp. 347-66. Turkle, S. (1984), The Second Self: Computers and Human Spirit, Simon & Schuster, New York, NY. Turkle, S. (1988), “Computational reticence: why women fear the intimate machine”, in Kramarae, C. (Ed.), Technology and Women’s Voices, Routledge, London.

261

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

Turkle, S. and Papert, S. (1990), “Epistemological pluralism: styles and voices within the computer culture”, Signs, Autumn, pp. 128-57. Viardot, E. (1998), Successful Marketing Strategy for Hightech Firms, Artech House, Boston, MA. Winter, S. (1993), “The symbolic potential of computer technology: differences among white collar workers”, in DeGrosss, J.I., Bostrom, R.P. and Robey, D. (Eds), Proceedings of the 14th International Conference on Information Systems, Orlando, Florida, 5-8, December, pp. 331-44. Witkin, H.A., Moore, C.A. and Raskin, E. (1977), “Role of the field-dependent and field-independent cognitive styles in academic revolution: a longitudinal study”, Journal of Educational Psychology, No. 69, pp. 197-211.

(5) Using a computer is too time consuming. (6) I feel that having my own computer would help me with my job. (7) I prefer not to learn how to use a computer. (8) I would like to own or do own a computer. (9) I like to play video games. (10) I feel that the use of computers in schools will help children to learn mathematics. (11) I prefer to use an automatic teller for most of my banking. (12) If I had children, I would not buy them computerised toys. (13) I have had bad experiences with computers. (14) I would prefer to order items in a store through a computer than wait for a store clerk. (15) I feel that the use of computers in schools will negatively affect children’s reading and writing abilities. (16) I do not like using computers because I cannot see how the work is being done. (17) I do not feel I have control over what I do when I use a computer. (18) I do not like to program computerized items such as VCRs and microwave ovens.

Further reading Bandura, A. (1997), Self-efficacy: The Exercise of Control, WH Freeman and Company, New York, NY. Bem, S.L. (1975), “Sex-role adaptability: one consequence of psychological androgyny”, Journal of Personality and Social Psychology, Vol. 33, pp. 634-43. Bernard, L.C. (1980), “Multivariate analysis of new sex role formulations and personality”, Journal of Personality and Social Psychology, Vol. 38, pp. 323-36. Eagly, A.H. and Carli, L.L. (1981), “Sex of researchers and sex-typed communications as determinants of sex differences in influenceability: a meta-analysis of social influence studies”, Psychological Bulletin, Vol. 90, pp. 1-20. Kelley, H.H. (1971), Attribution in Social Interaction, General Learning Press, Morristown, NJ. Lehman Brothers (2000), available at: http:// newscnetcom/news/0-1004-200-1611107html (accessed 5 June 2000). Ovum (2000), Mobile E-commerce: Market Strategies, Ovum, Burlington, MA. Spence, J.T., Helmreich, R. and Stapp, J. (1975), “Ratings of self and peers on sex-role attributes and their relation to self-esteem and conceptions of masculinity and femininity”, Journal of Personality and Social Psychology, Vol. 32, pp. 29-39.

Appendix. Example of questionnaire The list of characteristics used from ATCUS was scored on a five-point Likert scales of strongly agree to strongly disagree. Listed below are the 18-item scales used in this paper: (1) I would prefer to type on a word processor than on a typewriter. (2) Whenever I use something that is computerised, I am afraid I will break it. (3) I like to keep up with technological advances. (4) I know that I will not understand how to use computers.

The instrument included the following questions to obtain product consumption patterns: (1) Experience in computers: . Less than six months . Six months to less than one year. . One year to less than 1.5 years. . Over 1.5 years. (2) Main purpose of using computers: . For business. . For leisure. . For emergency. . Other. (3) Experience with Internet: . Less than six months. . Six months to less than one year. . One year to less than 1.5 years. . Over 1.5 years. (4) Main purpose of using Internet: . For business. . For school. . For leisure. . Other. (5) Experience with mobile phones: . Less than six months. . Six months to less than one year. . One year to less than 1.5 years. . Over 1.5 years.

262

Technophobia, gender influences and consumer decision-making

European Journal of Innovation Management

David Gilbert, Liz Lee-Kelley and Maya Barton

Volume 6 · Number 4 · 2003 · 253-263

(6) Main purpose of using mobile phones: . For business. . For leisure. . For emergency. . Other.

[ [ [ [

Product A

[

Please place a cross mark against the statement that best suits your opinion of what the Plate 1 looks like: [ ] Looks like a computer. [ ] Looks more like a computer than a phone. [ ] Looks both like a computer and a phone. [ ] Looks more like a phone than a computer. [ ] Looks like a phone. Product B

] Looks like a phone. ] Looks more like a phone than a computer. ] Looks both like a phone and a computer. ] Looks more like a computer than a phone. ] Looks like a computer.

Product preference Please place a cross mark against the options. In your opinion, which product appears easier to use. [ ] Product A. [ ] Product B. [ ] Neither. If you were offered a free coupon to buy either one of the above mobile Internet devices, which one would you choose? [ ] Product A. [ ] Product B. [ ] Neither. Androgyny scale items list utilised. The respondents were asked what best describes them, and the characteristics were scored on five-point Likert scales. The items were as listed in Table AI.

Table AI Androgyny scale items list

Please place a cross mark against the statement that best suits your opinion of what the Plate 2 looks like.

M items

F items

Neutral items

Stong personality Forceful Has leadership qualities Self-sufficient Dominant Independent Masculine Acts as a leader Makes decisions easily

Soft spoken Warm Tender Gentle Compassionate Affectionate Cheerful Feminine Loves children

Adaptable Tactful Moody Secretive Sincere Theatrical Conscientious Conventional Generous

263