No title

Introduction

Determinants of innovation in small food firms

Over the past decades, the Belgian food and drink industry (hereafter the ``food industry'') has faced dramatic changes in its competitive environment. As in some other countries, most notably the UK, internationalisation of the food market and mergers among supermarkets have put pressure on the prices which can be realised by the suppliers of both fresh and processed food. In addition, the Belgian government has increased minimum standards with respect to food safety. At the beginning of the twenty-first century, Belgium has one of the strictest food safety policies in the world. As a result of these pressures, food manufacturing, which is generally viewed as a mature and relatively low technology industry, has been forced to introduce changes that have affected all aspects of operation. On the one hand, food companies have needed to keep up with new regulatory standards. In this context, control and prevention of food contamination play a central role. On the other hand, increased competition has pushed food companies to become more efficient in processing, to reorganise management, develop new products and explore new markets in order to meet the needs and wants of consumers competitively. The aim of this research was to explore some of the influences on innovation amongst small Belgian food manufacturers as they react to these pressures. Three determinants of innovation were tested using data collected through a survey: the age of the company, the size of the company, and the economic performance of the region in which the firm is located. Several theories have been developed in these areas but both the theories themselves and the results of empirical testing are often contradictory and inconclusive (Grunert et al., 1997). One of the main reasons for the apparent inconsistencies in the economic literature is that the theories advanced and the empirical tests often seek to establish general patterns, whereas in reality firms tend to follow industry-specific patterns of innovation (Malerba and Orsenigo, 1995; Tether, 2002).

Tessa Avermaete Jacques Viaene Eleanor J. Morgan and Nick Crawford

The authors Tessa Avermaete is a Researcher and Jacques Viaene is a Professor, both in the Department of Agricultural Economics, University of Ghent, Belgium. Eleanor J. Morgan is a Doctor and Nick Crawford is a Researcher, both in the School of Management, University of Bath, Bath, UK. Keywords Innovation, Small firms, Food industry, Drinks industry, Belgium Abstract This paper focuses on innovation in small food- and drinkmanufacturing enterprises in two Belgian regions. The aim of the research was to identify patterns of innovative activities. Based on both literature and empirical analyses, a framework was developed to help in understanding some aspects of the determinants of innovation in small food firms. Two conclusions can be drawn from the research. On the one hand, it is clear that innovation was regarded as essential by most small food firms. Such firms tended to continuously introduce new products, develop new processes, make changes in the organisational structure and explore new markets. On the other hand, the research demonstrated that some aspects of innovation depend on the age of the company, company size and regional economic performance. 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

This research was undertaken with support from the European Union's fifth framework programme: project Innovaloc No. HPSE-CT-1999-00024. The content of the paper is the responsibility of the authors.

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . pp. 8-17 # MCB UP Limited . ISSN 1460-1060 DOI 10.1108/14601060310459163

8

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

The research was based on the study of a particular industry. Three main reasons motivated the decision to investigate small Belgian firms within the food industry. First, the food industry represents one of the most important industrial sectors in the Belgian economy and is characterised by a dual structure in which small and medium-sized enterprises dominate rather than large enterprises and multinationals (Traill, 1997). Second, it is argued that small firms are essential to economic development as an integral part of all market economies (Taylor, 2001). There is evidence that small food firms are particularly important in rural areas where they have developed to process local agricultural products (Traill, 1995). Small food firms are therefore thought of as a means of achieving sustainable economic growth in local economies, especially lagging regions which tend to have important agricultural sectors (McDonagh and Commins, 1999). Third, it is suggested that rurally located firms produce specialised regional products of a different nature from those produced by large firms. Large firms generally have a national or European market orientation and consequently tend to focus on products with more of a mass appeal. In this sense, such small firms can make an important contribution to Europe's highly valued cultural diversity (Traill, 1995). The paper is structured as follows. Section two discusses the conceptual framework, reviewing the literature on the concept of innovation and its determinants. Section three deals with the methodology. The results of the survey are presented in section four. The last section formulates the conclusions, discusses the managerial implications and suggests areas for further research.

of innovation have been developed and applied in the economic literature (Cumming, 1998; Grunert et al., 1997; Johannessen et al., 2001). Most researchers have focused on technology-related innovations, such as the introduction of products that require radical changes in the production process. The concept of innovation, however, can be seen as extending far beyond radical and technology-based product innovation. Innovation may also be taken to cover incremental changes in products and processes as well as changes in the organisational structure and moves to exploit new markets. This idea is reflected in Lundvall's (1992) definition of innovation as: an ongoing process of leaving, searching, and exploring which results in: (1) new products; (2) new techniques; (3) new forms of organisation; and (4) new markets.

This approach is visualised in Figure 1 where the four domains of innovation are shown. Innovation is often the result of simultaneous changes in different domains, as discussed below, and the arrows between the boxes in the diagram indicate the scope for such interaction. Product innovation can be seen as any good, service or idea that is perceived by someone as new (Kotler, 1991; Grunert et al., 1997). Therefore, a product may be considered an innovation to one person or organisation but not to another (Johannessen et al., 2001). Product innovation may result from changes in the organisational structure of the company. For example, this is the case when food quality is improved through a more efficient organisation of the firm's safety controls. Further, new products may arise through the exploitation of new market segments. Over the last decades, several new market segments have been introduced by the food industry, ranging from organic and nutritional foods to ready-made meals. However, product innovation is mostly associated with changes in processing. Process innovation includes the adaptation of existing production lines as well as the installation of an entirely new infrastructure and the implementation of new technologies. In general, process innovation allows the creation of new products. But process innovation may also be required as part of a reorganisation of the company or to enable the exploitation of new markets. An example

Conceptual framework The concept of innovation Although innovation has been studied extensively, there is no generally accepted way of measuring innovation. Some research is based on published R&D expenditures and patent data (Breschi, 1999; Malerba and Orsenigo, 1995), while other research relies on measurements derived from detailed surveys among companies (Diederen et al., 2000). Innovation itself is a very broad concept and, as a result, various classifications 9

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

Figure 1 Domains of innovation

exploitation of new territorial markets and the penetration of new market segments within existing markets. The shift from conventional to organic production illustrates that market innovation in the food industry is strongly interwoven with product innovation and organisational innovation and, to a lesser extent, with process innovation.

can be found in Grunert and Ottowitz (1997) who describe the conversion of traditional brewing to ecological brewing in a small German firm. While the conversion was foremost a market innovation, both organisational and process innovations were required to successfully make the shift to ecological brewing. Organisational innovation deals with changes in marketing, purchases and sales, administration, management and staff policy (Clarysse et al., 1998). Although studies on organisational innovation are limited, organisational innovation has gained importance in all industrial sectors. One can, for example, think of the success of the ISO standard, which prescribes rules in order to make processes transparent, documented, reproducible and controlled (Varzakas and Jukes, 1997). Although the food industry initially lagged behind in terms of ISO certification, increased competition and the growing power of retailers have forced food firms along the entire chain to re-organise quality systems and implement international standards (Boudouropoulos and Arvanitoyannis, 2000; Hoogland et al., 1998). Moreover, Maurer and Drescher (1996) show that the implementation of these standards in food firms may potentially result in innovation in products and processes and lead to a competitive advantage. The last domain of innovation concerns market innovation which is defined as the

Determinants of innovation Various determinants of innovation have been identified ranging from micro-economic characteristics and inter-firm linkages to macro-economic performance (Camagni, 1991; Cooke et al., 1997; Fanfani and Lagnevik, 1995; Nooteboom, 1999; Noronha Vaz and Nicolas, 2000). In this paper, the impact of three variables is analysed: (1) year of establishment of the company; (2) company size; and (3) regional economic performance. Literature on the relationship between the age of the company and innovation is rather limited. Studies in this domain go back to Shumpeter (1934), who is considered to be the founding father of the theory of innovation dynamics (Malerba and Orsenigo, 1995). In his first work, The Theory of Economic Development, Shumpeter (1934) examined the European industrial structure in the late nineteenth century that was, at the time, dominated by small firms. Shumpeter (1934) found that entry tended to be easy for 10

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

expenditures and employment in R&D as indicators of innovation (European Commission, 1999; Roper, 2000). Besides economic performance, the institutional, technological and political environment also plays a part in determining whether a region is a potentially innovative milieu (Camagni, 1991). From this perspective, industrial policy and food legislation, public research centres, universities, industry associations and membership of other types of networks all contribute to the innovative behaviour of firms (Antonelli and Calderini, 1999; Breschi, 1999).

firms with new technology to exploit and emphasised the role of new firms as drivers for innovation. New entrepreneurs start with new ideas, new products and new processes. Current ways of production, organisation and distribution are disrupted and quasi-rents, associated with previous innovations, are wiped out. This dynamic is referred to as creative destruction, or the Shumpeter (1934) Mark I pattern of innovation. The study of the relationship between firm size and innovation also goes back to Shumpeter. In his second work, Capitalism, Socialism and Democracy, Shumpeter (1942) claimed that large firms are more likely to innovate than small firms. With their accumulated stock of knowledge in specific technological areas, their advanced competence in large-scale R&D projects, production and distribution and their access to resources, large firms create barriers to entry for new entrepreneurs (te Velde, 2001). This is the Shumpeter Mark II pattern of innovation. In the footsteps of Shumpeter, the relationship between company size and innovation has been extensively studied (Antonelli and Calderini, 1999; Breschi, 1999; Le Bars et al., 1998; Malerba and Orsenigo, 1995). However, more than half a century after Shumpeter's work, the debate on the relationship between company size and innovation is still ongoing. Empirical studies have reached apparently contradictory conclusions. These are mainly due to the different measurements of innovation used (Grunert et al., 1997; Le Bars et al., 1998) but also to different sampling methods, with many studies taking data across industries to try to reach generalised conclusions rather than looking at industry-specific patterns of innovation. In addition, the size distributions of firms included within samples differ and it is worth noting that the smallest firms are often excluded from the analyses. Although it is clear that internal firm characteristics like size and length of establishment have an impact on the firm's innovative behaviour, scientists and policy makers are increasingly focusing on the environment in which innovation occurs. Data on European regions suggest a positive relationship between regional economic performance ± measured in terms of GDP per head ± and innovation ± measured by patent data. Similar results are found using R&D

Methodology The evaluation of innovative activities was based on primary data collected through a survey among small food firms in two Belgian provinces: Hainaut in the South and West Flanders in the North. The regions were chosen as each has a significant foodmanufacturing sector but they differ significantly in terms of economic performance with West Flanders being the more prosperous of the two. In 2001, official statistics show that the GRP per head in West Flanders was 22,000 euro and the unemployment rate was 3.3 percent. In contrast, the GRP per head in Hainaut was only 15,000 euro and there was an unemployment rate of 13.1 percent. The target population included small food and drink firms with between three and 50 employees[1]. These were identified with reference to the National Bank of Belgium and the Belgian Food Federation (FEVIA). Quota sampling was then used to select 60 enterprises to approach with the aim of studying 30 firms in each region. The final respondents included 55 small food firms, 26 based in Hainaut and 29 in West Flanders. The structure of the final sample reflected the importance of the sub-sectors in the provinces. In Hainaut, 30 percent of the sample were drink-processing enterprises. In West Flanders, 38 percent of the sample were meat processors. The firms had been established for 50 years on average and had a mean company size of 21 employees. Thirteen of them were micro enterprises, having up to ten employees. A pilot survey was carried out in May 2001. The final survey was carried out between July 11

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

and December 2001. It was based on indepth, face-to-face interviews with the top manager or owner of the firm. Each interview usually lasted between one hour and one and a half hours. The questionnaire included five indicators of innovation. The first indicator was product innovation. Entrepreneurs were asked whether they had introduced a product innovation (defined as a new or substantially modified product) over the last five years. The second measure deals with process innovation in terms of whether a new or substantially modified process had been introduced over the last five years. The third indicator was the certification of Hazard Analysis and Critical Control Point (HACCP). HACCP is internationally recognised as an effective method to guarantee safe food (Unnevehr and Jensen, 1999). In order to receive the HACCP certificate, a company has to install a strict mechanism to control and prevent food contamination. It is widely agreed that HACCP certification of small food firms requires incremental and/or radical changes in technologies, infrastructure and organisation (Taylor, 2001). The fourth indicator was ISO 9000 certification. Implementation of ISO 9000 in small food firms implies a radical change in the organisation and puts conditions on the technologies and materials used in the processing (Varzakas and Jukes, 1997). The fifth and last indicator was whether the firm participated in the organic food chain, which is principally a market innovation. All these five indicators were treated as dummies in the analysis and coded zero or one. Besides enquiring about the presence and absence of innovation, the questionnaire was designed to collect data to allow the extent of innovativeness to be assessed. Four indicators were used to measure this. These were first, the percentage of sales associated with products introduced in the last five years; second, the number of patents held; third, the share of turnover spent on R&D activities and fourth, the frequency of introducing products that were new to the market segment in which they competed. In order to identify the determinants of innovation in small food firms, independent sample t-tests and Chi-square tests were assessed. The impact of company size, company age and regional economic

performance on innovation patterns in small food firms was analysed using SPSS.

Results The empirical analysis of the sample comprises four parts. First, the general findings of the survey are summarised. Second, the results of the analysis of the relationship between the age of the company and innovativeness are discussed. The effect on innovation of the size of the firm within the lower end of the size distribution being considered here is then examined with a distinction being made between micro and small enterprises. The last analysis deals with the role of regional economic performance for innovation by comparing the innovativeness of the small food firms in the two provinces studied. General findings The results of the general analysis support the importance of innovation for small food firms. Out of 55, 48 respondents claimed to have introduced at least one of the five types of innovation discussed earlier. This finding is in line with an earlier study carried out among 2,783 companies for the French agro-food industry, where 70 percent of the respondents declared they had achieved an innovation during the past three years (Le Bars et al., 1998). Figure 2 shows the results of the survey, split up by the indicators for innovation. Over the past five years, 38 respondents introduced a product innovation and the same number had made a process innovation. These were mainly concerned with small changes in packaging, extension of existing production lines and mechanisation of the production process. Almost one fourth of the respondents achieved HACCP certification, whereas only ten implemented the ISO 9000 standard. Participation of food manufacturers in the organic chain is only a recent tendency. Nevertheless, six of the respondents had already obtained the Belgian organic logo. The analysis also supports the argument that was made earlier on the interrelationships between the different domains of innovation. Out of the 48 innovative firms, 37 firms had introduced more than one type of innovation and five firms had introduced four types of innovation over the last five years. However, a significant relationship could only be found between product 12

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

Figure 2 General outcomes for innovation in small food firms (N = 55)

innovation and process innovation (2 = 4.566; p = 0.033). Turning to the results relating to the degree of innovativeness, three results are apparent. First, patenting is not very common in small food firms, as only five of the respondents had a patent. Second, the frequency of market innovation is sector-related. In primary processing companies, like slaughtering and deepfrozen vegetables, there was relatively little product innovation. This is in contrast with the beverage sector: new alcoholic and non-alcoholic drinks were continuously being developed. Third, in 24 firms, innovations introduced in the previous five years accounted for more than 5 percent of the turnover. A small number of respondents had been very successful in their product innovation; in 11 cases, product innovation over the past five years represented over 20 percent of the firm's annual turnover.

group included 27 firms established before 1970; the second group included 28 companies established after 1970. Chi-square testing indicates that the age of the company was not significantly related either to patenting behaviour or to the share of turnover spent on R&D activities. Table I presents the cross-tabulation for the frequency of introducing products that were also new to the market segment in which the firm competes and for the impact of innovation on turnover, on the one hand, and the company age, on the other hand. Two conclusions emerge from the Table. First, older companies were more likely to introduce products that are new to the market segment in which they compete (2 = 6.058; p = 0.048). Eight of the 27 companies established before 1970 often introduced products that were also new to the market segment in which they competed, compared with only three of the 28 companies established after 1970. Second, the Table shows that the turnover accounted for by new products was higher in younger companies.

Company age As in the previous analysis, independent sample t-tests were conducted to identify the impact of company age on innovativeness. None of the five types of innovation discussed earlier is found to depend on the year the company was established. A remarkable though not significant difference was identified for participation in the organic food chain. The companies having an organic logo were younger on average than those without this logo with mean ages of the two groups of 28 years and 53 years respectively. In order to identify the role of company age on the degree of innovativeness, the respondents were split into roughly equal groups by age of establishment. The first

Company size Independent sample t-tests were applied to see if the company size had any relationship to the innovativeness of food firms within the lower end of the size distribution considered in this study. Whereas in other studies micro companies have often been excluded from the analysis because of the lack of data, this analysis makes a distinction between micro companies (having up to ten employees) and small companies (having over ten employees). The results indicate that product innovation, process innovation and HACCP certification were independent of company size. However, company size was related to 13

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

Table I Frequency and impact of innovations by company age and company size (number of companies) Year of establishment Before 1970 After 1970 (N = 27) (N = 28)

Size (number of employees) Up to 10 More than 10 (N = 13) (N = 42)

Frequency (frequency of introducing products new to the market segments in which the firm competes) Never 16 15 10 21 Sometimes 3 10 2 11 Often or always 8 3 1 10 Impact of innovation on turnover Less than 1 percent 1-10 percent >10 percent

13 9 5

13 2 13

8 3 2

18 8 16

Source: Own survey in Hainaut and West Flanders, 2001

the implementation of the ISO 9000 standard. The mean size of companies that implemented ISO 9000 was significantly higher than that of firms that did not implement the standard (t = 1.85; p = 0.00). The opposite tendency was found for participation in the organic food chain (t = 3.78; p = 0.07). Finally, cross-tabulations were developed to identify the relationship between company size and the degree of innovativeness. In Table I, a distinction is made between micro companies and small companies. No significant differences were observed either for the frequency of introducing products new to the market segment in which the firm competed or for the impact of innovation on turnover. This also held true for patenting: two micro companies and three small companies had intellectual property rights in the form of patents. However, a significant relationship was found between the company size and R&D activities (2 = 5.91; p = 0.015): 85 percent of the small companies carried out R&D activities, whereas only 54 percent of the micro companies had R&D activities. Regional economic performance The Chi-square test was applied to identify the relationship between regional economic performance and innovativeness. The sample was split into two groups: the first group comprised the companies situated in Hainaut and the second group included the companies from West Flanders. HACCP certification, ISO 9000 implementation and participation in the organic food chain were found to be independent of the province in which the company operated. A significant relationship was identified between the company's

location, on the one hand, and both product innovation (2 = 3.15; p = 0.08) and process innovation (2 = 4.08; p = 0.04), on the other hand. In both cases, the companies in Hainaut scored higher than the companies in West Flanders. Over the past five years, 21 of the 26 respondents in Hainaut introduced product innovations and the same number introduced process innovation. In West Flanders, these figures are, respectively, 17 and 16 out of 29. Chi-square analysis indicates no significant difference in patenting behaviour or R&D activities between the two provinces. Crosstabulations reveal that the impact of innovation on the turnover was similar for both provinces. In contrast, the frequency of introducing products that are new to the market segments in which the firm competed show significant regional differences (2 = 3.96; p = 0.05). Twenty of the respondents in West Flanders never introduced products that were new to the market segment in which they compete compared with only 11 in Hainaut. Recent studies indicate a positive relationship between regional economic performance and innovativeness (European Commission, 1999; Roper, 2000). However, the results of this survey show that the small food-processing companies in Hainaut were more innovative than those in West Flanders. Therefore, the region with the lowest GRP per head and the highest unemployment rate had the most innovative small firms in the food sector.

Conclusions This paper examined the determinants of innovation in small food firms in Belgium 14

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

using data based on an in-depth survey of such firms in two provinces. The results show the importance of innovation in the vast majority of small food firms. Although such firms are limited in terms of investment and research facilities, innovation appears to take place continuously. The survey demonstrates that almost 90 percent of the small food firms have recently implemented some type of innovation. In contrast with the position in high technology industries, this was mainly innovation in the sense of the introduction of something new from the firm's perspective rather than in the strict sense of the first time this type of change has ever been introduced. In consequence, innovation in small food firms is seldom patented and patent data fail to measure the innovativeness of small food firms (Le Bars et al., 1998). Three conclusions can be drawn from the comparison of specific segments within the sample. First, capital-intensive innovations ± such as the implementation of ISO 9000 ± are more likely to take place in small firms compared with micro firms. Conversely, innovation that does not need much investment ± such as participation in the organic food chain ± is more common amongst micro firms. Second, the research is ambiguous on the relationship between company age and innovativeness: whereas older firms are more likely to introduce products that are also new to the market segment in which they compete, young firms tend to introduce innovations that have a larger impact on the firm's turnover. Third, the research indicates that geographical location does affect innovation. However, in contrast with earlier studies, this survey implies that the region with the better economic performance has the least innovative firms. This supports the view that small food firms may play an important part in innovation activities within lagging regions.

products and new processes. The managerial implications of the research are twofold. Although the food industry is generally considered to be a mature and slow-changing sector, it is clear that most small food firms regard innovation as essential. The food sector is in fact forced continuously to make changes in the production process in order to fulfil rising safety and quality regulations, and, at the same time, small food firms still regularly implement innovations that are not compulsory. This research shows that the domain in which innovations are carried out depends greatly on the profile of the firm. One question raised is whether innovations in small food firms are profitable. The analyses show that product innovation, although usually incremental rather than radical, typically accounted for a significant percentage of the firms' turnover. Apparently incremental changes in product and processes in small food firms might lead to major strategic changes at all levels of the firm in line with the suggestion of Figure 1. The findings of this research also have implications for innovation policy towards small food firms. Taking into account the economic importance and the dual structure of the food sector, it leaves no doubt that encouraging innovation in small food firms can be regarded as a strategy to stimulate regional and national growth. The study also shows that innovations based on R&D are rare in small food firms, mainly due to the lack of means and know-how to invest in R&D activities. This explains the role of interfirm relations for innovation in these firms. Small food firms tend to rely heavily on information from customers, suppliers, similar enterprises and research institutes as a source of innovation. In this context, governments could play a role in decreasing the communication barriers between firms and external parties in order to stimulate innovation (Baardseth et al., 1999).

Managerial implications Economic literature on innovation in small firms in relatively low technology industries is in short supply. Entrepreneurs within such firms mainly rely on personal experience and on the success and failures of similar businesses to decide on their innovative strategy. Small-scale trial and error experiments, rather than large-scale R&D activities, precede the introduction of new

Areas of further research The research shows the need for an appropriate tool for measuring innovation in sectors where R&D intensity is low and patenting is rare. Several recent innovation studies provide alternative ways to account for innovation. An important contribution has been made by the Community Innovation Survey Project (Kaiser, 2002). However, the CIS database aims to measure innovation in 15

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

all manufacturing industries and does not provide data to assess innovation patterns in small food firms. Another approach was taken in the work of Diederen et al. (2000), where an innovation index was constructed to measure innovation in the agricultural sector. The use of an innovation index ± composed as a sum of variables ± would allow small food firms to be classified on the basis of their overall innovativeness rather than focusing on specific aspects of innovation. Finally, this research is limited to food firms in Belgian regions. It would be interesting to extend the analysis to other European regions and carry out comparative studies to identify innovation patterns that are either regionspecific or generalisable. Such cross-country analyses should also allow conclusions to be drawn for European innovation policy towards small food firms which appear to play a potentially important role in the development of peripheral regions.

Cooke, P., Uranga, M.G. and Etxebarria, G. (1997), ``Regional innovation systems: institutional and organisational dimensions'', Research Policy, Vol. 26 Nos. 4-5, pp. 475-91. Cumming, B.S. (1998), ``Innovation overview and future challenges'', European Journal of Innovation Management, Vol. 1 No. 1, pp. 21-9. Diederen, P., van Meyl, H. and Wolters, A. (2000), ``Innovation in agriculture: innovators, early adopters, and laggards'', paper presented at the XXIVth International Conference of Agricultural Economists, Berlin. European Commission (1999), 6th Periodic Report on the Social and Economic Situation and Development of the Regions of the European Union, Brussels. Fanfani, R. and Lagnevik, L. (1995), ``Industrial district and Porter diamonds'', paper presented at the Strategic Management Society 15th Annual Conference, Mexico City. Grunert, K.G. and Ottowitz, T. (1997), ``Neumarkt LammsbraÈu: brewing beer for Greens'', in Traill, B. and Grunert, K.G. (Eds), Product and Process Innovation in the Food Industry, Blackie Academic & Professional, London, pp. 99-111. Grunert, K.G., Harmsen, H., Meulenberg, M., Kuiper, E., Ottowitz, T., Declerck, F., Traill, B. and GoÈransson, G. (1997), ``A framework for analysing innovation in the food sector'', in Traill, B. and Grunert, K.G. (Eds), Product and Process Innovation in the Food Industry, Blackie Academic & Professional, London, pp. 1-37. Hoogland, J.P., Jellema, A. and Jongen, W.M.F. (1998), ``Quality assurance systems'', in Jongen, W. and Meulenberg, M. (Eds), Innovation of Food Production Systems: Product Quality and Consumer Acceptance, Wageningen Pers, Wageningen, pp. 139-58. Johannessen, J.-A., Olsen, B. and Lumpkin, G.T. (2001), ``Innovation as newness: what is new, how new, and new to whom?'', European Journal of Innovation Management, Vol. 4 No. 1, pp. 20-31. Kaiser, U. (2002), ``Measuring knowledge spill-overs in manufacturing and services: an empirical assessment of alternative approaches'', Research Policy, Vol. 31 No. 1, pp. 125-44. Kotler, P. (1991), Marketing Management ± Analysis, Planning, Implementation and Control, PrenticeHall, London. Le Bars, A., Mangematin, V. and Nesta, L. (1998), ``Innovation in SMEs: the missing link'', paper presented at the High Technology Small Firms Conference, University of Twente, Enschede. Lundvall, B.-A. (1992), National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning, Frances Pinter, London. McDonagh, P. and Commins, P. (1999), ``Globalisation and rural development: demographic revitalisation, entrepreneurs and small business formation in the West of Ireland'', in Kasimis, C. and Papadopoulos, A.G. (Eds), Local Responses to Global Integration, Ashgate, Aldershot, pp. 179-202. Malerba, F. and Orsenigo, L. (1995), ``Schumpeterian patterns of innovation'', Cambridge Journal of Economics, Vol. 19 No. 1, pp. 47-65. Maurer, O. and Drescher, K. (1996), ``Industrial standards as driving forces of corporate innovation and

Note 1 Bakeries were excluded to avoid the inclusion of small retail shops with on-site bakeries. One firm in the sample had 56 employees but most of its data relate to the period when it had fewer than 50 employees.

References Antonelli, C. and Calderini, M. (1999), ``The dynamics of localised technology change'', in Gambardella, A. and Malerba, F. (Eds), The Organisation of Economic Innovation in Europe, Cambridge University Press, New York, NY, pp. 158-76. Baardseth, P., Dalen, G.A. and Tandberg, A. (1999), ``Innovation/technology transfer to food SMEs'', Trends in Food Science & Technology, Vol. 10 No. 6-7, pp. 234-8. Boudouropoulos, I.D. and Arvanitoyannis, I.S. (2000), ``Potential and perspectives for application of environmental management system (EMS) and ISO 14000 to food industries'', Food Review International, Vol. 16 No. 2, pp. 177-237. Breschi, S. (1999), ``Spatial patterns of innovation: evidence from patent data'', in Gambardella, A. and Malerba, F. (Eds), The Organisation of Economic Innovation in Europe, Cambridge University Press, New York, NY, pp. 71-103. Camagni, R. (1991), Innovative Network: Spatial Perspectives, Belhaven Press, London. Clarysse, B., Van Dierdonck, R., GabrieÈls, W., Lambrechts, J. and Uytterhaegen, M. (1998), ``Strategische verschillen tussen innovatieve KMOs: een kijkje in de zwarte doos'', Publication No. 5, IWT, Brussels.

16

Determinants of innovation in small food firms

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 8-17

T. Avermaete, J. Viaene, E.J. Morgan and N. Crawford

Traill, B. (1995), ``Small and medium food-manufacturing enterprises in the EU'', Discussion Paper, No. 11, AAIR Programme, of DGXII, University of Reading, Reading. Traill, B. (1997), ``Structural changes in the European food industry: consequences for innovation'', in Traill, B. and Grunert, K.G. (Eds), Product and Process Innovation in the Food Industry, Blackie Academic & Professional, London, pp. 38-60. Unnevehr, L.J. and Jensen, H.H. (1999), ``The economic implications of using HACCP as a food safety regulatory standard'', Food Policy, Vol. 24 No. 6, pp. 625-35. Varzakas, T. and Jukes, D.J. (1997), ``The globalisation of food regulation and market quality: a study of the Greek food market'', in Loader, R.J., Henson, S.J. and Traill, W.B. (Eds), Globalisation of the Food Industry: Policy Implications, The University of Reading, Reading.

internationalisation'', in Galizzi, G. and Venturini, L. (Eds), Economics of Innovation: The Case of the Food Industry, Physica Verlag, Heidelberg, pp. 221-39. Nooteboom, B. (1999), ``Innovation and inter-firm linkages: new implications for policy'', Research Policy, Vol. 28 No. 8, pp. 793-805. Noronha Vaz, M.T. and Nicolas, F. (2000), ``Innovation in small firms and dynamics of local development'', Workshop ISEG, Lisbon. Roper, S. (2000), ``Benchmarking regional innovation: a comparison of Baden-WuÈrttemberg, Bavaria, Northern Ireland and the Republic of Ireland'', Working Paper, No. 56, Northern Ireland Economic Research Centre, Belfast. Shumpeter, J.A. (1934), The Theory of Economic Development, Harvard Economic Studies, Cambridge, MA. Shumpeter, J.A. (1942), Capitalism, Socialism and Democracy, Harper, New York, NY. Taylor, E. (2001), ``HACCP in small companies: benefit or burden?'', Food Control, Vol. 12, pp. 217-22. Tether, B.S. (2002), ``Who co-operates for innovation, and why. An empirical analysis'', Research Policy, Vol. 31, pp. 947-67. te Velde, R.A. (2001), ``Shumpeter's theory of economic development revised'', paper presented at the ECIS Congress on the Future of Innovation Studies, Eindhoven.

Further reading Johne, A. (1999), ``Successful market innovation'', European Journal of Innovation Management, Vol. 2 No. 1, pp. 6-11.

17

The problem of managing technological innovations

Strategic management of technological innovations in the small to medium enterprise

An innovation strategy arises from the need to establish a linkage between customer needs and the needs satisfied by a company product (whether new or modified). This linkage must be not only better and stronger than competitors', but also sustainable over time (so that it may translate into a true competitive advantage), something that only the application of advanced technologies or proprietary know-how can achieve. Establishing and maintaining such linkage in order to best satisfy customers' needs (current and potential) through offerings that incorporate new technologies is what defines technological innovation. Implementing a strategy for such innovation involves pursuing two basic goals: (1) improving product/service quality with respect to two fundamental market dimensions: customers and competitors (translating in the short term into increased product competitiveness); (2) improving the company's technological level, once again relative to two dimensions: the current state of technological development; and competitors' positioning with regard to such technologies (translating in the medium to long term into increased competitiveness of company technologies).

Paolo Pratali The author Paolo Pratali is a Professor in the Department of Electrical Systems and Automation, Pisa University, Pisa, Italy. Keywords Small to medium-sized enterprises, Technological innovation, Competitiveness, Product innovation, Scenario planning, Innovation Abstract This paper addresses the problems inherent in identifying technological innovations that can improve company competitiveness with the ultimate aim of increasing the value of a specific enterprise. A model is proposed that, starting with the competitive weight of a technological innovation to processes or products, yields a strategic weight that enables decision makers to evaluate the increase in business value consequent on application of such innovation. The proposed model is composed of four sub-models: the first is an analysis of process/product competitiveness aimed at identifying competitive priorities and therefore appropriate technologies; the second sub-model identifies the priorities of technological intervention from amongst the competitive technologies selected; the third sub-model correlates the two previous sub-models and thereby expresses a ``strategic weight'' of the technological projects with respect to the competitive priorities of the processes or products; the fourth and last sub-model applies scenario simulation and sustainable growth verification to estimate the impact of strategic project innovations in terms of increased business value.

Achieving these two strategic goals requires an essential correspondence between the strategic decision-making process with regard to technological innovations and the dynamics of market and technological evolution. Strategic choices therefore involve evaluating the appropriateness of technological investments in order to improve existing products with the ultimate aim of furnishing the firm with an achievable, yet substantial ability to compete[1]. In fact, factors such as product quality, service level, lead-time, and so on, may have a fundamental impact on the company's possibilities for development and very survival, in that they represent prerequisites for achieving longterm success[2].

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

Interrelations between technology and product competitiveness Abell (1986) asserts that a product is the physical manifestation of application of a

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . pp. 18-31 # MCB UP Limited . ISSN 1460-1060 DOI 10.1108/14601060310456300

18

Strategic management

Paolo Pratali

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

particular technology. Companies must therefore seek to improve (or acquire ex novo) those technologies able to offer the greatest competitive advantage in terms of meeting customers' expectations. Of the various physical product properties (such as shape, colour, etc. but also, and especially, the ability to provide a series of functionalities), consumers deem some more relevant than others; these priority factors are represented by ``performance features'', or simply ``quality'', and in these terms the product itself simply represents the means to achieve such functionality. Through this interpretation, it is possible to identify a series of performance features whose effectiveness is linked to the technologies adopted to produce either the components themselves or the interrelations among such components that provide function. The introduction of technological innovations must therefore reap advantages in terms of product quality, intended as the set of performance features, which must translate into a ``capacity gap'' with respect to competitors that comes from applying better technologies. Therefore, it is necessary to consider the complex relations thus established between a company product, its performance features, the physical components making it up and the relative production processes, while using as reference points, on the one hand, customer requirements and, on the other, the competitors' offerings (Buzzell and Gale, 1988). Analysing these relations enables one to define the strong and weak points of various products in relation to market demands and what the competition offers. As mentioned, each performance feature must be analysed in relation to a product's constituent components and/or their interrelations, both of which can be achieved through various technological processes. Such analysis, conducted for each and every feature, allows one to identify a ``range'' of possible innovative solutions achievable through improvements to already existing technologies and/or through the acquisition of new ones. The innovations thus engendered will enable the company to substantially improve its competitive position through greater correspondence between market demands for a certain feature and the company's ability to offer it.

Technology as a dynamic factor The duration of the competitive ``capacity gap'' gained through the adoption of a given technological innovation, in terms of the time competitors need to imitate new features and improve their own production, has enormous consequences on the effectiveness of the company's strategies in the medium to long term: the longer this period is, the greater the advantages accruable by the company first introducing the innovation, thereby guaranteeing a less ephemeral success to the strategy adopted[3]. Hence, there is a clear need to formulate strategies able to combine technological with market opportunities, the aim being to achieve the aforementioned dynamic correspondence between innovative change and environment/market forces. It is only through such dynamic strategies that the goal of effective and lasting competitiveness can be achieved through innovation. The timeframe of technological advances is thus a fundamental consideration in adopting innovation strategies and calls for careful consideration of the dynamics of a given technology in analysing its potential effects on product quality. Therefore, due account must be taken of the technology's maturity, that is to say, its current stage of development with respect to its foreseeable life cycle. In fact, the set of technologies adopted by a company goes to make up its ``technological pool'' (comprising both ``hard'' and ``soft'' elements), which in time inevitably becomes obsolete (with respect to the performance evolution of various technologies) and must therefore be constantly renewed. In other words, the decision to implement or to acquire a given technology must also be assessed on the basis of its place within the company's technological pool. From this perspective, the various technologies will have to be evaluated, not only with regard to their possible future evolution (their maturity), but also with respect to the relationship between the company's technological positioning and that of its competitors. The results obtained through these two comparisons provide an assessment of the potential impact on the company's technological capacities of the technological innovation chosen to improve production quality. Furthermore, the more the company's technological capacities are based on the accrued technologies providing 19

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

performance advantages over competitors (at comparable levels of investment), the greater the competitive advantage they will afford. To sum up then, managing technological innovation involves the two simultaneous, interrelated fundamental objectives of competitiveness: (1) improving product quality (a prerequisite to success); (2) improving the company's overall technological quality (a prerequisite to lasting success).

The medium- to long-term results, which reflect company performance and can be linked to the strategic initiatives undertaken, are measurable through increases in company value (financial performance), according to the perspectives of survival, success and growth. Such perspectives involve various strategic-financial indices, of which the most fundamental is the cash flow engendered by improvements in the company's competitive abilities (increases in sales, market-share, profitability and self-financing ability). However, apart from the a posteriori stocktaking needed to assess the true effectiveness of the strategies followed, the investments called for in order to implement such decisions must be evaluated a priori with respect to the various alternative courses of action and their estimated chances of success. Only thus can the costs and the risks inherent in each be appraised before decisions are made.

Evaluating technological innovation strategies Innovation strategies must therefore stem from decisions of a strategic financial nature. If the main goals are short-term results, the strategy is to be oriented towards market defence (consolidating one's position) by pursuing costs reduction and resources optimisation. If, on the other hand, the financial strategy is oriented to medium- to long-term objectives, then priorities shift to the creation and/or development of new markets through new, more advanced products and technologies. The specific choices are determined by a number of factors, including the company's current position, financial market conditions, interest rates, self-financing ability, and market receptivity to innovative developments[4]. Moreover, attention to quality, service, product diversification, and so on, can transform traditional costs-oriented results into strategic results in terms of company performance. Therefore, the decision-making process with regard to strategic choices must ultimately be linked to overall company performance. From such considerations, it follows that the measurement criteria must be able to: . link strategies to objectives; . integrate accounting and non-accounting measures.

Managing technological innovation in small and medium-sized enterprises (SMEs) Medium- to small-sized firms are generally unprepared to tackle complex problems, for which the decision-making process requires, more than technical skills, experience or sound business sense, the ability to conduct far-reaching, systematic analysis of data and events for the most part extraneous to the company. Traditionally, the major difficulties facing SMEs with regard to strategic management of technological innovation involve their lack of managerial skills, the inadequate attention generally paid to technologies as a strategic variable, the reduced scope and varying stability of their field of operations (or niche), and the lack of qualified competitors. Recently, a new generation of executives has brought with them a greater awareness of and preparedness to address the new issues facing modern management: more and more, the potentialities inherent in technological innovations have given new impetus to creative drives and success-minded insight; the potential field of operations has been broadened considerably through the progressive elimination of borders and the opening-up of global markets; and the constant, rapid changes in customer needs are continuously eliminating old markets, while at the same time creating new ones.

Thus, the primary (medium- to long-term) objective of increasing company value can be divided schematically into two sub-goals: (1) market aims: increasing market share, expanding into various markets, and so on; (2) financial aims: increasing profitability, cash flow and, therefore, the medium- to long-term financial equilibrium. 20

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

Thus, we undertook to contribute what we believe is a useful ``how to'' method for managing technological innovations. This contribution, precisely because it is aimed at SMEs in particular, could not but be oriented to management activities and, therefore, to the formulation of a working model for the decision-making process.

Therefore, the foregoing considerations are more applicable than ever, and for all the more reason, to SMEs, simply because they find themselves less equipped to deal with such change. Often, the difficulty in obtaining the information necessary to assessing one's own competitive situation and conducting scenario analyses is invoked as the main factor keeping SMEs from performing strategic planning. We believe this to be, at least in many cases, more of an apparent difficulty than a real constraint on management; or, at least, we do not see it as sufficient to prevent SMEs' management from undertaking the formulation of wellfounded strategies. However, given the highly interdisciplinary nature of approaches to managing technological innovations, it cannot be denied that some difficulties do indeed exist. But, if properly tackled, such management can be facilitated by the realisation that much of the needed information can be deduced through experience and by analysing other data and information that are either already in the company's possession or easily accessible. Another related difficulty that is also more apparent than real involves the fact that information is not collected systematically, thereby rendering burdensome and incomplete any understanding of market evolution and the connected potential technological opportunities. Instead, a concrete limiting factor for the SME (and not only) on the implementation of innovative strategies is represented by the highly complex nature of the phenomena to be managed. Moreover, there is the related problem of defining clear, unequivocal objectives, and therefore evaluating the impact of the possible strategic alternatives on enterprise performance (which becomes perceptible in the medium to long term). Even the best-prepared manager, when faced with the multifarious complexity of innovation phenomena, may lose sight of the true goals of the decision-making process. Moreover, no manager can be expected to systematically follow the complex task of information processing, which opens up a myriad new alternatives, corresponding to as many as possible scenarios, and thereby further amplifies the already exacting number of interrelations and their consequent effects on overall objectives.

The methodology adopted Generally, decision-making problems of any complexity are addressed through application of quantitative DSS (Decision Support System) methodologies. Such methods, however, are aimed at ``problem solving'' in rather narrow, well-defined domains and provide the best results in stable contexts. They moreover call for specialised information-science expertise. Therefore, given the uncertainties inherent in potential future scenarios and the evolutionary dynamics of the financial, market and technological settings, tools such as Executive Information Systems (EIS) and Executive Support Systems (ESS) appeared more suitable. However, because of the need for a method oriented towards the development of information-science technologies, we resolved to adopt a socalled Intelligent Decision Support System (IDSS) (Pratali, 1986). IDSS systems integrate semi-structured processing models for evaluating alternatives. They yield prospectuses and reports that are able to evidence the rationale underlying the choices being made on the basis of the values of only a few factors, considered to be crucial. Such methodologies currently appear to be the most efficient and effective for the information processing underpinning strategic decisions. They leave decision makers the freedom to decide at what level of detail to handle the available information, thereby facilitating and fostering in-depth analysis and systematic diagnosis of the issues, an important prerequisite for coming to well-considered, foresighted decisions. The skills and creativity of individual decision makers are therefore not stifled; their insights can instead be examined and compared, one with the other, as well as with respect to the overall set of elements and factors (deriving from the synergistic 21

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

dimensions, customers and competition, with the aim of identifying the features in need of improvement. Then, the processes fulfilling these features are identified, and their influence (weight) measured. The last step in this stage then calls for identifying any technological alternatives able to improve the given features and their weight with respect to the performance benefits they can offer. (2) Stage 2: selecting the technologies able to improve the firm's technological capacity (evaluation of the technological intervention priorities (TIP), a mediumto long-term aspect). This consists of evaluating the company's position in the technologies selected in the previous stage with respect to the competition, as well as the maturity of the technologies themselves (i.e. their stage of development). (3) Stage 3: linking the two indicators (CIP and TIP) in order to evaluate the potential overall strategic benefits of adopting the selected technologies (strategic value of the technologies (SV)). (4) Stage 4: evaluating the increase in company ``value'' (in the medium to long term) consequent on implementation of the technological innovations with the highest SVs (company performance index (CP)).

relationships among the variables considered), which are then represented in a simple, yet thorough fashion. From this perspective, the adopted qualitative and quantitative analysis tools become useful in coming to decisions under conditions of uncertainty with regard to problems that cannot be completely structured in rigid mathematical models. However, it seems worthwhile cautioning that, however sophisticated, such analytical tools are all ultimately guided and managed by man. Therefore, the skills and expertise of those defining the input to, and the conditions of, the scenario, as well as the interpretative abilities of the decision makers, all represent determining factors in their correct application and, above all, for the reliability of the results they can yield.

Design elements of the innovationstrategies management model At this point, having defined our objectives and chosen the type of methodology to adopt, the next step becomes to design a conceptual scheme of the proposed system for the analysis and evaluation of innovative strategies according to the specifications presented in the foregoing. The model's framework is represented by the relations between technology, the market, and the firm's effective capacity to implement innovations. This last factor, in turn, involves, on the one hand, the resources a company is willing and/or able to commit and the consequent acquiring power and, on the other, the benefits that may be gained in the medium to long term (in terms of improved company performance). Therefore, the model serves to help select those technological strategies that prove to be the most promising for the ultimate aim of garnering a competitive advantage for the company. Four stages of the analysis, corresponding to as many distinct sub-models, can be distinguished: (1) Stage 1: selecting the technologies best able to improve the company's market competitiveness (evaluation of competitiveness intervention priorities (CIP), a short-term aspect). This is performed by assessing the position of product performance with respect to the two fundamental market

In short, the first two sub-models are aimed at analysing the technological factors determining the company's strategic position, and therefore serve to identify the innovative strategies able to enhance the company's competitive and technological position (that is, improve the company's production and overall technological quality). The third submodel serves to integrate the two previous choices, selected separately for the two distinct dimensions, into a synoptic indicator of the importance of specific technologies for the company's overall competitiveness, termed the ``strategic value'' of the innovation. Finally, through the fourth submodel, we seek to measure the effects of the highest-ranking potential strategies (that is, with the highest SVs) on company performance, in terms of the potential increase in ``company value'' in the medium to long term (Figure 1). 22

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

Figure 1 The four sub-models

Sub-model 1: the competitive priority of technology Analysis of competitiveness Concerning a company's ability to compete, we began with the basic consideration that the more a product's performance features correspond to market demands, as well as to those of competing products, the more competitive will be the company's position in the market. Therefore, company performance must be defined and assessed with respect to market offerings. Defining performance signifies establishing the parameters that target customers equate with ``quality''. This involves compiling a company ranking with regard to such parameters, a ranking within which the firm must then evaluate its position by determining the correspondence between the quality ``demanded'' and that which it ``supplies'' in consideration of both of the aforementioned dimensions of customers and competitors. Therefore, the first step in defining the sub-model's framework consists of identifying those variables capable of ``explaining'' such a complex phenomenon as the market position of company products. The first relation to take into account is that between customers and the company. As product features can be assumed to be the parameters measuring quality, this relation can be determined by analysing the correspondence between the quality level sought for by customers and that of company offerings. Such relationship is expressed through the variable termed the index of ``called-for improvement'' (CI), which represents the qualitative distance, or gap, between market demands and company offerings. Therefore, the wider this gap is, the worse will be the ability of company products

to satisfy customers and, consequently, the greater the improvements ``called for'' by customers. Analogously, the relationship between the company and its competitors, defined as the ``level of competitive capacity'' (CC), can be analysed and estimated by comparing the level of quality offered by the company with that offered by the competition. Thus, the better the features (sought for by customers) offered by company products in comparison with those of the competition, the greater will be the company's competitive capacity. Figure 2 illustrates the process by which the values of the two variables CI and CC are determined. Each of the two identified variables (CI and CC) is assigned a qualitative rating (high, medium or low) and positioned on a twodimensional matrix (such as a classical Ansoff or BCG matrix, or similar approaches), in which the abscissa represents the ``level of competitive capacity'' (CC) and the ordinate the ``called-for improvement'' (CI). By means of this matrix, the competitively weak features (i.e. in need of improvement) can be easily identified as those exhibiting the maximum distances between company and market, and company and competitors (Figure 3)[5]. Such features are therefore characterised by a high level of called-for improvement (expressing the discrepancy with regard to market demands) and low competitive capacity (revealing the company's weakness vis-aÁ-vis the competition). However, as not all features will have the same importance for users, each must be assigned a weight in order to account for the degree of market demand for that particular feature, and therefore furnish a basis for 23

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

Figure 2 Selection of product performance features needing improvement in consideration of market demands and competing products

Figure 3 Measure of the improvement priority (IP)

evaluating the need for improvement. Thus, the different quadrants of the matrix will contain features with differing weights, so that, in order to estimate the overall competitive situation, it is necessary to define an idealised, theoretical reference feature characterised by the highest called-for improvement, the lowest competitive capacity and the highest weight. Such feature is thereby assigned a maximum value, termed ``absolute priority''. Then, by comparing all the other, real features with this hypothetical ideal, it is possible to define a set of those features most in need of improvement, according to the resulting relative values, called the ``improvement priorities'' (IP), obtained by combining each feature's matrix position with its weight. The use of a weighted element allows improvements in the competitive situation to focus on those features that, on the one hand, represent the company's weakest points and, on the other, have a determining influence on user preferences.

The advantage of this type of analysis lies in the fact that it can be applied to already existing products, as well as to new ones. In the latter case, the evaluation can be performed through two alternative methods. The first one consists of assigning a null value to the quality currently offered, so as to automatically insert the new product's features into the list of those to be improved. The second possibility consists of precisely defining, rather than the quality currently offered, that which could potentially be offered by the company based on its actual capacities. Although this second method clearly involves greater complexity of the analysis, it offers the advantage of delineating a more thorough, realistic picture of the company's competitiveness, one that corresponds to its effective needs and capacities. Potential innovations can therefore be best identified as market opportunities that 24

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

The overall procedures involved in this first sub-model (analysis of the competitive capacities of technology) can therefore be schematised in the following steps (Figure 5): . identifying the product's most important performance features; . evaluating the competitive value of each feature and defining its improvement priority (IP); . identifying the process(es) influencing performance features and assigning a weight to each of these features (FW = feature weight); . identifying technological alternatives and evaluating their importance in providing for critical processes (TW = technology weight); . defining the technological intervention priority (CIP) for each technology, which is represented by the sum of the various values obtained by multiplying the improvement priority (IP), the weight of each process on the feature (FW) and the importance of the technology to the process (TW).

present themselves from time in time. Moreover, the analysis enables the company to arrive at an estimate of the competitiveness of a product even before it is made available[6]. Analysis of technological capacity The second stage (still within the first submodel) consists of determining what technologies could be acquired or improved in order to achieve concrete competitive advantages. To this end, products must be viewed as combinations of processes, each of which contributes to a varying extent to providing various performance features (Figure 4a). This ``contribution'' will then represent the ``weight'' of the process(es) with respect to the feature in question. Thus, the higher this value is, the greater the need to improve the process(es) will be. In practical terms, this means that, by multiplying this weight by the ``improvement priority'' (IP), we obtain a value that can be thought of as an index representing the perceived power of that process to improve the feature, and consequently the product and, ultimately, the company's competitive capacity. Considering that every process can be carried out through various alternative technologies (Figure 4b), the technology to be acquired or improved can be identified by assigning a ``technological weight'' (TW) to each technology with respect to each process, and then calculating the product between this value and the index representing the need for improvement of the process(es). The resulting value represents the given technology's power to improve the product (competitiveness intervention priority (CIP)[7]. Therefore, for each technology, the sum of its priorities of competitive intervention for the various processes represents an index of its power to improve products (and, consequently, the company's competitiveness).

Sub-model 2: the competitiveness of technology All technologies have limits to their applicability, and such restrictions must clearly be adequately accounted for by companies considering innovations. Such limits do not stem solely from a technology's power to contribute (in various ways and to varying degrees) to product improvement, but rather the given technology's proximity to a stage of discontinuity, that is to say, the moment in which its utility begins to decline due to the advent of new and more effective means to the same end. However, although we cannot speak of a true ``technostructure'' when dealing with SMEs, we must

Figure 4 Linking technology to performance via process(es)

25

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

Figure 5 Summary of the first sub-model

which an evolving technology can be placed within its foreseeable life cycle. Such evaluation involves some evident difficulties that can, however, be resolved through recourse to the opinions of experts in the given field. The second parameter, instead, represents the company's ability to leverage the technology in question better than its competitors do. It can therefore be determined by evaluating the company's know-how, patents, human and financial resources and R&D investment, and comparing them with the corresponding values and situations within competing firms. As before, the type of approach allows a two-dimensional matrix to be constructed (Figure 7). In this case, we plot the variable TM along the abscissa and variable TC on the ordinate. The use of qualitative ratings once again renders the values homogeneous, thereby allowing identification of those technologies to be fostered and those to be eschewed with the aim of improving the technological quality of a given production unit. Given the difficulties involved in applying such an approach (which go beyond those of garnering the necessary information), the SME may elect to forgo such a technological quality assessment of the company. In this case, however, an indirect assessment of the company's technological quality can nevertheless be included in the relations between technologies and processes by simply increasing the weight of those technologies that will presumably undergo greater future development.

nonetheless recognise the existence of ``technological quality'', represented by the capacity of a company's accrued technical means to yield the best results possible in terms of cost to performance ratio. Such considerations underlie the following procedure for determining a sort of company technological position through the interrelations existing between its technological capacity, that of competitors, and the state of maturity of the technology in question (Figure 6). To this end, by analogy to the procedure outlined in the foregoing for the relations linking products and features, each selected technology must be evaluated with regard to its possible future evolution (``level of technological maturity'' = TM), as well as to the company's ability to compete technologically with other comparable firms (the company's ``technological capacity'' = TC). The value of the first parameter can be determined by estimating the current stage in Figure 6 Measuring the priority of technologies (TIP)

26

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

Figure 7 The priority of technologies (TIP) matrix

Sub-model 3: the strategic value of technology At this point, we can proceed to an evaluation of the ``strategic value'' of the technologies considered, an index of which is obtained by crossing the values of the ``competitiveness intervention priority'' (CIP) with those of the ``technological intervention priority'' (TIP). This is clearly possible only because these values are homogeneous, as they both stem from qualitative judgements, and can therefore be linked in a matrix (Figure 8), in a manner similar to that previously described for technologies and processes. Each technology is assigned a strategic value (SV), which represents its ability to confer benefits on the company in terms of improvements to both its products and its overall technological standing. The matrix therefore provides an overall, synoptic view of the ``strategic'' position that a company can achieve by

fostering those technologies with high values of SV. To provide an overview of the steps covered so far, Figure 9 shows a schematic, integrated outline of the three sub-models described. Sub-model 4: company performance assessment As mentioned in the foreword, the effects of adopting technological strategies are manifest through measurable increases in market shares and/or expansion into previously untapped markets. Therefore, any parameter used to measure the consequent improvements in company performance (on which its future success and prosperity depend) must necessarily be expressed in economic-financial terms. However, as also mentioned, traditional financial measures are unsuitable for estimating such phenomena. Return on investment (ROI) and other static indicators are unable to reflect the dynamic nature of technological strategies: they cannot take into account the supplementary investments necessary to maintain competitiveness, nor are they able to express the variability in results in the medium to long term. In times of rapid change, the assessment system must be capable of accounting for the impact of market dynamics on company performance. The problem therefore becomes one of formulating a measure of company ``financial performance''. Various approaches to solving this problem have been advanced. Recently, many seem to be oriented to measuring performance in terms of the creation of company ``value'',

Figure 8 Matrix for the ``strategic value'' of technology (SV)

27

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

Figure 9 Integration of the three sub-models for evaluating technological priorities

decision-making process for technological innovations. The fundamental objectives are to endow the model with the following abilities: . to measure variations in company value; . to verify the ``practicability'' of development; . to simulate scenarios (both internal and external); . to conduct sensitivity analysis in order to estimate the incidence of risk factors, the reliability of the processed information and the varying degrees of uncertainty inherent in the assumed scenarios; . to express results in the form of synoptic reports that, as previously stated, can evidence the criticality and significance of the different parameters and variables defined.

that is to say, a system (such as economic value added (EVA)) that enables one to estimate decisions with the potential to increase a company's economic value in the medium to long term. As is clear from the foregoing, such a ``value'' system has been adopted in formulating the current assessment model[8]. By improving its technology, a company, by virtue of its enhanced competitiveness, can increase not only its sales volume, but also its productivity, thereby reducing costs. On the other hand, the problem of the availability of the financial means necessary to implement innovations may be aggravated by the increase in circulating capital consequent on greater sales revenues. Implementing innovation raises a far-ranging series of economic and financial issues inextricably linked to the size of the investments involved. In fact, a company decision to make a certain investment, rather than another (also in relation to the amounts involved), is not a matter of indifference. Therefore, it is necessary, not only to coordinate a set of factors entailing variable degrees of uncertainty (and therefore risk), but to link the various assessment parameters of the decision-making process (costs to sustain vs. anticipated benefits). The foregoing considerations have led to definition of the specifications for a performance assessment model of the

Measuring variations in company value In order to achieve this first objective, the model adopts the well-known concept of measuring company ``value'', formulated for our current purposes in terms of the increase in economic value consequent on implementation of a given strategy. The parameter adopted to reflect such increase is the projected additional net cash flow consequent on adoption of the strategy (over a four- to five-year forecast horizon) (Chiavaccini and Pratali, 2000). The increase in company value is therefore given by the following relation[9]: 28

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

EVA …
Value† ˆ Discounted ‰NCF …forecast horizon†Š ‡ Final value Initial value where NCF ˆ Net Operating Profit Net Added Investments: The only modification made to the original formulation is the introduction of a ``desired'' return on venture capital for calculation of the discount rate (the average, weighted with respect to indebtedness, between the return on venture capital and the cost of borrowed capital)[10]. Thus, the model aims to calculate the increase in value achievable through the selected (i.e. high-priority) innovation strategies for three hypothetical levels of sales increases (minimum, medium, or maximum, corresponding to pessimistic, positive, or optimistic expectations) under conditions of fixed capital costs, indebtedness, and so on.

Figure 10 summarises the structure of the submodel up to this point. The input data relative to each selected technology (output from the first three sub-models) must be derived from: . the estimated growth in sales revenues for three possible scenarios (less favourable, medium, and more favourable) consequent on varying degrees of market responsiveness and/or competitor reactions (new offerings); . the estimated variation in both variable and fixed operating costs consequent on implementation of the technology, the former being linked to hypothetical increases in productivity, the latter to an enhanced structural capacity; . the estimated circulating capital requirements; linked to the increase in sales volume; . definition of financing means, the return on capital and the cost of money. Reporting results synoptically As previously stated, the final results of the analysis must be structured in such a way as to provide easily interpreted, synoptic information, which must, however, be thorough in terms of both the details provided relative to each of the various technologies considered and the comparative data furnished. Moreover, the reporting must allow for varying the values of the significant parameters and presenting the results of simulation in an easily readable, summary form for all the technologies, including clear indications of the degrees of uncertainty involved in each parameter (increase in value and development practicability)[11].

Development practicability We use the term ``development practicability'' to refer to the combination of the ability to sustain the sales increase (with a given return on investment (ROE) and dividends distribution policy), on the one hand, with the duration of the value created, on the other. In other words, the ultimate aim is to verify the feasibility of the financial growth that is compatible with the created value, in the light of financing (indebtedness ratios) and dividends-distribution policies. Therefore, the model must calculate the maximum practicable growth rate of sales, as well as that rate which will no longer be able to furnish any increase in value at all (with the given indebtedness and dividends policies).

Conclusions

Simulating scenarios and conducting sensitivity analysis This consists of endowing the model with the ability to effect ``what-if'' analysis through the processing of those parameters deemed to be most meaningful to the creation of added value. Clearly, the analysis must account for the risk factors stemming from the varying levels of uncertainty and/or reliability inherent in the chosen parameters. Of the various possible parameters, we shall consider the following: . indebtedness ratio; . incidence (on turnover) of working capital; . percentage of hypothecated earnings; . variation in the percentage of operating costs (productivity).

Before drawing any final conclusions, it must first be said that we do not purport the Figure 10 Sub-model for evaluation of company value

29

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

proposed model to be exhaustive. It most certainly suffers from limitations in its resolving power with respect to the many and multifarious effects on performance consequent on the decision-making processes involved in adopting innovation strategies. Perhaps the most important of such effects are those due to the synergies activated between old and new technologies, though those stemming from the simultaneous action of the various assessment parameters (flexibility, cycle times, service levels, and so on) should not be underestimated. Moreover, technological innovation, whether it involves products or processes, nearly always leads to a revision of the design criteria underlying the production system (methods, cycles, lay-out, programming and control, and so forth), and will therefore have an impact on the organisational-managerial system as well; in other words, it almost inevitably involves ``systems innovation''. Nonetheless, our belief is that this model, despite such limitations, represents a first important step in defining a method enabling the decision maker to quickly, efficiently and interactively obtain important information for the selection of the most promising strategies in terms of increasing company competitive capacity. We, moreover, believe that the model, through its various stages (analysis of competitiveness, technology and strategic value), offers the major advantages of helping management, first, to ``understand'' the market, its customers, products and competitors and, second, to reveal the close interrelations between the dynamics of technological innovations, product quality and company know-how and, lastly, to adopt a forward-looking perspective, thereby promoting commitment to strategic planning. In closing, its seems worthwhile stressing the novelty of the approach, which resides in its shifting the treatment of the issues at hand from a ``problem solving'' perspective towards one of ``problem setting''.

Notes 1 Customers' perceptions of product or service quality is a key factor in market competition, which more and more often depends on the company's ability to properly define performance features in conformity with its customers' wishes and equip its products accordingly through production processes able to guarantee high quality and reliability (De Witt, 1993).

30

2 Competitive capacity is measured in terms of the future; therefore it is necessary to pursue effectiveness rather than efficiency strategies (Kaplan and Norton, 1992). 3 Quality contributes to company performance in two ways: in the short term, higher quality produces greater profits through price increases; in the long term, higher relative and/or improving quality constitutes the most effective way to expand and maintain business. 4 The fear of failure may attain such levels that the perceived risks impede investment decisions. Alternatively, the investments needed in order to implement the strategic design may be so burdensome as to jeopardise the company's financial equilibrium. 5 In the working model, the qualitative evaluations are converted to scores of 1 to 5 (1 = low, 5 = high). 6 The analysis procedure described requires the company to gather a good deal of information and data: the performance features and their weights, the quality demanded and that offered by both the company and competitors. Such data can be garnered, for example, through specific surveying methods (market trials) or input into the model interactively. The first system is undoubtedly preferable, provided that reliable data are accessible. The second, which should instead be used when the available information is incomplete, consists of proceeding by successive approximations, in which the analyst modifies the underlying assumptions depending on the step-wise processing results of the data input. 7 There are information issues in this case as well: analysis of the technological variables calls for the support of ``technologists''. Such specialists, apart from general expertise in the field of technology, must also possess specific experience in the company's field of operations, a good knowledge of the product and the ability to formulate objective judgements. 8 However, above and beyond the indicators chosen to measure the effects of the adopted strategies, the effectiveness of the hypothesised model instead resides in its ability to process various hypotheses and scenarios with which, through the selected indicator, to compare the different decision-making alternatives according to different perspectives (profitability, risk, significance, etc.) 9 Assuming that company activities at the start of the forecast horizon (initial value), as well as those at its end (residual value), generate positive cash flows in the precise amounts necessary to cover the investments costs of maintaining a constant company-economic value (i.e. a ``perpetual flow'' system). 10 The return on capital should be measured in relation to the minimum return (obviously, after accounting for projected inflation) that, on the one hand, would not prompt disinvestment, but, on the other, is sufficient to attract new injections. 11 For example, the uncertainty in and/or reliability of the various parameters' values can be linked to the resulting degree of variability in the solutions in order to determine those factors deserving of the

Strategic management

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 18-31

Paolo Pratali

Bellandi, G. (1989), ``L'innovazione tecnologica e la gestione della qualitaÁ'', in Raffa, M. (Ed.), Innovazione Tecnologica ed Impresa, CUEN, Napoli. Biois, K.L. (1980), ``The manufacturing/marketing orientation and its information needs'', European Journal of Marketing, Vol. 14, No. 5/6, pp. 339-53. Burgelman, R.A. and Maidique, M.A. (1988), Strategic Management of Technology and Innovation, Irwin, Homewood, IL. Christensen, C.M. (1997), The Innovator's Dilemma, Harvard Business School Press, Boston, MA. Corti, E., Guglielemini, M. and Zollo, G. (1989), ``A decision support system for the technology strategic planning'', paper presented at the TII European Conference, Nizza (Nice). Corti, E., Guglielemini, M. and Zollo, G. (1990), ``Il processo decisionale per la valutazione della tecnologia nella grande impresa: la metodologia Valutech'', Convegno AIRO, Sorrento. Coyne, K.P. (1986), ``Sustainable competitive advantage: what it is, what it isn't'', Business Horizons, Vol. 1, January-February. Foster, R. (1987), Innovation: il vantaggio di chi attacca, Sperling e Kupfer, Milano. Gilardoni, A. (1984), Le politiche tecnologiche delle imprese industriali, F. Angeli, Milano. Gottardi G. (1986), ``Modelli di diffusione tecnologica come strumenti di pianificazione'', in Atti Workshop, Bressanone, Ed. Cedam, Padova, pp. 95-118. Lynch, R.L. and Cross, K.F. (1992), Migliorare la performance aziendale, F. Angeli, Milano. Pratali, P. (1996), Progettare il profitto 1, 2 e 3, F. Angeli, Milano. Quinn, J.B., Baruch, J. and Zien, K.A. (1997), Innovation Explosion: Using Intellect and Software to Revolutionize Growth Strategies, Free Press, New York, NY. Rappaport, A. (1989), La strategia del valore, F. Angeli, Milano. Rolfo, S. (2000), Innovazione e piccole imprese in Piemonte, F. Angeli, Milano.

most attention (thereby prompting greater focus on the most reliable values that produce significant variations in the solutions, as opposed to those presenting greater uncertainty and low significance).

References Abell, D.A. (1986), Business and Company Choice, IPSOA, Milano. Buzzell, R.D. and Gale, B.T. (1988), I principi PIMS, Sperling e Kupfer, Milano. Chiavaccini, R. and Pratali, P. (2000), Progettare i processi d'impresa, Cap. 2, F. Angeli, Milano. De Witt, G. (1993), ``Produzione snella: eccellenza competitiva'', L'impresa, n. 2. Hax, A.C. and Majluf, N.S. (1987), Direzione Strategica, IPSOA, Milano. Kaplan, R.S. and Norton, D.P. (1992), ``The balanced scorecard: measures that drive performance'', Harvard Business Review, January-February, pp. 71-9. Kaplan, R.S. and Norton, D.P. (1996), ``Using the balanced scorecard as a strategic management system'', Harvard Business Review, January-February, pp. 75-85. Pratali, P. (1986), ``I Decision Support System nella gestione strategica delle innovazioni: una proposta di ricerca'', in Atti Workshop, Bressanone, Ed. Cedam, Padova, pp. 119-42.

Further reading Aprea, G. (1998), Prodotto innovativo o innovazione tecnologica? Vade-mecum per le piccole e medie imprese, F. Angeli, Milano. Azzone, G. and BerteleÁ, U. (1992), Techniques for Measuring the Economic Effectiveness of Automation and Manufacturing Systems, Quaderni MIP, Consorzio MIP, Milano.

31

Introduction

Cross-functionality and leadership of the new product development teams

Differentiation through new product development (NPD) is one of the most effective strategies for achieving success. The working teams that the firms select to carry out their NPD processes are the core of this strategy. They are responsible for understanding customer needs and for obtaining information about the markets, technologies, competitors and resources, with the aim of converting these needs into powerful, effectively manufactured product concepts that are profitable for the firm and valuable for customers. In this sense, crossfunctionality has become one of the fundamental characteristics of the NPD teams, its enormous popularity growing in the last few years. So much so that a recent study (McDonough, 2000) found that 97 percent of the firms analysed had used cross-functional teams and a third of those firms used them 100 percent of the time. Likewise, the impact of these cross-functional teams on the NPD processes is strongly influenced by their leadership (Jassawalla and Sashittal, 2000), and thus the choice of an effective leader who guides the team throughout the whole NPD process has also become an essential issue. In fact, literature on the functions that effective team leaders should assume is rich and abundant (Barczak and Wilemon, 1989, 1991, 1992; McDonough and Barczak, 1991; Ancona and Caldwell, 1990a, b, 1992a, b; Rossy, 1992; Henke et al., 1993; Hershock and Braun, 1993; McDonough, 1993; Cooper and Kleinschmidt, 1994; Jassawalla and Sashittal, 2000). The aim of this work is to analyse the impact of the use of cross-functional teams and effective leaders on the success of the NPD processes of Spanish firms. Specifically, we shall attempt to analyse if the use of crossfunctional teams and of solid and influential leaders results in better development times, more efficient processes and superior products, on the one hand, and higher percentages of successful new products in the market, higher frequency of introduction of new products and higher level of customer satisfaction, on the other. With this goal, we have carried out an empirical study in the most innovative industries in Spain through a

Sandra Valle and LucõÂa Avella

The authors Sandra Valle is Assistant Professor and LucõÂa Avella is Associate Professor, both in the Business Administration Department, University of Oviedo, Oviedo, Spain. Keywords New product development, Effectiveness, Internal control, Leadership, Innovation, Spain Abstract This study analyses the effect of the use of crossfunctional teams and effective leaders on the success of the new product development (NPD) process. With this aim, a sample of 125 firms representing the most innovative industries in Spain has been used. Results show that firms using cross-functional teams obtain a more effective NPD process (that is, better development times and costs, and superior products) and a higher percentage of new products that are successful in the market. Likewise, the firms that use effective leaders achieve better development times, superior products and a higher level of customer satisfaction. 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

This work was partially financed by the research project PB-EJS01-09. The authors are sincerely grateful for the invaluable comments and suggestions of Esteban FernaÂndez and Beatriz Junquera (University of Oviedo, Spain).

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . pp. 32-47 # MCB UP Limited . ISSN 1460-1060 DOI 10.1108/14601060310456319

32

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

sample of 125 manufacturing firms representing those industries. Thus, the purpose of this work is threefold: first, to delve deeply into the understanding of the practices adopted by Spanish firms in their NPD processes; second, to obtain results that are an empirical contribution to the literature on NPD success factors; and, finally, to offer management effective recommendations that allow them to increase the likelihood of success of their NPD processes. This work has been organised into four sections. In the first one, we include the main contributions on the subject found in the literature and present the hypotheses derived from them. In the second section, we present the methodology used in the research. In the third, the analysis carried out and the results obtained are shown. Finally, in the fourth section, we draw the conclusions of the work, together with a series of recommendations for the firms' management.

directly or indirectly related to the design, engineering, manufacturing and marketing of the product for which they are responsible (Henke et al., 1993). In this way, we obtain cross-functional teams formed by members from different specialities and with complementary skills. Each member must have an equal stake in the project and, as a team, must be accountable for the entire process, from beginning to end (not just one phase) (Cooper, 1999). The functional diversity of the development teams increases the quantity and variety of information available to design new products. Thus, teams with various specialities have certain advantages with respect to the least diverse ones (Rochford and Rudelius, 1992). The existence of more information helps team members to understand the design process more quickly and fully and from a variety of perspectives, making it possible, in this way, to improve process performance (Brown and Eisenhardt, 1995). In fact, cross-functional teams have the acknowledged merit of reducing errors, improving new product decisions, effectively organising work flow and producing more creative solutions (Thamhain, 1990; Clark and Fujimoto, 1991; Donnellon, 1993; Henke et al., 1993; Brown and Eisenhardt, 1995; Frohman, 1995). Numerous research works have examined the relationship between the use of crossfunctional teams and the success of the NPD process (Takeuchi and Nonaka, 1986; Souder, 1987; Hayes et al., 1988; Larson and Gobeli, 1988; Gupta and Wilemon, 1990; Zirger and Maidique, 1990; Clark and Wheelwright, 1992; Dougherty, 1992; Ford and McLaughlin, 1992; Millson et al., 1992; Rosenthal, 1992; Henke et al., 1993; Page, 1993; Zangwill, 1993; Cooper and Kleinschmidt, 1993, 1994). Although it is true that some of the studies have not found a really clear relationship (Ancona and Caldwell, 1992a; Clark and Wheelwright, 1992), most of them agree that the use of cross-functional teams has a positive effect on the performance of the NPD processes (Cooper and Kleinschmidt, 1990; Zirger and Maidique, 1990; Clark and Fujimoto, 1991; Dougherty, 1992; McDonough and Griffin, 1997; McDonough, 2000). In this sense, it has been observed not only that the use of cross-functional teams allows the quick incorporation of new products into the market (Takeuchi and Nonaka, 1986; Gupta

1. Literature review and hypotheses 1.1. Cross-functional NPD teams The development of a new product is a very complex process that requires the participation of the different functional areas of the firm, basically engineering, manufacturing and marketing. In general, good designs, wellexecuted testing and high-quality prototypes are needed from engineering; from marketing, thoughtful product positioning, solid customer analysis and well-thought-out product plans; and from manufacturing, capable processes, precise cost estimates, skilful pilot production and ramp-up (Wheelwright and Clark, 1992). Participation of areas such as purchasing, finances, human resources and other support functions is also important. In short, in order to carry out the NPD task, it is necessary to simulate the business as a whole. Before a new product reaches the market, it will have needed the specialised knowledge of each of the organisation's functions to one degree or another. Therefore, it is recommended that the teams dedicated to NPD should be crossfunctional, including specialists from the different departments and even extending beyond the limits of the firm to include suppliers and customers. The functional areas that are represented in a cross-functional team vary from one firm to another and from industry to industry, but they generally include the functions that are 33

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

and Wilemon, 1990; Cooper and Kleinschmidt, 1994; Towner, 1994; Eisenhardt and Tabrizi, 1995; Choperana, 1996; Zirger and Hartley, 1996; Griffin, 1997b), but also that its use is positively associated with the fulfilment of the premises stated for development, the design of highquality products, team member satisfaction (McDonough, 2000) and increased product success in the market (Larson and Gobeli, 1988; Griffin, 1997a). In spite of the existing empirical evidence, testing and cross-validation of hypotheses and theories are necessary to advance in the field of NPD. When the same results are obtained from different research studies using completely different databases, greater confidence is gained in such results. The replication of the studies diminishes the concern about the possibility of obtaining spurious results and is the main motivation of a large number of empirical works. Owing to the above, we have put forward the following hypotheses in this work: H1. The use of cross-functional teams has a positive impact on the success of the NPD process. H1a. Firms that use highly cross-functional development teams achieve a greater level of internal success of the NPD process; that is, they obtain better development times, are more efficient and obtain superior products. H1b. Firms that use highly functional development teams achieve a greater level of external success of the NPD process; that is, they are characterised by a higher percentage of new products that are successful in the market, a higher frequency of introduction of new products and a higher level of customer satisfaction. 1.2. Effective NPD team leaders The NPD team leader[1] is responsible for actively and effectively directing the interaction between the team and senior management, acting as a bridge between both. Moreover, he has the function of facilitating the NPD process, and of obtaining resources, diversity and autonomous conditions necessary for success. His global responsibility is to manage and support the entire NPD process from beginning to end, maximising the value of the product for customers and integrating the different segments of business into a 34

strategically calculated whole (Murphy and Gorchels, 1996). Therefore, the choice of an appropriate leader to run the team is essential insofar as he is the pivotal figure in the process. In this sense, the NPD team leader must possess the following characteristics: 1. Power and leadership. Power is the capacity to influence, while leadership is the process or act of influencing (Frischer, 1993). Leaders with power are those who have significant decision-making responsibility, organisational authority and a high-ranking hierarchical level. Thereby, these leaders are highly effective in obtaining resources (both human and budgetary) for the project team (Brown and Eisenhardt, 1995). Ancona and Caldwell (1992b) observed that powerful leaders are highly effective when it comes to pressing to obtain the necessary resources, protect the group from outside interference and manage the impressions of outsiders. In contrast, less powerful leaders are likely to be less successful in gaining needed talent and financial support and in shielding the team from outside interference. On the other hand, powerful leaders are capable of inspiring more respect and so attract the best members for the development team, keeping them focused and motivated (Clark and Fujimoto, 1991). However, rather than the power and management position of the leader, it is his behaviour in itself and his leadership style that in the end are going to stimulate the labour climate, which results in successful NPD processes (Norrgren and Schaller, 1999). Leaders with leadership capacity are those who like power and exercise it so that everybody works in the interest of a common objective. These leaders must be strong and influential and must not try to please or be popular at any price (Frischer, 1993). A project leader who tends to avoid conflicts and problems, saying yes to everything and everybody so as not to upset the others, will end up losing the respect and confidence of his workers. Moreover, these leaders must exercise their power through a participatory, open and apolitical leadership style, which consists of delegating responsibility and freedom to the team members for taking decisions, and in sharing information and knowledge with them and other groups in the organisation. Likewise, they must be capable of guaranteeing the personal commitment of the team participants, of building information-intensive environments, of strongly centring on human

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

(3) Planner. This role consists of planning the development activities in accordance with the strategic vision of the firm so that these activities can be performed in the necessary form and time limits. Insufficient planning may bring about inter-functional conflicts and discontinuity in the workflow. (4) Mediator. The leader must act as a link between the team, the rest of the organisation and the firm's customers.

interaction and of promoting learning within the team (Jassawalla and Sashittal, 2000). According to the empirical research work carried out, a leadership style of this kind is associated with higher NPD process performance (McDonough and Barczak, 1991; McDonough and Griffin, 1998). 2. Vision. The project leader must have vision; that is, he must have the cognitive ability to mesh together firm competencies and strategies with the needs of the market to create an effective product concept (Brown and Eisenhardt, 1995). These characteristics clarify the meaning of the product for the firm and are necessary for its success. Team leaders, together with senior management, frequently shape the overall product concept and communicate it to team members (Clark and Fujimoto, 1991). Subsequently, they must defend the concept, ensuring that the decisions taken are coherent and harmonious. In addition, they must attend to the matter of making the team members understand the concept, guaranteeing in this way the systematic integrity of the product (Clark and Wheelwright, 1993). Similarly, leaders with vision play a key role when it comes to guaranteeing clear and stable project goals and providing the resources the team needs to reach these goals (Lynn et al., 1999). Having a clear vision and sharing it with the team members, reaching an agreement among all of them and having a strong commitment are some of the recurrent factors in the NPD success (Lynn, 1998). 3. Management skill. The team leader must have enough ability to play the four roles Barczak and Wilemon (1989) identified as fundamental: (1) Communicator. This role consists of transmitting to the team the project goals, the tasks involved in it and the individual responsibilities of team members, apart from facilitating communication during the process. (2) Creator of an adequate climate. The leader must transmit to the team an awareness of the significance and importance of the task in which they are engaged as innovation creators. He must have the ability to integrate individual objectives and needs of the project personnel with the project objectives. In addition, he must have the capacity to create personal enthusiasm for the work and a motivating atmosphere.

4. Technical capacity. Team leaders must maintain and develop their own technical capacity in the work field. Without an understanding of the technology they are employing, they will not gain the confidence of the team members and will not create credibility in the customers (Thamhain and Wilemon, 1977). Nor, however, does the leader need to be an expert in the technology that is being used, since in this case he would run the risk of entering too deeply into technical questions and forgetting his fundamental role as a guide and promoter of the correct operation of the team. As a result of the above, we obtain effective leaders capable of overcoming the functional differentiation, fostering corporate decision taking, concurrently organising the work flow and making better use of the internal intellectual assets. Moreover, they provide objective advice about the emergent questions, interpret needs, balance the different points of view and arbitrate in case of conflict of interests (Topalian, 2000). In the same way, they promote open-mindedness and encourage risk-taking acceptance so that highly creative products can be developed faster and cheaper (Jassawalla and Sashittal, 2000). Effective leaders protect team autonomy (Dougherty, 1990; Ancona and Caldwell, 1992b), break traditional loyalties to specific departments, create a common and unified approach to product innovation and increase the speed of development, at the same time as they reduce cost and increase creativity (Ancona and Caldwell, 1992b). In order to be able to effectively undertake all these functions, leaders must have autonomy to run the project, act with total senior management support and fully understand and identify themselves with the objectives set out for the new product (Barczak and Wilemon, 1992; Cooper and Kleinschmidt, 1994; Murphy and Gorchels, 1996). 35

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

Bearing in mind all the above, team leaders become ``facilitators'' (not heroes) of the NPD process. Some studies have observed that they have a critical effect both on the process performance (that is, the speed and productivity of the development process) and on the superiority of the product (Katz and Allen, 1985; Joyce, 1986; Clark and Fujimoto, 1991; Cooper, 1996; McDonough, 2000). Nevertheless, empirical research that examines the relationship between cross-functional team leadership and NPD process performance is surprisingly little (Gemmill and Wilemon, 1994; Brown and Eisenhardt, 1995). As a consequence, from all the arguments offered we have derived the following hypotheses: H2. The type of leader selected to run the team has an effect on the success of the NPD process. H2a. Firms that use effective team leaders (that is, leaders with power, technical capacity and management skills) achieve a higher level of internal success of the NPD process; that is, they obtain better development times, more efficient processes and superior products. H2b. Firms that use effective team leaders (that is, leaders with power, technical capacity and management skills) achieve a higher level of external success of the NPD process; that is, they are characterised by a higher percentage of new products that are successful in the market, a higher frequency of introduction of new products and a higher level of customer satisfaction.

.

.

the information obtained in previous research works performed at the University of Oviedo (Spain) regarding the capacity for innovation of Spanish firms; and the different works that classify the industries in terms of their innovative effort, such as the studies by Phillips (1966), Lafuente et al. (1985) and Buesa and Molero (1989, 1992).

Thus, taking into account all the available information and following the standard industrial classification (SIC), the code most frequently used at an international level, as representative of the innovative activity in Spain, the following industries were selected: food (SIC 20 and 21), chemicals and plastics (SIC 28 and 30), iron and steel industry (SIC 33 and 34), machinery (SIC 35), electrical and electronic machinery (SIC 36) and transport equipment (SIC 37). 2.2. Field work This research project began in February 2000 with the design of a questionnaire that allowed the attainment of the necessary information to test the hypotheses formulated. For this purpose, the questionnaire was fashioned with a series of items that the existing literature on NPD allowed us to consider as a representative and adequate instrument for measuring the NPD process in the firms analysed. In this work we have used only those questionnaire items that allow the measurement of the crossfunctionality level of the NPD teams used, the type of leaders selected in each case, as well as the success achieved in the NPD process[2]. In order to facilitate the maximum response to the questionnaire, it was considered fitting that most of the questions included be closed, offering a five-point Likert scale for the selection of an option. This method also facilitates the later tabulation and joint processing of the data. In order to check the validity of the designed questionnaire, in March 2000, experts in NPD processes and surveys were consulted and a pre-test was performed on a small sample of firms in varying industries and of various sizes, using in-depth interviews. In April 2000, the definitive questionnaire was mailed together with a covering letter to the 1,269 manufacturing firms making up the target population. Of these firms, 89

2. Research methodology 2.1. Selection of the target population The database used to carry out the proposed research is made up of the information obtained from a mail survey aimed at firms belonging to the most innovative manufacturing industries in Spain and during financial period 2000, according to the list drawn up by Dun & Bradstreet, having over 25 workers. Consequently, the target population was composed of 1,269 firms. The selection of the industries considered as most innovative was carried out considering: 36

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

and, therefore, are more willing to report on their situation than those in a worse condition. On the other hand, the internal validity of the study requires that requested information be obtained from the appropriate source. For this reason, we selected the person in charge of NPD or R&D (technical engineering or similar) in the firm to reply to the survey. In addition, in the covering letter it was explicitly recommended that, if the initial addressee considered it appropriate, he could pass the questionnaire on to a person with greater knowledge and experience of the question at hand. As shown in Table III, the questionnaires were filled in, basically, by the people in charge of the R&D/engineering/ NPD departments or by top management. Likewise, we analysed the characteristics of the respondent managers regarding the number of years they had worked in the firm and the number of years they had carried out their current position. It could be observed that the returned questionnaires were filled in by managers who, on average, had been in the firm for over 14 years and in their current post for over seven years. Therefore, the responsibility and experience of the managers who replied to the questionnaires enable the internal validity of the study to be confirmed; that is to say, the information has been obtained from the most reliable or suitable sources.

communicated that they did not perform NPD processes and so were dropped from the target population. In addition, 41 of the surveys were returned because of an incorrect address or closure of business. Owing to this, a real target population of 1,139 firms was considered. In the months of May and June, second and third mailings, respectively, were carried out. Of the total number of surveys sent out, 146 questionnaires were returned. At the end of August, following a detailed review of the returned surveys, faxes and letters were sent out to resolve the shortcomings detected in the replies of the firms that had filled in the questionnaire. In September 2000, the field work was considered complete, obtaining a total of 125 valid surveys. Table I compiles the technical data of the research performed, that is, the target population, geographical scope, time reference, unit of analysis, sample size and error, duration of field work, response rate and profile of the respondent manager. 2.3. External and internal validity of the study The external validity of a study implies that the results can be generalised for the scope of analysis considered, in this case, Spanish innovative manufacturing firms. Table II includes the representativeness of the sample with respect to the target population in terms of industry and size. The attached data reveal that the sample analysed is reasonably representative with regard to the size and activity sector of the target population. Nevertheless, with respect to size, it can be observed that the sample is slightly biased toward large companies. Thus, the very large companies (with 1,000 or more workers) are observed as having a higher response rate. This is logical since the companies with more resources are generally the ones that contribute the ``innovative force''

3. Analysis and results 3.1. Measurement of the success of the NPD process Since this work deals with evaluating the influence of the use of highly cross-functional teams and effective leaders on the success of the NPD process, it is necessary to indicate, first of all, the variables used to measure such success, which has been analysed in a dual dimension: internal (which reflects the level of

Table I Technical data Characteristics

Survey

Target population Geographical/time reference Unit of analysis Sample size Sample error/level of confidence Date of field work Response rate Respondent

Spanish industrial firms of innovative industries Spain/financial year 1999-2000 Innovative business 125 valid surveys 8.4 percent/95.5 percent March-September 2000 12.81 percent NPD manager/R&D manager/general manager

37

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

Table II Representativeness of the sample Population (%) Industries: Chemical products Electrical and electronic machinery Transport equipment Machinery Iron and steel industry Food and tobacco

376 249 214 240 115 75

Total

1,269

Size: Fewer than 100 workers Between 100-249 Between 250-499 Between 500-999 1.000 or more workers

242 501 325 123 78

Total

1,269

These three indicators have been considered both individually and collectively. In this sense, since the first two indicators Table III Respondents' responsibility

(19.1) (39.5) (25.6) (9.7) (6.1) (100)

125 22 39 27 10 18 125

(27.2) (20) (17.6) (15.2) (12) (8) (100) (19.9) (33.6) (23.3) (8.6) (15.5) (100)

3.2. Testing of the first hypothesis: impact of the use of cross-functional teams on the success of the NPD process In order to judge the level of crossfunctionality of the NPD teams used by the firms sampled, we have used the following items:

Number of firms Percentage of firms 60 41 19 3

(100)

34 25 22 19 15 10

(development time and cost) are those that define the process performance, they are analysed jointly. For this purpose, a new variable termed ``process performance'' (PROPER) has been created, which was calculated as an arithmetic mean of both items. Finally, the three indicators are processed together, creating a new variable for this purpose, the result of their arithmetic mean termed ``process effectiveness'' (PROEF)[3]. On the other hand, the external success of the NPD process has been measured using the following items: percentage of new products that are successful in the market (EXSU1), frequency of introduction of new products into the market (EXSU2) and level of customer satisfaction (EXSU3). All these variables have been measured using a Likert (1-5) scale where a value of 1 indicates ``far below the industry average'' and a value of 5 ``far above the industry average''. Just as in the above case, these three indicators have been processed both independently and together. In this sense, the global measurement of external success is given by the arithmetic mean of the three items indicated, obtaining as a result a new variable termed ``overall measure of external success'' (GLOEXSU)[4].

effectiveness reached in the development process) and external (which reflects the success of NPD in the market). The effectiveness reached in the NPD process is determined by the achievement of three objectives (Hayes et al., 1988; Wheelwright and Clark, 1992): (1) minimising the development time of new products; (2) minimising the development cost of new products; and (3) maximising the superiority of new products developed. Consequently, in this research work, the internal success of the NPD processes has been measured by means of the following items: development time (INSU1), development cost (INSU2) and superiority of the product (INSU3). All of these variables have been measured using a Likert (1-5) scale where a value of 1 indicates ``far below the industry average'' and a value of 5 ``far above the industry average''.

R&D/engineering/NPD Top management Marketing/personnel/finances Others

(29.6) (19.6) (16.9) (18.9) (9.1) (5.9)

Sample (%)

48.8 33.3 15.5 2.4

Note: Frequencies and percentages have been calculated on 123 valid cases

38

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

These results allow us to draw relevant conclusions. Thus, taking into account the importance of each of the participants in the process on an individual basis, we can observe that the participation of marketing in the development team (MT5) is significantly correlated with the commercial success of NPD, both with the global indicator (GLOEXSU) and with the frequency of introduction of new products in the market (EXSU2). On the other hand, we observe a significant correlation between the participation of manufacturing in the NPD team (MT3) and the internal success achieved, this being reflected in a better process performance (less development time and costs) and in its greater effectiveness. However, it is relevant to point out that, when we previously described the variables used, it could be observed that the percentage of Spanish firms involving the manufacturing department in their NPD process was not very high. Another significant fact is that the participation of suppliers in the NPD process (MT6) is significantly correlated with the development time of new products. However, it had also been previously observed (Table III) that only a small number of the firms included in the sample involve suppliers in their NPD processes. With the aim of studying the existence of different groups of firms depending on the level of cross-functionality of the NPD teams used, we resorted to the technique of cluster analysis, using all the items previously described. The proximity measure used for this was the squared Euclidean distance of each case to the centroid of the groups. On the other hand, the classification algorithm selected was the k-means[6]. This method requires the researcher to specify the number of clusters he wishes to obtain. For this reason, we applied a hierarchical cluster[7], which enabled us to conclude that two was the optimal number of conglomerates for this analysis. The results of the cluster analysis are shown in Table VI. In order to interpret the different groups obtained, it is necessary to note the scores obtained by each variable considered in each group. In this case, given the nature of the scale that measures the scoring of the different items, the higher a value is, the higher the level of participation of the corresponding function. However, before interpreting the conglomerates obtained, we wanted to

(1) level of participation of R&D in the teams selected to undertake the NPD process (MT1); (2) level of participation of engineering and design in the NPD teams (MT2); (3) level of participation of manufacturing in the NPD teams (MT3); (4) level of participation of finances in the NPD teams (MT4); (5) level of participation of marketing in the NPD teams (MT5); (6) level of participation of suppliers in the NPD teams (MT6); and (7) level of participation of customers in the NPD teams (MT7). All of these have been measured by means of a Likert (1-5) scale, where a value of 1 represents ``a very low level of participation'' and a value of 5 ``a very high level of participation''. The results of the descriptive analysis of all these variables are in Table IV. Starting from these results, we observe that the highest levels of participation correspond to the R&D (92.7 percent), engineering and design (79.8 percent) and marketing (61.7 percent) departments, followed by the manufacturing department (40.4 percent) and, finally, the finance department (14.5 percent). With respect to suppliers and customers, although 54.5 percent of the firms include customers in their NPD teams, only 22.8 percent intensely include suppliers. In turn, Table V shows the existing correlation between each of the items that measure the cross-functionality of the development teams and the indicators of the NPD process's success, having used the Spearman correlation coefficient for this purpose due to the lack of normality of the variables analysed[5]. Table IV Level of participation of the different functions in the NPD teams 1 MT1 MT2 MT3 MT4 MT5 MT6 MT7

1 3 7 4 31 27 5

(0.8) (2.5) (5.6) (3.3) (25.0) (22.0) (4.1)

2 4 9 24 16 43 32 19

(3.3) (7.6) (19.4) (13.0) (34.7) (26.0) (15.4)

3 4 12 43 27 32 36 32

(3.3) (10.1) (34.7) (22.0) (25.8) (29.3) (26.0)

4 38 38 41 42 15 24 52

(30.9) (31.9) (33.1) (34.1) (12.1) (19.5) (42.3)

76 57 9 34 3 4 15

5

Mean

(61.8) (47.9) (7.3) (27.6) (2.4) (3.3) (12.2)

4.4959 4.1513 3.1694 2.3296 3.6992 2.5610 3.4309

Notes: 123, 124, 123, 124, 123, 124 and 123 valid cases, respectively: the number of cases in each category is indicated in rows and the percentage with respect to the total of the sample in parentheses; the last column shows the mean value of each indicator for the analysed sample

39

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

Table V Correlation between the team cross-functionality and the success indicators of NPD

MT1 MT2 MT3 MT4 MT5 MT6 MT7

GLOEXSU

EXSU1

EXSU2

EXSU3

INSU1

INSU2

INSU3

PROPER

PROEF

ns ns ns ns 0.203* ns ns

ns ns ns ns ns ns ns

ns ns ns ns 0.250** ns ns

ns ns ns ns ns ns ns

ns ns ±0.220* ns ns ±0.185* ns

ns ns ns ns ns ns ns

ns ns ns ns ns ns ns

ns ns ±0.209* ns ns ns ns

ns ns ±0.192* ns ns ns ns

Notes: *Confidence level at 95 percent; **Confidence level at 99 percent; ns: statistically non-significant results

normality among the variables analysed, we applied the Mann-Whitney test, obtaining the results shown in Table VIII. The results of this test show that there are significant differences between the two groups of firms identified for the following variables: percentage of new products that are successful in the market (EXSU1), development time (INSU1), process performance (PROPER) and process effectiveness (PROEF). Specifically, if we observe the mean value reached by each of the groups in Table IX for each variable for which significant differences were found, we can conclude that the more crossfunctional the development teams are, the higher the percentage of new products that turn out to be successful in the market. Similarly, we observe that the more crossfunctional the teams are, the lower the development times and costs achieved and, thereby, the higher the process performance. Finally, the cross-functionality of the teams also influences the effectiveness of the process as a whole, so much so that the more cross-functional teams achieve more effective NPD processes.

Table VI Typology of firms in terms of the degree of multifunctionality of the NPD teams

Cluster 1 Cluster 2

MT1

MT2

MT3

MT4

MT5

MT6

MT7

Number

4.41 4.56

3.76 4.54

2.58 3.70

1.68 2.93

3.75 3.64

1.85 3.32

3.12 3.74

59 57

analyse which variables (among those taken as input to carry out the cluster analysis) maintained significant differences with respect to the resultant groups, something which can be observed in the ANOVA test shown in Table VII. The results obtained in this Table confirm the existence of significant differences between the two groups of firms for all the variables considered. If we interpret the conglomerates starting from here, we can observe that Cluster 2, formed by 57 firms, obtains higher scores for all the variables, except for that referring to marketing. Therefore, although the 59 firms in Cluster 1 are characterised by a higher participation of marketing in their teams, we can assert that the firms in Cluster 2 use more cross-functional teams than the firms in Cluster 1. Thus, Cluster 1 has been termed firms that use less multifunctional NPD teams and Cluster 2 firms that use more cross-functional NPD teams. For verifying the existence of significant differences with respect to the NPD success between both groups and due to the lack of

3.3. Testing of the second hypothesis: impact of the use of effective leaders on the success of the NPD process The items used for this analysis were the following: (1) the team leader enjoys great power in the firm, superior to the functional managers (LEAD1);

Table VII Anova of the items measuring the degree of cross-functionality of the NPD teams

MT1 MT2 MT3 MT4 MT5 MT6 MT7

F

Sig.

1.093 18.198 54.843 62.705 0.290 86.958 12.004

0.038 0.000 0.000 0.000 0.021 0.000 0.001

Table VIII Differences with respect to success following the degree of cross-functionality of the NPD teams EXSU1 U of Mann-Whitney Level of significance

40

INSU1

PROPER

PROEF

1330.5 1229.5 1228.5 1191.0 0.045 0.018 0.031 0.025

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

Table IX Degree of cross-functionality of the teams ± mean values for the success of the NPD process

Firms that use less cross-functional teams Firms that use more cross-functional teams

EXSU1

INSU1

PROPER

PROEF

3.3750 3.6780

3.2241 2.8750

3.1930 2.8929

2.9006 2.7030

capacity, we can see that most of the leaders in the sample firms are characterised by sufficiently high technical capacity, while their technological skills (67.4 percent) exceed their marketing skills (48.4 percent). On the other hand, we can observe that in most of the firms analysed the leaders are characterised by a wide vision of business (65.9 percent) and by being responsible for selecting the members that are to be part of the NPD teams they are going to direct (45.5 percent). However, they are not responsible for establishing the reward and incentive systems of the participants in the process (19.5 percent). Finally, we can observe that most of the sample firms choose team leaders with strong management skills, since they foster the autonomy and responsibility of the team (61.7 percent), have the ability to organise and motivate the participants in the process (65 percent), have the ability to interact with the resource sources (61.8 percent), and maintain fluid communication with the external environment of the firm (71.3 percent). Following the same plan as in the previous case, we first analysed the existing correlation between each of the characteristics of the leader considered and the indicators that measure NPD success. In order to identify these correlations, we calculated the Spearman correlation coefficient, obtaining the results presented in Table XI.

(2) the team leader has thorough knowledge of the technology and knowledge used in the project (LEAD2); (3) the team leader has the necessary marketing skills and knows the market (LEAD3); (4) the team leader has a wide vision of the firm's business (LEAD4); (5) the team leader is responsible for selecting the members of the team (LEAD5); (6) the team leader is in charge of establishing an adequate incentive and reward system (LEAD6); (7) the team leader promotes the autonomy and responsibility of the team (LEAD7); (8) the team leader has the ability to organise and motivate people (LEAD8); (9) the team leader has the ability to interact with the resource sources (LEAD9); and (10) the team leader maintains fluid communication with the firm's external environment (LEAD10). All these items were measured using a Likert (1-5) scale, where a value of 1 represents ``total disagreement'' and a value of 5 ``total agreement'' with the leader characteristic indicated in each case. The profiles of the sample for each of the variables indicated are included in Table X. Thus, we can observe that in most firms team leaders have intermediate authority (37.9 percent). With respect to their technical Table X Leader characteristics 1 LEAD1 LEAD2 LEAD3 LEAD4 LEAD5 LEAD6 LEAD7 LEAD8 LEAD9 LEAD10

16 1 1 2 10 36 3 1

2 (12.9) (0.8) (0.8) (1.6) (8.1) (29.3) (2.4) (0.8)

± ±

24 5 18 6 21 36 11 3 13 6

3 (19.4) (4.0) (14.5) (4.9) (17.1) (29.3) (8.9) (2.4) (10.6) (4.9)

47 22 45 34 36 27 33 39 34 29

4 (37.9) (17.7) (36.3) (27.6) (29.3) (22.0) (26.8) (31.7) (27.6) (23.8)

29 62 46 54 38 13 57 61 56 65

5 (23.4) (50.0) (37.1) (43.9) (30.9) (10.6) (46.3) (49.6) (45.5) (53.3)

8 34 14 27 18 11 19 19 20 22

Mean (6.5) (27.4) (11.3) (22.0) (14.6) (8.9) (15.4) (15.4) (16.3) (18.0)

2.9113 3.9919 3.4355 3.7967 3.2683 2.4065 3.6341 3.7642 3.6748 3.8443

Notes: Valid cases: 124, 124, 124, 123, 123, 123, 123, 123, 123 and 122 firms, respectively; the number of cases in each category is indicated in rows and the percentage with respect to the total of the sample in parentheses; the last column shows the mean value of each indicator for the sample analysed

41

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

Table XI Correlation between the characteristics of the team leader and the indicators of NPD success

LEAD1 LEAD2 LEAD3 LEAD4 LEAD5 LEAD6 LEAD7 LEAD8 LEAD9 LEAD10

GLOEXSU

EXSU1

EXSU2

EXSU3

INSU1

INSU2

INSU3

PROPER

PROEF

0.224* ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns ns ns

ns 0.196* ns ns ns ns ns ns ns ns

ns 0.262** ns 0.240** ns ns ns ns 0.227* ns

ns ±0.219* ns ns ns ±0.179* ns ns ns ±0.195*

ns ns ns ns ns ns ns ns ns ns

ns 0.195* ns ns ns ns 0.191* ns ns ns

ns ns ns ns ns ns ns ns ns ns

±0.204* ns ns ns ns ns ns ns ns ns

Notes: *Confidence level at 95 percent; **Confidence level at 99 percent; ns: statistically non-significant results

their NPD teams, we have applied a cluster analysis. Previously, three new variables were created, with the following denominations: power, technical capacity, and management skill. The variable ``power'' is an arithmetic mean of the following items: ``the leader wields great power in the firm'', ``the leader is responsible for selecting team members'' and ``the leader is in charge of establishing an adequate incentive and reward system''. The variable ``technical capacity'' was created as an arithmetic mean of two items: ``the team leader has a thorough comprehension of the technology and knowledge used in the project'' and ``the team leader has the necessary marketing skills and knows the market''. Finally, the variable ``management skill'' is an arithmetic mean of the following items: ``the team leader promotes the autonomy and responsibility of the team'', ``the team leader has the ability to organise and motivate people'', ``the team leader has the ability to interact with the resource sources'' and ``the team leader maintains fluid communication with the external environment of the firm''. The means and standard deviations of the new variables created[8] are in Table XII. In order to detect the optimal number of conglomerates for this analysis, we applied a hierarchical cluster from which two groups were derived. Starting from this result, we applied the cluster analysis using the three variables indicated. The results are indicated in Table XIII. The interpretation of the two conglomerates obtained is carried out noting the scores of the variables studied in each group. It is worth pointing out that, given the nature of the scale with which the variables have been measured, the higher the value, the

As can be observed, the level of power the team leader has (LEAD1) is significantly correlated, from the external dimension, with the global measurement of success (GLOEXSU), and from the internal dimension, with the effectiveness of the process (PROEF). On the other hand, the level of technological knowledge the team leader has (LEAD2) is significantly correlated with the development time (INSU1) and the superiority of the product (INSU3). From the external perspective, this fact is also significantly correlated with the frequency of introduction of new products into the market (EXSU2) and with the level of customer satisfaction (EXSU3). Similarly, the fact that the team leader has a wide vision of business (LEAD4) is significantly correlated with one of the indicators that measures the external success of the NPD, particularly with the level of customer satisfaction (EXSU3). Moreover, it can be observed that the fact that the team leader is in charge of establishing an adequate incentive and reward system (LEAD6) is significantly correlated with the development time of the new products (INSU1). Likewise, the fact that the team leader fosters the autonomy and responsibility of the team (LEAD7) is significantly correlated with the superiority of the product (INSU3). On the other hand, the fact that the leader has the ability to interact with the resource sources (LEAD9) is significantly correlated with the level of customer satisfaction (EXSU3). Finally, the fact that the leader maintains fluid communication with the external environment of the company (LEAD10) is significantly correlated with the obtaining of better development times (INSU1). Second, for detecting a typology of firms according to the type of leader assigned to 42

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

firms detected in the cluster analysis. Specifically, we observed significant differences between them for the following variables: level of customer satisfaction (EXSU3), development time (INSU1) and superiority of the product (INSU3). The calculation of the mean values reached by each of the groups in each of the variables for which there were significant differences (Table XVI) allowed us to conclude that the firms that have more effective leaders to run their NPD teams are characterised by a higher level of customer satisfaction, lower development times and superior products in comparison with firms with less effective leaders.

Table XII Description of the variables that define the type of leader

Power Technical capacity Management skill

N

Mean

Standard deviation

123 124 122

3.0955 3.7137 3.7295

0.8139 0.66 0.6262

Table XIII Typology of firms according to the type of project leader

Cluster 1 Cluster 2

Power

Technical capacity

3.67 2.63

4.22 3.30

Management skill Number 4.18 3.34

56 66

greater the power, the technical capacity or the management skill of the team leader. Prior to the interpretation of the conglomerates, we studied whether there are significant differences for each when it comes to judging each variable included or, on the contrary, whether the conglomerates are determined by only some of the variables. In Table XIV, we present the results of the ANOVA test obtained. When assessing each of the variables considered for the groups of firms detected according to the type of leader that characterises them, significant differences can be observed at a 99 percent confidence level. When we interpret the conglomerates, we can observe that Cluster 1 obtains higher scores than Cluster 2 for all the variables considered (power, technical capacity and management skill). Therefore, Cluster 1 has been termed firms characterised by more effective leaders and Cluster 2 firms characterised by less effective leaders. The application of the Mann-Whitney test, whose results appear in Table XV, allowed us to detect the existence of significant differences regarding the level of success of the NPD process between the two groups of

4. Conclusions and recommendations for management The objective of this work was to analyse the impact of the use of cross-functional teams and effective leaders on the success of the NPD process. Specifically, we attempted to analyse the influence of both practices on the following variables: on the one hand, the variables that measure the effectiveness reached in the NPD process (internal success: development time, development cost and superiority of the product) and, on the other hand, the variables that measure the success of this process in the market (external success: percentage of new products that are successful in the market, frequency of introduction of new products and level of customer satisfaction). For this purpose, we used a sample of 125 firms representative of the most innovative manufacturing industries in Spain. The results obtained show, on the one hand, that the firms that use cross-functional teams obtain better development times, better performance in their projects and, in general, more effective NPD processes. Similarly, these firms achieve a higher percentage of new products that are successful in the market. On the other hand, the firms that use effective leaders to run these teams obtain better development times and superior products.

Table XIV ANOVA of the items that measure the type of project leader

Power Technical capacity Management skill

F

Sig.

82.967 116.604 97.680

0.000 0.000 0.000

Table XV Differences with respect to success considering the type of project leader

U of Mann-Whitney Level of significance

EXSU3

INSU1

INSU3

1235.5 0.001

1399 0.030

1388 0.024

Table XVI Type of project leader ± mean values for success

Firms with more effective leaders Firms with less effective leaders

43

EXSU3

INSU1

INSU3

4.1273 3.7231

2.8545 3.2154

3.7818 3.5313

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

promote a working environment with all the adequate factors to enhance the success of any NPD project under his responsibility.

Largely, this result can justify the fact that the use of this type of leader is also positively associated with the level of customer satisfaction. Consequently, we can conclude that the use of cross-functional teams and the use of effective leaders have a positive impact on the success of the NPD processes. Thus, it seems appropriate to recommend to the people in charge of managing and organising the NPD in their firms that they continually apply both practices all the time. It is necessary that top management seriously consider the advantages of using cross-disciplinary teams of experts that are involved from the beginning of the NPD process and of selecting leaders with power, technical capacity and management skill to effectively direct and conduct such teams throughout the whole process. Nevertheless, it is necessary to warn firms that the simple fact of using cross-functional teams and of selecting leaders that present the characteristics previously mentioned does not guarantee the immediate achievement of positive results; rather, as with any method or practice, it needs time to efficiently adopt the method and learn how to use it. It is not enough to gather all the functional areas that affect the NPD process and call them a team; all these functions must learn to interact and communicate openly, exchanging information and collaborating intimately, thus supporting one another in the achievement of shared goals. Therefore, management must learn to overcome all the barriers and obstacles and adopt policies and procedures that foster this interfunctional cooperation so that the different elements are mutually reinforced and the communication among the diverse functions can be productive, mutually respected and continuous. The firm must be aware of the fact that the successful adoption of cross-functional teams is going to be an interactive process where problems and lessons to be learned will constantly arise. Likewise, management must bear in mind that being a good NPD team leader is an ability that is acquired with time. Therefore, it must be concerned with not only selecting the person with the appropriate characteristics, but, once a good leader has been detected, with providing him with the rewards and incentives that will make him want to continue his career in this area. Management must help him to develop leadership skills that

Notes 1 Two types of leaders may be distinguished: (a) formal team leaders and (b) informal team leaders. The ones described in this section correspond to the first type of leader. The second type corresponds with what in literature are the so-called ``executive champions'' and are considered informal leaders because they normally work outside official roles. These informal leaders are essentially actors outside organisational rules and procedures (Shane, 1994) who strongly believe in the product, push it, fight for it and make sure that it receives serious consideration (Chakrabarti, 1974). Despite the interest in thoroughly studying the relationship between the use of executive champions and the NPD process's success, given that the existing evidence is quite anecdotal and due to the lack of the necessary information in the sample of manufacturers analysed, we were not able to include its study in this work. 2 This work is part of a wider research that analyses the effect of multiple practices of management and organisation of the NPD process on its success in innovative Spanish firms. Therefore, the sevenpage-long questionnaire designed for the research contains numerous items that have not been used in this work. 3 In order to check the reliability of the two scales proposed to measure the internal success of the NPD process (process performance and effectiveness), we went on to calculate the Cronbach alpha coefficients, obtaining a value over 0.55 in both cases, this being generally accepted for exploratory studies. Although the criterion usually considered to denote a strict internal consistency is a coefficient over 0.70, certain authors such as Van de Ven and Ferry (1979) argue that a certain or moderate internal consistency can be considered to exist if the coefficients range between 0.55-0.70. Similarly, in order to check the uni-dimensionality of the scales proposed, the corresponding exploratory factorial analyses were performed. In both cases, results showed that the items included in each of the scales can be summarised in a single factor, which indicates that these items are inter-related. In particular, the factor loadings were over 0.60 and the percentages of explained variance 69.94 percent and 51.17 percent, respectively. The KMO test and the Bartlett sphericity test enabled us, in turn, to check the suitability of the application of the principal components analysis. 4 Just as in the above case, the reliability and unidimensionality of the proposed scale to measure the external success of the NPD process were checked, obtaining the desired results. 5 With the aim of verifying the normality of the individual variables analysed, we applied the Kolgomorov-Smirnov test. The results of this test enabled us to reject the null hypothesis of normality

44

Cross-functionality and leadership

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Sandra Valle and LucõÂa Avella

performance'', Organization Science, Vol. 3, pp. 321-41. Ancona, D.G. and Caldwell, D.F. (1992b), ``Bridging the boundary: external process and performance in organisational teams'', Administrative Science Quarterly, Vol. 37, pp. 634-65. Barczak, G. and Wilemon, D.L. (1989), ``Leadership differences in new product development teams'', Journal of Product Innovation Management, Vol. 6, pp. 259-67. Barczak, G. and Wilemon, D.L. (1991), ``Communication patterns of new product development team leaders'', IEEE Transactions on Engineering Management, Vol. 38 No. 2, pp. 101-9. Barczak, G. and Wilemon, D.L. (1992), ``Successful new product team leaders'', Industrial Marketing Journal, Vol. 21, pp. 61-8. Bisquerra, R. (1987), IntroduccioÂn Conceptual al AnaÂlisis Multivariable. Un Enfoque InformaÂtico con los Paquetes SPSS-X, BMDP, LISREL y SPAD, Vols. I and II, PPU, Barcelona. Brown, S.L. and Eisenhardt, K.M. (1995), ``Product development: past research, present findings, and future directions'', The Academy of Management Review, Vol. 20 No. 2, pp. 343-78. Buesa, M. and Molero, J. (1989), InnovacioÂn Industrial y Dependencia TecnoloÂgica en EspanÄa, Eudema Universidad, Madrid. Buesa, M. and Molero, J. (1992), Patrones de Cambio TecnoloÂgico y PolõÂtica Industrial, Civitas, Madrid. Chakrabarti, A. (1974), ``The role of champion in product innovation'', California Management Review, Vol. 17, pp. 58-62. Choperana, A.M. (1996), ``Fast cycle time: driver of innovation and quality'', Research Technology Management, May-June, pp. 36-40. Clark, K.B. and Fujimoto, T. (1991), Product Development Performance, Harvard Business School Press, Boston, MA. Clark, K.B. and Wheelwright, S.C. (1992), ``Organising and leading heavyweight development teams'', California Management Review, No. 34, pp. 9-28. Clark, K.B. and Wheelwright, S.C. (1993), ``El desarrollo de productos como ventaja competitiva'', HarvardDeusto Business Review, No. 56, pp. 72-84. Cooper, R.G. (1996), ``Overhauling the new product process'', Industrial Marketing Management, Vol. 25, pp. 465-82. Cooper, R.G. (1999), ``From experience: the invisible success factors in product innovation'', Journal of Product Innovation Management, Vol. 16, pp. 115-33. Cooper, R.G. and Kleinschmidt, E.J. (1990), New Products: The Key Factors in Success, American Marketing Association, Chicago, IL. Cooper, R.G. and Kleinschmidt, E.J. (1993), ``Major new products: what distinguishes the winners in the chemical industry?'', Journal of Product Innovation Management, Vol. 10 No. 2, pp. 90-111. Cooper, R.G. and Kleinschmidt, E.J. (1994), ``Determinants of timeliness in product development'', Journal of Product Innovation Management, Vol. 11, pp. 381-96. Donnellon, A. (1993), ``Cross-functional teams in product development: accommodating the structure to the process'', Journal of Product Innovation Management, Vol. 10, pp. 377-92.

with a confidence level of 99 percent. For this reason, in order to measure the correlation among the items under analysis, we used the Spearman correlation coefficient, which is the non-parametric version of the Pearson correlation coefficient and is suitable for measuring the correlations between ordinal variables or with interval data that do not satisfy the assumption of normality. The rest of the variables used in this work were equally analysed, with the result that none meets the assumption of normality. 6 The k-means method is a separate procedure belonging to the methods group of reassignment or iterative partition. Accordingly, at the end of the process, each case is assigned to the cluster with the nearest centroid, keeping the initial assumption that the resultant clusters are separate (Bisquerra, 1987). Data processing by means of this methodology was carried out with the Quick Cluster module of the SPSS statistical package (version 10.0). 7 Cluster analyses occasionally make artificial classifications. In order to avoid this inconvenience, Punj and Stewart (1983) recommend first applying a hierarchical cluster that enables the researcher to determine the number of groups and, later, a re-allocation method (in this case, the k-means method). 8 In order to check the reliability of these scales as indicators of the leader type, we calculated the Cronbach alpha coefficients, obtaining a value higher than 0.70 for the case of ``power'' and ``management skill''. However, for the case of ``technical capacity'' we obtained a value lower than 0.55, which would be generally accepted for exploratory studies. In this respect, it is necessary to point out that the Cronbach alpha coefficient is sensitive to the number of items, that is, the lower the number of items that measure the same reality, the lower the value of the coefficient tends to be. Likewise, in order to check the uni-dimensionality of these scales, we performed the corresponding exploratory factorial analyses, with the result that, in the three cases, the items included in each scale are summarised in a single factor, which indicates that such items are inter-related. In particular, the factor loadings were over 0.7 and the percentages of explained variance 54.137 percent, 57.10 percent and 55.66 percent, respectively. The KMO test and the Bartlett test enabled us, in turn, to check the suitability of the application of the principal components analysis.

References Ancona, D.G. and Caldwell, D.F. (1990a), ``Improving the performance of new product teams'', Research Technology Management, Vol. 33, pp. 25-9. Ancona, D.G. and Caldwell, D.F. (1990b), ``Outward bound: strategies for team survival in an organisation'', Academy of Management Journal, Vol. 33, pp. 334-66. Ancona, D.G. and Caldwell, D.F. (1992a), ``Demography and design: predictors of new product team

45

Cross-functionality and leadership

Sandra Valle and LucõÂa Avella

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Dougherty, D.A. (1990), ``Understanding new markets for new products'', Strategic Management Journal, Vol. 11, pp. 59-79. Dougherty, D.A. (1992), ``Interpretative barriers to successful product innovation in large firms'', Organization Science, Vol. 3, pp. 179-202. Eisenhardt, K.M. and Tabrizi, B.N. (1995), ``Accelerating adaptive processes: product innovation in the global computer industry'', Administrative Science Quarterly, Vol. 40, pp. 84-110. Ford, R.C. and McLaughlin, F.S. (1992), ``Successful project teams: a study of MIS managers'', IEEE Transactions on Engineering Management, Vol. 39, pp. 312-17. Frischer, J. (1993), ``Empowering management in new product development units'', Journal of Product Innovation Management, Vol. 10. pp. 393-401. Frohman, M.A. (1995), ``Do teams . . . but do them right'', Industry Week, Vol. 3, April, pp. 21-4. Gemmill, G. and Wilemon, D. (1994), ``The hidden side of leadership in technical team management'', Research Technology Management, NovemberDecember, pp. 25-32. Griffin, A. (1997a), ``PDMA research on new product development practices: updating trends and benchmarking best practices'', Journal of Product Innovation Management, Vol. 14, pp. 429-58. Griffin, A. (1997b), Drivers of NPD Success: The 1997 PDMA Report, Product Development & Management Association, Chicago, IL. Gupta, A.K. and Wilemon, D.L. (1990), ``Accelerating the development of technology-based new products'', California Management Review, Vol. 32 No. 2, Winter, pp. 24-44. Hayes, R.H., Wheelwright, S.C. and Clark, K.B. (1988), Dynamic Manufacturing, Free Press, New York, NY. Henke, J.W., Krachenberg, A.R. and Lyons, T.F. (1993), ``Cross-functional teams: good concept, poor implementation!'', Journal of Product Innovation Management, Vol. 10, pp. 216-29. Hershock, R.J. and Braun, D.L. (1993), ``Cross-functional teams drive change'', Executive Excellence, Vol. 10 No. 7, pp. 16-17. Jassawalla, A.R. and Sashittal, H.C. (2000), ``Strategies of effective new product team leaders'', California Management Review, Vol. 42 No. 2, Winter, pp. 35-51. Joyce, W. F. (1986), ``Matrix organisations: a social experiment'', Academy of Management Journal, Vol. 3, pp. 536-61. Katz, R. and Allen, T.J. (1985), ``Project performance and the locus of influence in the R&D matrix'', Academy of Management Journal, Vol. 28, pp. 67-87. Lafuente, A., Salas, V. and YaguÈe, M.J. (1985), Productividad, Capital TecnoloÂgico e InvestigacioÂn en la EconomõÂa EspanÄola, MINER, Madrid. Larson, E.W. and Gobeli, D.H. (1988), ``Organising for product development projects'', Journal of Product Innovation Management, Vol. 5 No. 3, pp. 180-90. Lynn, G.S. (1998), ``New product team learning: developing and profiting from your knowledge capital'', California Management Review, Vol. 40 No. 4, pp. 74-93. Lynn, G.S., Skov, R.B. and Abel, K.D. (1999), ``Practices that support team learning and their impact on speed to market and new product success'', Journal of Product Innovation Management, Vol. 16, pp. 439-54.

McDonough, E.F. III (1993), ``Faster new product development: investigating the effects of technology and characteristics of the project leader team'', Journal of Product Innovation Management, Vol. 10, pp. 241-50. McDonough, E.F. III (2000), ``Investigation of factors contributing to the success of cross-functional teams'', Journal of Product Innovation Management, Vol. 17, pp. 221-35. McDonough, E.F. III and Barczak, G. (1991), ``Speeding up new product development: the effects of leadership style and source of technology'', Journal of Product Innovation Management, Vol. 8, pp. 203-11. McDonough, E.F. III and Griffin, A. (1997), ``The impact of organisational tools on new product development efficiency and effectiveness'', Proceedings of the International PDMA Conference, Monterrey, October. McDonough, E.F. III and Griffin, A. (1998), ``Creating systemic capability for consistent high performance new product development'', in Jurgens, U. (Ed.), New Product Development and Manufacturing Networks: Learning from Experiences in Different Industries and Countries, Springer Verlag, New York, NY. Millson, M.R., Raj, S.P. and Wilemon, D. (1992), ``A survey of major approaches for accelerating new product development'', Journal of Product Innovation Management, Vol. 9, pp. 53-69. Murphy, W.H. and Gorchels, L. (1996), ``How to improve product management effectiveness'', Industrial Marketing Management, Vol. 25, pp. 47-58. Norrgren, F. and Schaller, J. (1999), ``Leadership style: its impact on cross-functional product development'', Journal of Product Innovation Management, Vol. 16, pp. 377-84. Page, A.L. (1993), ``Assessing new product development practices and performance: establishing crucial norms'', Journal of Product Innovation Management, Vol. 10 No. 4, pp. 273-90. Phillips, M. (1966), ``Patents, potential competition and technical progress'', American Economic Review, Vol. 2, pp. 13-22. Punj, G. and Stewart, D.W. (1983), ``Cluster analysis in marketing research: review and suggestions for application'', Journal of Marketing Research, Vol. 20, May, pp. 134-48. Rochford, L. and Rudelius, W. (1992), ``How involving more functional areas within a firm affects the new product process'', Journal of Product Innovation Management, Vol. 9, pp. 287-99. Rosenthal, S.R. (1992), Effective Product Design and Development: How to Cut Lead Time and Increase Customer Satisfaction, Business One Irwin, Homewood, IL. Rossy, G.L. (1992), ``Building commitment in project teams'', Project Management Journal, Vol. 23, pp. 5-14. Shane, S.A. (1994), ``Are champions different from nonchampions?'', Journal of Business Venturing, Vol. 9 No. 5, pp. 397-421. Souder, W.E. (1987), Managing New Product Innovations, Lexington Books, Lexington, MA. Takeuchi, H. and Nonaka, I. (1986), ``The new product development game'', Harvard Business Review, No. 64, pp. 137-46.

46

Cross-functionality and leadership

Sandra Valle and LucõÂa Avella

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 32-47

Thamhain, H.J. (1990), ``Managing technologically innovative team efforts toward new product success'', Journal of Product Innovation Management, Vol. 7, pp. 5-18. Thamhain, H.J. and Wilemon, D.L. (1977), ``Leadership, conflict, and project-management effectiveness'', Sloan Management Review, Vol. 19 No. 1, Autumn, pp. 17-29. Topalian, A. (2000), ``The role of innovation leaders in developing long-term products'', International Journal of Innovation Management, Vol. 4 No. 2, Special Issue, pp. 149-71. Towner, S.J. (1994), ``Four ways to accelerate new product development'', Long Range Planning, Vol. 27 No. 2, pp. 57-65.

Van de Ven, A. and Ferry, D. (1979), Measuring and Assessing Organisations, Wiley, New York, NY. Wheelwright, S.C. and Clark, K.B. (1992), Revolutionising Product Development, Free Press, New York, NY. Zangwill, W.L. (1993), Lightning Strategies for Innovation, Lexington Books, New York, NY. Zirger, B.J. and Hartley, J.L. (1996), ``The effect of acceleration techniques on product development time'', IEEE Transactions on Engineering Management, Vol. 43, pp. 143-52. Zirger, B.J. and Maidique, M.A. (1990), ``A model of new product development: an empirical test'', Management Science, Vol. 36 No. 7, pp. 867-83.

47

Introduction

A hierarchical framework of new products development: an example from biotechnology

New products serve as an important part of economies. They drive the long-term growth of organizations, and so the long-term economic welfare of societies. Organizational growth leads to higher levels of employment. For example, a recent report from the US Senate's Joint Economic Committee's biotechnology summit (1999) states that ``In 1998, the industry generated revenues of about $19 billion, spent $10 billion on R&D, and employed about 150,000 highly-skilled workers. Most biotech companies are fairly small, with two-thirds of firms having fewer than 135 employees.'' The market value of corporations is based, to a considerable extent, on their expected growth in earnings. There is a limit to the number and newness of new products that may be developed from existing technologies. The development of new technologies provides a fresh source of growth. So individual corporations are driven to manage growth both from existing products and also by investing in the development of new products. Further, society places a high value on entrepreneurs who build new technologies into products that in turn increase economic welfare and social capital. Not all new products succeed. Some because they do not meet market needs at the price that the market is willing to pay for them, others because of poor planning and execution leading to a mismatch between when and where customers would buy a product and when and where it is made available. In other words, due to management issues. Other products may fail as they are perceived as being detrimental to the environment or to and by a sufficiently large group of people. One of the potentially important new technologies that have emerged recently is genetic engineering that allows the development of a variety of new products, including genetically modified organisms (GMOs). There is considerable controversy about whether products based on this technology should even be allowed to exist in the marketplace, let alone allowed to succeed. And then there are new products that have failed even before they have been conceived either because they have not been thought of, deemed feasible, or because their components were killed even earlier. New products do not just contribute to economic wellbeing. They often make a real

Maria Tereza Alexandre Olivier Furrer and D. Sudharshan The authors Maria Tereza Alexandre is a Visiting Assistant Professor and D. Sudharshan is Professor of Business Administration at the College of Commerce and Business Administration, both at the University of Illinois at Urbana-Champaign, Illinois, USA. Olivier Furrer is an Associate Professor of Strategic Management at Nijmegen School of Management, Nijmegen, The Netherlands.

Keywords New product development, Hierarchy, Biotechnology, Economic conditions

Abstract Many new products are based on new technologies, which may in turn be based on new scientific discoveries. The extant literature on new product development has focused on how a firm may successfully commercialize new products. There is a corporate cost associated with new product failure, which extends beyond the final product-manufacturing corporation to all the parties involved in the supply chain for the failed product. The new product development community has developed frameworks for managing the new product development process to minimize new product failure, notably by incorporating customer preferences into a cross-functional approach to new product design and by creating a set of decision points or stage gates. The focus of these has been on the latter stages of the new product development process. Besides corporate decisions, society and its various institutions play a role in the shaping of new products from knowledge discoveries. Identifies how other participants may indeed influence the development of new products. Permits a more deliberate understanding of the possible impact of aiding or preventing a movement up the development hierarchy and so a clearer understanding of the potential benefits and opportunity costs may arise.

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 1 . 2003 . pp. 48-63 # MCB UP Limited . ISSN 1460-1060 DOI 10.1108/14601060310456328

48

A hierarchical framework of new products development

Maria Tereza Alexandre et al.

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

difference in our quality of life. For example, again quoting from the US Senate's Joint Committee report (1999): ``The biotechnology industry offers immense potential for cures to many diseases as it takes advantage of rapid gains in scientists' understanding of human genomics. About 80 biotechnology drugs and vaccines are already on the market and have helped millions of patients. Hundreds of additional products are being researched, or are currently in clinical trials. Biotech medicines approved for use include products to treat anemia, cystic fibrosis, hemophilia, cancer, and other diseases. In agriculture, biotechnology research is leading to greater yields of higher quality crops at lower costs.'' Since the successful new products are so important for our wellbeing, it is therefore important for us to have a comprehensive picture of framework of the possible pathways between scientific knowledge and its development into new products. Several issues have emerged as being of importance to scholars and practitioners in the area of new product management specifically, and asset management in general. The area of new product management has been focused on responding to the competitive pressures in the market from the pace of introduction of number of new products, and the higher capital requirements for the development of new products based on the newer technologies. Both of these factors require attention to time. The faster a new product is launched, the sooner it will garner revenue and, as future revenue flows are discounted, the earlier a product is launched, the higher will be its present value. With an increasing pace of new product introductions, it is also felt that products will become obsolete faster. Also, because of the complexity of many of the newer products, including GMO-based seeds, there is a feeling that the market can only bear a few variants of a product leading to the concentration of the industry (see Goldsmith, 2001), and to a winner-take-all mindset. To combat the danger of being late firms focus on being faster at converting a product concept into a commercial product (or cycle time). The issues that have received most attention in the literature have been that of reducing the cycle time or the time to market for a new product and reducing the failure rate of new product introductions. Since new

product success is dependent on the communication between and among many departments as well as other commercial partners, much of the focus has also been on studying cross-functional process management tools such as quality function deployment (QFD) and concurrent engineering (Griffin and Hauser, 1993; Hauser and Clausing, 1988) to lead to more successful new products. New product development is fraught with uncertainty. To ensure that new information is incorporated into decisions on investments on a new product under development, the development process has been broken up into stages. The various stages of new product introduction are opportunity identification, design and prototyping, testing, and product introduction. At each stage a go/no go decision is to be made based on the latest information. Cooper (1990, 1994) provides an elaborate study of the stage-gate mapping and decision frameworks used for managing new product development. An important aspect of the recent developments in new product development management is the renewed attention paid to bring the voice of the customer into the various stages of the process. They do not as yet formally provide room for incorporating the perhaps low level murmur of those who may not be customers but yet may feel that they will be profoundly affected by the new products. Such murmurs may develop into a ground swell (e.g. the social movement organizations presented by Reisner (2001)) that may indeed lead to either the success or the failure of the new product under question. The strategy literature has indeed addressed the issue of stakeholder analysis that seeks to achieve the same goal, but it is surprising that it has not yet found itself in new product development processes. In the literature on product design and development (e.g. Ulrich and Eppinger, 1995) there is no formal framework linking technology, or scientific knowledge to possible products or product markets. In the strategic management literature there has been a growing body of literature that calls for properly valuing and leveraging the resources (both tangible and intangible) of a firm, as resources are said to be ineluctably linked to both enduring competitive advantage and rent (e.g. Amit and Schoemaker, 1993; Barney, 1991; Furrer et al., 2001; Grant, 1991; Hall, 1992, 1993; Mahoney and 49

A hierarchical framework of new products development

Maria Tereza Alexandre et al.

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Pandian, 1992). Along with the growth in the focus on resources of a firm has also come the development of using real options value (the value of making a small investment in order to have an opportunity to be involved at a later stage) calculations to properly value the resources of a firm (Amran and Kulatilaka, 1999; Trigeorgis, 1996). Technologies and customers are considered important resources for firms. To determine the option value of a technology it is necessary to understand both the possible options for its application and the probability and value of success of each possible application. The value of a technology is its ability to be used in products that provide customer value and generate value for a firm. A framework that would enable a systematic understanding of the potential applications of a technology as well as the technology needed to fulfill customer needs would aid in determining and managing the value of the technology resources of a firm. The objective of this paper is to sketch out one possible framework that may serve as a platform for further research. The biotechnology industry is a rapidly growing industry. New research developments like the impending full mapping of the human genome are expected to lead to an enormous amount of new product activity that will benefit mankind and create value for firms. A framework for viewing and analyzing the links between science, technology and product markets is likely of considerable importance both to practitioners and scholars. Two major insights from the literature underpin our framework. First is that the problem of linking a technology to product markets may be viewed hierarchically (Day, 1990). Starting from the top, a new product may be viewed as being composed of a bundle of different technologies. Each technology in a bundle fulfills a different function. A technology is selected for membership in a bundle from a set of technologies, called a technology building-block. All the members of a technology building-block can deliver the same functionality but differ otherwise. Second, the probability of a technology being used in a product depends on the probabilities of its being considered first as part of a relevant building-block, and then as part of the relevant bundle (Capon and Glazer, 1987). We postulate that boundaries hold back the consideration of a member of a set

from being considered as a member of another set. Various boundaries, cognitive and otherwise, keep a technology from moving through the various intervening steps to being part of a product and thus define the probabilities of belonging to a set. The possible outcome depends on a bundle being introduced, the marketing context in which it is introduced and the strategy used for its introduction. We can foresee based on the recent experiences with the dotcom revolution that for-profit institutions will arise, each specializing in identifying and clearing out the boundaries that prevent the transition from one stage to the other and ultimately leading to new products. It is quite possible that many of the ``experiments'' (technology product links) not carried out may have had the biggest impact, had they occurred. By more systematically and comprehensively viewing the product creation process, perhaps, more experiments will be carried out. Bender and Westgren (2001) examine the social processes that drive the construction of the market(s) for genetically modified and non-modified crops. Our framework is intended to help to view and interpret the various levels at which social construction can, does and may take place by highlighting the various stages of the product creation process. It is entirely possible that the social processes that emerge at a higher (or closer to the market) level of our hierarchy of stages may be substantially dependent on the discourse and dialogue, and construction shaping done at earlier levels. It requires careful theoretical and empirical work to understand whether indeed later construction processes can be predicted by the semantics, syntax, and the context of discourse at the preceding level. The impact of construction at one level on another may not be unidirectional, but an interplay over time which shapes the emergence of new product markets by helping shape the cognitive boundaries of the feasibility and viability of incorporating an element from a lower level of the hierarchy in an entity at the next level. The paper is organized in the following manner. First, we present a hierarchical process of product emergence and development. Second, we describe the development of a new product as the process of crossing boundaries, and discuss the example of the role of boundaries in the case of the development of the Flavr Savr2 tomato 50

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

decomposing the linkages into stages is new and, we feel, will enable a more detailed understanding of the commercial possibilities for technology use. So, our framework attempts to provide both an integrative (i.e. a bird's eye view of the technologies to new product link), as well as a systematic or stepby-step view. Existing models focus on the process stages of new product development from a firm's perspective. For example, Cooper (1994) describes the stages of ideation, preliminary investigation, detailed investigation (or building a business case), development, testing and validation, full production and marketing launch, and post-implementation review. Each stage is followed by a go/no go decision. Ulrich and Eppinger (1995) divide the product development process into five phases. They are concept development, system-level design, detail design, testing and refinement, and production ramp-up. The concept development phase is further divided into eight steps. These are: identifying customer needs, establishing target specifications for a product based on the prior step, analysis of competitor products, concept generation, concept selection, refinement of specifications, economic analysis, and project planning. The above process of Ulrich and Eppinger may be called a top-down process. They also suggest that a bottom-up process may be used when a firm starts with a technology and builds it into a product. Our framework is complementary to that of Ulrich and Eppinger (1995) in that we provide a view that allows alternative technologies to be more comprehensively identified in attempts to fulfill customer needs in top-down processes and for alternative customer needs and segments to be more comprehensively identified in attempts to find the appropriate opportunities for a technology. It will thus allow for a better calibration of the option value of available technologies or technologies that may be pursued for further development. For the purposes of visualization and analytical decomposition, we are of the view that competition for consumer demand occurs in the space of product variants (i.e. at Level 4 in Figure 1). Each element in the product variant space is an available product variant in the market. Each product is made using a bundle of technologies. The bundle used for a product-variant is chosen by a firm from elements in the technology bundle

and other GMOs. Third, we present two processes to develop new products: a topdown process from product-variants to knowledge, and a bottom-up process from knowledge to product-variant. Finally, we conclude with the description of some implications of the framework. We use examples of GMOs to provide a unifying context throughout the paper. We wish to emphasize that our use of the Flavr Savr2 tomato example is just an illustration and does not describe the application, calibration, or validation of our framework.

A hierarchy of spaces in the process of product emergence Products are made up of many technologies, and products succeed because they fulfill customer needs, are economically feasible and are socially acceptable. Old products are made obsolete by new products, but not all new products succeed, nor are all customer needs well satisfied. Many technologies are developed but are not incorporated into commercially available products, or at least may not be used to their fullest potentials. Therefore, we feel there is value to a representation of new product processes that will allow a broader view of the possibilities for new technologies as well as of marketopportunities and thus allow a better calibration of models calculating the real option value of a technology. Further, other participants, or would-be participants, could look at the knowledge development and anticipate the likely products that may or could emerge. Thus society could take more charge of the new product development process and not just the new technology development process, as is common practice. For example, again quoting from the US Senate report, ``the advances in biotechnology have been made possible by the twin strengths of federally-sponsored medical research carried out by the National Institutes of Health (NIH) and other agencies, and the entrepreneurial leadership of about 1,300 US biotech companies.'' While the idea of visualizing the existing linkages between technologies and products is not exactly a new idea (e.g. morphological analysis was first used by Zwicky (1969) and has also been written about by Tauber (1975), and Myers (1976)), that of 51

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

Figure 1 Five-level hierarchy diagram

space. Thus, each element in the technology bundle space is a bundle of technologies. Each technology bundle is made up of various technology building-blocks, which have entered the bundle based mainly on design knowledge and economic considerations and are themselves elements of the technology building-block space. Some elements of the space of technologies become members of different technology bundles based on their functional capabilities being associated with specific bundles. An element of the technology space emerges as an element of the knowledge space is discovered to have a certain functionality associated with it. The framework (see Figure 1) decomposes the overall process of movement from the discovery of a new knowledge element in knowledge space to its being incorporated into a new product, as the creation of new elements in various intervening analytical spaces. The space in which new products are represented is called product-variant space. These intervening spaces are labeled technology, technology building-block, and technology bundle respectively. We contend

that, if a position in any of the above intervening spaces or the product-variant space is unoccupied, it is because of the existence of boundaries that prevent that position from being occupied by a new element[1]. At the base of the hierarchy is the space of knowledge. The elements of this space are discovered as the result of research efforts. Examples of results of these efforts in biotechnology are the discovery of Mendel's law, the DNA structure by Watson and Crick, or the function of each gene. An example of a conceptual mapping of knowledge space is provided by Pelc (1996, pp. 13, 17). Each technology can be represented as addressing a particular functionality with a certain value of efficiency. These two dimensions are similar to those developed by Miller (1978) and Van Wyk (1996). So, the Technology space consists of all technologies that are represented by these two dimensions. An example of technology is antisense technology that relates antisense compounds to the blocking of specific proteins. The first dimension orders technologies based on their 52

A hierarchical framework of new products development

Maria Tereza Alexandre et al.

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

similarity on functionality. The second orders them based on their similarity on efficiency. The distinction made between knowledge and technology in our paper is the same as that made by Capon and Glazer (1987). They distinguish technology from the general notion of knowledge, and define technology as knowledge intended for a use. When a potential usage is ascribed to an element of the knowledge space, it along with its intended usage or function becomes an element of the technology space. We define technologies that are considered for use in products as technology buildingblocks. This term is similar to the usage by Meyer and Lehnard (1997). Technology building-blocks can be grouped into categories consisting of substitutes. For example, gene therapy and antisense technology are two elements of the same building-block. They may both be used to block the action of genes. Each technology building-block category can be viewed as consisting of a set of technologies that are similar in terms of their functionality and efficiency. A collection of technology building-block elements that can be used jointly is called a technology bundle. A technology bundle will usually contain technologies from several technology building-block categories. Sometimes, more than one building-block from the same category may be present in the same technology bundle. Also, a technology may appear in multiple bundles. Different bundles may use the same building-block. For example, Remicade, used for the treatment of Crohn's disease, and Synargis, used to prevent serious lower respiratory tract disease caused by syncytial virus, both use the building-block monoclonal antibody technology. The functionality and form of the resulting goods produced by using a technology bundle will depend on the production process and the mix of ingredients used. These goods, when combined with the marketing elements of branding, pricing, image, the choice of communications and distribution channels, and other services, are offered to customers. Customers are viewed as distinguishing between the alternatives offered to them in terms of benefits expected and the occasion(s) for which they can be used. Each alternative is referred to as a product-variant. This follows from the work of Haley (1968), Srivastava

et al. (1978), Dickson (1982), Sheth et al. (1991), Urban and Hauser (1993), Sudharshan (1995), and Green and Srinivasan (1978). Next, we provide (see Figure 2) illustrative tomato examples for each level of the framework. At the Product-Variant level, tomatoes provide various benefits sought for by customers. The variants differ on the benefits they are perceived to offer and the usage (salad, cooking, etc.) for which they are perceived to be suited. Several tomato variants are commercially available. MacGregor's, Carmel, Hot House, Pride Max, and Grape brands in Roma, vine ripened, and plum varieties, and packaged in different weight units stand as examples of the various tomato variants that are offered to customers. The production of each of these tomato variants requires a different technology bundle. For example, the technology bundle required for the MacGregor's tomato variant is made up of a specific gene transfer technology (Agrobacterium-mediated transfer), a specific gene expression technology (called ``antisense'' gene technology), a traditional breeding technology, and other building-block technologies. The bundles for different tomato variants may be similar in that they require the inclusion of a technology building-block from the same technology building-block categories. They may differ in that each bundle may have a different building-block technology from the same technology category. For example, the MacGregor's tomato variant, like conventional variants, like the Hot House variant, needs a traditional breeding technology (Kramer and Redenbaugh, 1994). Other building-block categories are growing technology (e.g. hydroponics, in soil), harvesting technology (e.g. hand, mechanical), ripening (e.g. on the vine, artificial or ethylene-based ripening), etc. Some of the various technology building-blocks that make up the gene transfer technology category are Agroinfection technology, Agrobacterium-mediated technology, pollen tube pathway-based technology, etc. While various other gene transfer technologies may exist, only those technologies that practitioners consider for use become technology buildingblocks. It is important to understand that the members of each level of the framework may change over time. New knowledge and new technologies certainly do emerge. New 53

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

Figure 2 Five level hierarchy diagram

technology building-block categories are created over time, some possibly because of the re-categorization or because of the splitting-up of existing categories, and others because of the choice for use of new technologies that create their own categories. It is also entirely possible that a member of a technology building-block category will stop being a technology building-block. This may happen if it is surpassed by another technology because of functionality, or efficiency reasons, or other factors such as regulations and social resistance. An example of a technology that was a building-block technology but ceased to be one is the socalled ``Terminator technology''. This technology was developed by Delta and Pine Land (later acquired by Monsanto) to enable crops to kill their own seeds in the second generation (Crouch, 1998). Because of the social, economic, and environmental implications of this technology, several farmer and consumer associations launched a campaign against this technology. The opposition's campaign was so virulent that the

company announced that it would not pursue the commercial development of this technology. When a knowledge element is included as a technology, or a technology becomes a member of a building-block, or a building block technology becomes part of a bundle, or a bundle is commercialized by a firm, a new entity comes to exist in a particular position in the corresponding space. It is our view that the entity did not exist before because it was held back by one or more boundaries that held it from coming into existence. Therefore, we view the emergence of a new entity as the result of the crossing of one or more boundaries.

Boundary crossing The existence of a product is delimited by technology boundaries, usage boundaries, needs boundaries, political boundaries, regulatory boundaries, etc. Boundaries of any kind within a space or across spaces may 54

A hierarchical framework of new products development

Maria Tereza Alexandre et al.

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

prevent the emergence of a new element. Some of the most important types of boundaries for new products are: scientific, technological, political, regulatory, economic, cultural, social, and cognitive boundaries. Our focus in this paper is on new product and market development through the process by which technology advancements lead to product-variants that when matched with customer needs lead to new markets. One such example is the Flavr Savr2 tomato story[2]. In winter, most of the northern regions of the USA and Europe must rely on tomatoes shipped from the south. To withstand the rigors of shipping, tomatoes must be picked at a stage the growers call ``mature-green.'' Mature-green tomatoes have already absorbed all the vitamins and nutrients from the plant that they can, but have not started to produce the natural ethylene gas that triggers ripening. To give tomatoes their natural red color, an operation called ``degreening'' or ``ripening initiation'' is necessary. This operation involves putting the green tomatoes in ripening rooms where ethylene gas is released. The green tomatoes spend three to four days in the ripening room before they are shipped. However, most consumers find that shippedin winter tomatoes lack the taste and texture of vine-ripened tomatoes. There was a customer need that needed to be fulfilled. To solve this problem, Calgene, Inc., a biotechnology company with headquarters in Davis, California, has developed a tomato with a gene that slows the natural softening process that accompanies ripening. This genetically engineered tomato, called ``Flavr Savr2,'' spends more days on the vine than other tomatoes, resulting in more flavor, yet remains firm enough to be shipped. To develop this tomato, Calgene had first to understand how the ripening process works. It was the first boundary, in the Knowledge space, to be crossed. Pectin, used to make jelly thicken or gel, occurs naturally in many fruits, giving them their firmness. The pectin in ripening tomatoes is degraded by an enzyme called polygalacturonase (PG). As the pectin is destroyed, the cell walls of tomatoes break down and they soften, making them difficult, if not impossible, to ship successfully. Reducing the amount of PG in tomatoes slows cell wall breakdown and produces a firmer fruit for a longer time. Calgene's scientists isolated the PG gene in

tomato plants and converted it into a reverse image of itself called an antisense orientation. The scientists called this ``reverse'' tomato gene the Flavr Savr2 gene and reintroduced it into tomato plants. Once in a tomato plant, with the Flavr Savr2 gene adhering to it, the PG gene cannot give the necessary signals to produce the PG enzyme that destroys pectin. With the specific work completed and a specific usage assigned to it, the isolation of the PG gene and its use for suppressing pectin-destroying signals in tomatoes became a member of the technology space. The scientific boundaries in knowledge space were not the only boundaries that Calgene had to cross to bring the Flavr Savr2 to the market; other inter-level boundaries also had to be crossed before a new productvariant with the PG-pectin technology was available in the market, or emerge in the product-variant space. In 1992, Calgene, Inc. established a wholly owned subsidiary named Calgene Fresh, Inc. to produce, market, and sell high-quality branded fresh produce to the retail grocery and food service markets. Domestic consumption of fresh tomatoes was estimated at about 5 billion lbs a year with an estimated retail value of approximately $3-3.5 billion. Calgene Fresh, Inc. estimated that 85 percent of US households purchase fresh tomatoes each year, with more than 50 million consumers purchasing 3 lbs of fresh tomatoes in a typical month. This level of consumption was occurring despite consumer dissatisfaction with the quality of fresh tomatoes. It was felt that a need existed for a ``fresher'' tomato to be introduced based on Calgene's technology. Calgene had created the technology, but it was up to farmers to decide whether to include the new seed as part of their technology bundle in growing tomatoes for their markets. It is just as important for farmers to be conscious of the impact of the new technology on their market as it is for Calgene. It might be more difficult for farmers to diversify their risk relative to Calgene's ability to do so by working on several technologies. Other boundaries crossed are discussed below: . Resistance to intellectual property protection: in February 1989, Calgene, Inc. was issued a US patent on the use of the tomato polygalacturonase (PG) gene sequence, including the antisense orientation of the gene. In April 1992, the company crossed a legal boundary 55

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

.

of the new tomatoes grown from Flavr Savr2 seeds under the MacGregor's brand in selected Midwestern and California markets. This signaled the emergence of the PG technology into the product-variant space: . Customer acceptance: Many boundaries had been crossed and a new product variant had emerged. However, the story does not end there. The Flavr Savr2 tomato was a good idea, but Calgene Fresh did not have access to the top commercial tomato variants (or cultivars). Therefore, they sold the Flavr Savr2 in limited quantities as a part of a non-premium tomato cultivar at a premium price. Meanwhile, between 1995 and 1997, Calgene was being bought up incrementally by Monsanto. In 1997, Flavr Savr2 production was stopped and Monsanto scientists began moving the genetic ripening trait into premium tomato variants. The company said that, because tomatoes are hybrids, this is a slow process and at the earliest a premium variety Flavr Savr2 tomato could be on the market again in two to three years. So, the bundles were not acceptable to the market, and new bundles would have to be conceived of and then introduced into the product-market space.

between the building-block and bundle spaces when it was issued a broad patent covering the use of the antisense technology in all plants to partially or completely inhibit specific gene expression. However, other legal boundaries emerged as at least two companies challenged Calgene's patent for the Flavr Savr2. FDA and USDA approval: both the FDA and the USDA require a company to conduct rigorous pre-market testing of genetically engineered food products before they become commercially available. Flavr Savr2 underwent more than four years of comprehensive premarket tests that examined its nutritional value, potential toxins, processing and horticultural traits, fungal resistance, softening rate, and other characteristics. At this stage the PG technology could be considered as part of bundles based on which product-variants could be introduced. In addition, Calgene Fresh, Inc. voluntarily submitted its safety data for rigorous review by an external panel of nationally recognized food safety experts. Their studies demonstrated the Flavr Savr2 tomato to be as safe and nutritious as other fresh tomatoes.

Not only should scientific and technology boundaries be crossed for a product to succeed, but also social and psychological boundaries should be crossed for the product to be accepted by the market. Nelson (2001) describes how consumers' risks and opportunities perceptions create such social and psychological boundaries. He shows that perception of dangers and opportunities of GMOs should be carefully managed for GMOs to be accepted by consumers. The importance of social and psychological boundaries can be illustrated by another example. In the case of the Terminator (RAFI, 1999), farmers perceived the dangers of adopting the ``Terminator'' technology as high, and the incremental opportunities resulting from its adoption as low. The boundaries that emerged once Terminator was introduced forced it to be retracted from the market. While the product was physically in the market, social and psychological boundaries kept it from being within the cognitive definitions of the market, and so led to its withdrawal.

In October 1991, Calgene requested the FDA to issue an Advisory Opinion on the status of the Flavr Savr2 tomato as a food. To assure a thorough review of the safety of the new product, in May 1992, the company filed a Petition for Determination with the USDA requesting that the agency determine that the Flavr Savr2 tomato is a non-regulated article under federal law. In October 1992, the USDA determined that the Flavr Savr2 tomato did not present a plant pest risk and therefore need not be regulated. In April 1994, outside experts of the FDA's Food Advisory Committee discussed the agency's evaluation of the Flavr Savr2 tomato in a public meeting. Members of the committee agreed with the FDA's preliminary assessment that all relevant safety questions about the new tomato had been resolved. On May 18, 1994, the FDA announced its findings that the Flavr Savr2 tomato is as safe as tomatoes bred by conventional means, in effect giving Calgene Fresh approval to market its new product. Calgene Fresh immediately began offering limited quantities 56

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

as pests will become resistant to BT and that more toxic pesticides than are being presently used will be needed in the future. Dozens of other farmers have joined with organic growers' organizations, Greenpeace, and other organizations in a lawsuit that seeks to revoke the registration of BT crops with the Environmental Protection Agency (Consumer Reports, 1999). There is also concern that BT farm products might affect other species in unknowing ways leading to a cascading set of environmental problems. Reisner (2001) shows that a wide variety of social movements may potentially be opposed to genetically engineered food. The first group, she presents, are social movement organizations directly concerned with ``natural foods'' or ``natural agriculture'' in all of its forms. But she shows that there are a wide variety of other types of movements that have adopted genetic engineering as an issue. The alternative agriculture movement, the environmental movement, the consumer movement and health movement, peace organizations, labor, human rights, international and nationalist, and animal rights organizations are already joining coalitions to oppose genetic engineering. That is, the usage of genetic engineering in agricultural and food productions has an unusually wide array of organizations opposing it. All these movements create boundaries to the development of genetically engineered food. The fears of the organic growers have been supported by the findings of a study published in Nature by Losey et al. (1999). In this lab study, Losey and his colleagues reported that pollen from BT corn could spell trouble for the Monarch butterflies. In their experiment, they scattered pollen from BT corn on to milk-weed (the butterfly's only food during its larval or caterpillar stage) and noticed that the caterpillars that ate these leaves either died or were stunted (Time, 1999; Consumer Reports, 1999). In the examples above, the technical and regulatory boundaries were successfully crossed, but the social and psychological ones could not be crossed. The importance of social and psychological boundaries is further illustrated by the announcements made on March 18, 1999 by seven large European food retail chains ± among them Migros, Carrefour, Sainsbury's, and Marks & Spencer ± in spite of initial strong, off-the-shelf

In March 1998, Delta and Pine Land Company (a seed company later to be purchased by Monsanto), in collaboration with the USDA, was awarded a patent over the control of plant gene expression. Although the patent was broad and covered many applications, one of these applications favored by the company was a scheme to engineer crops to kill their own seeds in the second generation. With a specific usage assigned to it, the knowledge became a technology. For the Delta and Pine Land Company, this technology had two important features: (1) it prevented the dissemination of GMOs into the environment; and (2) it made it impossible for farmers to save and replant seeds, thus obliging them to buy new seeds every year. The usage was needed to fulfill the needs of Delta and Pine Land Company's seedmanufacturing customers. This lock-up of the farmers pushed the Rural Advancement Foundation International (RAFI) to fight this technology that they nicknamed ``Terminator.'' RAFI's campaign against this technology was so virulent that in October 1999 Monsanto announced that it would not pursue the commercial development of this technology (Goldsmith (2001) discusses the market power of GMO seed producers and describes other techniques used by producers to lock up farmers). Such social and psychological boundaries have also affected the development of other biotech products. For example, biotechnology makes it possible for plants to protect themselves against certain insects. The protection comes from a naturally occurring micro-organism, called Bacillus Thuringiensis (or BT). BT has been used for more than 30 years by home gardeners, organic growers and other farmers. BT's DNA has been genetically engineered directly into corn, potatoes, cotton, and will soon be engineered into soybeans and other crops, to make them resistant to pests. While bundles may be created incorporating these technologies, their emergence as product-variants will depend on corporate strengths in overcoming the social and psychological boundaries. Organic farmers are furious about BT crops, because they fear that the future is being endangered 57

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

performance, that they would not sell any product containing GMOs. Other psychological boundaries that should be crossed by GMO producers are communication boundaries. More and more, governments worried about consumer reaction to GMOs have begun to impose labeling such products as containing GMOs. For example, in Switzerland, a new regulation was passed in January 2000 to oblige producers and distributors to properly label all products in which GMOs constituted more than 1 percent of the total contents. Such regulations will require GMO producers to communicate and convince consumers that their products are safe. That is, cognitive and social boundaries must be crossed. It was forecast that in 2000 the effect of the resistance of European and Asian consumers might force producers to reduce the surface used in 1999 to grow GMO crops by 25 percent. Juanillo (2001) argues for improved communication between the scientific community and the public to assess more accurately the risks of agricultural biotechnology and to reduce social and psychological boundaries to the acceptance of GM food and help governments to take more informed decisions. In Table I, we provide an example list of some of the requests for approval made to the USA for biotechnology products and the speed with which they were or are being approved. In the above, we have provided examples and described new product development as a process of crossing various boundaries within and across levels. There are two canonical ways, top-down or bottom-up, by which new product emergence may be modeled and managed by systematically addressing the boundaries that either block or retard the emergence of new elements. We will next provide a stage-by-stage description of the two processes.

emergence of a new element in the Knowledge space or in the Technology space. The new element is then linked upwards to the Product-Variant space through all the intermediary spaces. A knowledge element may be embodied in several elements of the Product-Variant space. If a firm identifies a good market potential for seedless tomatoes, they will search for an existing technology bundle that could be used to serve this market. If such a bundle is not found (remember that it may exist but may not be found by the particular firm involved, due to the existence of knowledge or cognitive boundaries that prevent it from discovering such a bundle), growers will search for building-blocks that would allow this bundle to be formed. If the appropriate buildingblocks do not exist, a new technology(ies) must be developed. The development of the new technologies required would be based on the combination of existing biology knowledge, agricultural knowledge, and other new knowledge developed from research projects. A bottom-up process would start with the discovery of an element (or elements) in knowledge space. Prior to the experiments conducted by Mendel in the late 1850s and early 1860s, there was a scientific boundary (i.e. lack of knowledge) that would not allow systematic plant hybridization. With advances in knowledge of the structure of DNA, in other words, with the crossing of knowledge boundaries, scientists could investigate the functions of different genes. With the discovery of the uses of such knowledge for gene transfer, gene isolation and the ability to carry out both routinely, new technologies were born. These genetic engineering technologies could now be bundled with technologies from other building-block categories such as growing, harvesting, ripening, etc. Firms could decide to create product-variants from such bundles. Other types of boundaries also play important roles in the example of GMOs. A good example is the FDA approval regulatory boundary that needs to be crossed for bringing drugs and foods to the market in the USA. Other countries have similar organizations for approving such products. In the case of GMOs cultural and social boundaries made consumers reluctant to accept products that were genetically altered. These boundaries

New product emergence: top-down and bottom-up processes The top-down process starts, in its extreme case, with the identification of an open position in the Product Variant space. It leads to successive searches in the Bundle, Building Block, Technology and Knowledge spaces to develop the product-variant to be introduced. The bottom-up process starts with the 58

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

Table I Requests for approval made to the USA for biotechnology products Company

Application Plant

Aventis Bejo DNA Plant Tech Monsanto

99 97 94 97

Mycogen Monsanto

99 97

Agritope AgrEvo

98 95

U. of Saskatchewan AgrEvo AgrEvo

98 98 98

AgrEvo Monsanto Novartis Seeds AgrEvo Pioneer

98 98 98 96 97

Monsanto

97

Monsanto

97

AgrEvo Monsanto AgrEvo

97 97 97

AgrEvo Monsanto

97 97

Bejo Monsanto AgrEvo Monsanto Calgene

97 97 97 97 97

Du Pont AgrEvo Du Pont Monsanto Monsanto

97 96 96 96 96

DeKalb Calgene Monsanto

96 92 96

DeKalb AgrEvo Cornell U. Monsanto Cornell U. Asgrow

96 96 96 95 95 95

Monsanto AgrEvo Agritope

95 95 95

Type of alteration

Rapeseed Chicory Tomato Potato

Herbicide tolerant Agronomic properties Product quality Insect resistant Viral resistant Corn Insect resistant Potato Insect resistant Viral resistant Melon Agronomic properties Corn Agronomic properties Herbicide tolerant Flax Agronomic properties Rice Herbicide tolerant Rapeseed Agronomic properties Herbicide tolerant Soybean Herbicide tolerant Rapeseed Herbicide tolerant Beet Herbicide tolerant Soybean Herbicide tolerant Corn Agronomic properties Herbicide tolerant Potato Insect resistant Viral resistant Potato Insect resistant Viral resistant Beet Herbicide tolerant Tomato Insect resistant Corn Herbicide tolerant Insect resistant Rapeseed Herbicide tolerant Potato Insect resistant Viral resistant Cichorium intybus Agronomic properties Corn Herbicide tolerant Rapeseed Herbicide tolerant Cotton Insect resistant Cotton Herbicide tolerant Insect resistant Soybean Product quality Rapeseed Herbicide tolerant Soybean Product quality Corn Herbicide tolerant Corn Herbicide tolerant Insect resistant Corn Insect resistant Tomato Product quality Corn Herbicide tolerant Insect resistant Corn Insect resistant Soybean Herbicide tolerant Papaya Viral resistant Corn Insect resistant Papaya Viral resistant Squash Viral resistant Viral resistant Viral resistant Potato Insect resistant Soybean Herbicide tolerant Tomato Product quality

59

Effect

Result

Date of result

Bromoxynil tolerant Male sterile Fruit ripening altered Colorado potato beetle resistant PLRV resistant European Corn Borer resistant Colorado potato beetle resistant PLRV resistant Fruit ripening delayed Male sterile Phosphinothricin tolerant Tolerant to soil residues of sulfonylurea Phosphinothricin tolerant Male sterile Phosphinothricin tolerant Phosphinothricin tolerant Glyphosate tolerant Glyphosate tolerant Phosphinothricin tolerant Male sterile Phosphinothricin tolerant Colorado potato beetle resistant PVY resistant Colorado potato beetle resistant PVY resistant Phosphinothricin tolerant Lepidopteran resistant Phosphinothricin tolerant Lepidopteran resistant Phosphinothricin tolerant Colorado potato beetle resistant PLRV resistant Male sterile Glyphosate tolerant Phosphinothricin tolerant Lepidopteran resistant Bromoxynil tolerant Lepidopteran resistant Oil profile altered Phosphinothricin tolerant Oil profile altered Glyphosate tolerant Glyphosate tolerant European Corn Borer resistant European Corn Borer resistant Fruit ripening altered Glyphosate tolerant European Corn Borer resistant European Corn Borer resistant Phosphinothricin tolerant PRSV resistant European Corn Borer resistant PRSV resistant CMV resistant WMV2 resistant ZYMV resistant Colorado potato beetle resistant Phosphinothricin tolerant Fruit ripening altered

Withdrawn Withdrawn Withdrawn Approved

00 00 99 00

Withdrawn Withdrawn

99 99

Withdrawn Approved

99 99

Approved Approved Approved

99 99 99

Approved Approved Approved Approved Approved

98 99 98 98 98

Approved

99

Withdrawn

97

Approved Approved Approved

98 98 98

Approved Approved

98 98

Approved Approved Withdrawn Withdrawn Approved

97 97 97 97 97

Approved Withdrawn Withdrawn Withdrawn Approved

97 97 96 97 97

Approved Approved Withdrawn

97 96 96

Withdrawn Approved Approved Approved Withdrawn Approved

96 96 96 96 96 96

Approved Withdrawn Approved

96 95 96

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

based on the expected dynamics to achieve the goals set for the Product-Variant space. If a new knowledge element emerges then its value can be assessed by systematically following the path of its similar or related knowledge elements across the levels of the product-market hierarchy. A fuller set of alternatives and their associated probabilities generated by the systematic process will allow for a better valuation and decision. By explicitly understanding the links, and the benefits that will be delivered, a manager will have richer information. By comprehending the nature of the boundaries present: . a manager can allocate resources to their crossing; . understand how the market of interest is likely to change over time; and . incorporate the understanding of both gaps in the spaces and likely scenarios in strategic planning.

also need to be crossed for the success of a new product. The bottom-up and top-down processes, described above, are mainly concerned with how a gap at any level is filled using elements from levels below it. So, a new product variant may emerge based on either existing or new bundles. A new product-variant based on a previously known bundle can change the product market landscape to some degree. A new product is likely to emerge as a result of a novel bundling of technologies by combining technologies from categories not combined before, or by incorporating a technology from a newly emerged technology building-block category. The benefits and usage occasions served by new products are also likely to be distinctly different from those being served in the market and thus are likely to lead to new product categories and product markets.

Another implication of this paper is the use of the hierarchical framework to identify the connections between elements of different levels that point to clear changes in the market. A change in the technology space may represent changes in more than one domain of product-variants that use the same technology as a building-block. In some other circumstances, a change in the technology space element that is used in different product-variant domains could also break the link between these product variants if the new technology dominates the old one for one product variant domain and not for the other. Finally, the representation of the boundaries that define the existent set of elements in each space level, and the understanding of the impact of crossing each boundary, may help a firm to identify its technology and new product priorities and thus the direction that its efforts should take. We have presented an integrated framework for the study of the development of new products. Our discussion has used several examples from the biotechnology industry. The framework provides an exposition of the primary level-by-level decomposition of the problem of new product emergence from new knowledge to technology, to technology building-blocks, to technology bundles, to new product variants. We have detailed bottom-up and top-down processes for product-market emergence, and discussed managerial implications using these processes as guidelines.

Conclusion Some of the implications of the framework may be viewed in the context of the topdown and bottom-up processes of emergence that we just discussed. To use a top-down process, a manager is expected to start with identifying a gap in the Product-Variant space. The first question to be asked is ``Why does the gap exist?'' In other the words, what boundaries are protecting that location from the emergence of an element there? Is the gap because of need boundaries? Is it because customers do not perceive the need for a product variant that would fill the identified gap? If so, a latent need has perhaps been identified. A latent need is very likely to be associated with customer perceptual boundaries in product-variant space. It is also very likely to be associated with boundaries at other levels that either exist or have existed till quite recently. If it is anticipated that the identified gap is associated with an adequate market potential, then analysis and resource allocation can be performed and devoted, respectively, to the appropriate spaces of the other levels of the product-market hierarchy. Strategic planning would then involve the understanding and documentation of the locations, strengths and types of boundaries and their dynamics. Then, attempts at crossing boundaries in Building-Block, Technology, or Knowledge spaces could be 60

A hierarchical framework of new products development

Maria Tereza Alexandre et al.

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

The biotechnology industry is relatively young. The associated knowledge discoveries are occurring at a heightened pace. The connections between scientific knowledge and product variants are not well-known. The knowledge elements may yet not have become technology elements, let alone be introduced as product-variants. There is a need for being able to put a value on the various knowledge elements or even technologies to identify priorities for further resource allocation. By better identifying the possibilities associated with each, and by thinking through and developing the associated probabilities, better valuations, through real options analysis, may be possible. The consideration of a technology as a member of a building-block set may open a whole host of bundle and product-variant possibilities. In the biotechnology industry, the value of real options analysis is likely to be high and a need exists for a framework such as the one presented in this paper to make better possible calibration of such analysis and therefore lead to better resource allocation decisions. Several research questions remain to be raised and answered. For example, methodologies for operationalizing the framework (a combination of mental mapping, environmental analysis, multidimensional scaling of similarities/ dissimilarities, and stochastic process modeling) need to be carefully articulated and tested. Theoretical propositions need to be developed for the emergence of the types and positions of new products and new productmarkets. The nature of social organizations that can or should emerge to better direct the development of new products needs to be studied. These organizations may provide the language through social construction with which the entities at each level of the hierarchy are cognitively accessed, addressed, and processed. The interaction between discourses at the various levels needs to be studied. For example, how does the type of approval provided by a regulatory body affect the limit or broaden the possibilities for a new technology? Cutting off the development process too early may lead to a tremendous opportunity cost; on the other hand, not cutting off some developments may lead to tragedy. Natural social processes may sometimes be too slow at recognizing and responding to danger before it is too late ± the

danger either of lost opportunity or of tragedy. We need to study how social knowledge management systems may be built so as to facilitate wise choices in the movement of science to new products. Biological processes contain technologies that perform every function known to man. Biotechnological processes, therefore, have the possibility of affecting every type of product in the marketplace, from food, to transportation, to books, to computing, to energy sources, to entertainment. At the same time, societies may perceive themselves to have less control of the environmental impact of the results of the usage of biotechnologybased products compared with products based on other technologies. In the language of our framework, it would appear that a key part is missing from the bundle necessary for product success. This would be a control technology. To find a control technology, the appropriate building-block space should be searched for and, if no candidate exists there, then efforts should be directed at the appropriate-level technology or basic research to create the necessary building-block technology(ies). It is important to ensure that boundaries be anticipated and managed. Or that they be quickly identified and managed. We go out on a limb, because of our belief in technology, that it is more than likely that the solution to crossing a boundary lies in either adding or subtracting technologies from the technology bundle used to create products for the marketplace. As mentioned in our introduction, new product development is often a long and arduous task. The rewards are high both to individual corporations and to their shareholders, but also to society at large. The consequences may be serious of delaying the development of new products or of the failure of new products because they have important benefits but do not have the necessary technology to combat their ill effects (see Table II). provides a few examples of the time from science to new products. We hope that a proactive management of the boundaries that prevent scientific knowledge from being embedded in products will shorten the time taken as well as lead to products that are successful and are beneficial to consumers. We hope that our framework contributes in some way to a better understanding of the link between science and new products and provides support to human creativity and 61

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

Table II Time from Invention to commercialization of some technologies Silicone

Insulin

Magnetic recording

1899 ± first published paper on organo-silicon 1930 ± first silicones could be produced in laboratory 1946 ± General Electric produced silicone commercially

1920 ± first ideas for extraction of insulin 1922 ± first human patient received insulin 1933 ± production of the full-day insulin supply in one injection

1893 ± magnetic recording was invented 1898 ± magnetic recording was patented 1937 ± first tape recorder was produced

entrepreneurship in improving the quality of life on this planet.

Day, G.S. (1990), Market-Driven Strategy: Process for Creating Value, The Free Press, New York, NY. Dickson, P.R. (1982), ``Person-situation: segmentation's missing link'', Journal of Marketing, Vol. 46, Fall, pp. 56-64. Furrer, O., Sudharshan, D. and Thomas, H. (2001), ``Organizational structure in a global context: the structure-intangible asset portfolio link'', in Contractor, F.J. (Ed.), Valuation of Intangible Assets in Global Operations, Quorum Books, Westport, CT, pp. 334-53. Goldsmith, P.D. (2001), ``Innovation, market power and the welfare of farmers: the economics of genetically modified seeds'', American Behavioral Scientist. Grant, R.M. (1991), ``The resource-based theory of competitive advantage: implications for strategy formulation'', California Management Review, Vol. 33 No. 3, pp. 114-35. Green, P.E. and Srinivasan, V. (1978), ``Conjoint analysis in consumer research: issues and outlook'', Journal of Consumer Research, Vol. 5, September, pp. 103-23. Griffin, A.J. and Hauser, J.R. (1993), ``Patterns of communication among marketing engineering and manufacturing ± a comparison between two new product teams'', Management Science, Vol. 38 No. 3, pp. 360-73. Haley, E.I. (1968), ``Benefit segmentation, a decisionoriented research tool'', Journal of Marketing, Vol. 32, July, pp. 30-5. Hall, R. (1992), ``The strategic analysis of intangible resource'', Strategic Management Journal, Vol. 13 No. 2, pp. 135-44. Hall, R. (1993), ``A framework linking intangible resources and capabilities to sustainable competitive advantage'', Strategic Management Journal, Vol. 14 No. 8, pp. 607-18. Hauser, J.R. and Clausing, D. (1988), ``The house of quality'', Harvard Business Review, May-June, pp. 63-73. Juanillo, N.K. Jr (2001), ``The risks and benefits of agricultural biotechnology: can scientific and public talk meet?'', American Behavioral Scientist. Kramer, M.G. and Redenbaugh, K. (1994), ``Commercialization of a tomato with an antisense polygalacturonase gene: the FlavrSavr2 tomato story'', Euphytica, Vol. 79 No. 3, pp. 293-7. Losey, J.E., Rayor, L.S. and Carter, M.E. (1999), ``Transgenic pollen harms Monarch larvae'', Nature, Vol. 399 No. 6733, p. 214. Mahoney, J.T. and Pandian, J.R. (1992), ``The resourcebased view within the conversation of strategic management'', Strategic Management Journal, Vol. 13 No. 5, pp. 363-80.

Notes 1 Elements in knowledge space may need to be viewed somewhat differently. From a phenomenological perspective, it may be argued that all things knowable already exist, but that boundaries prevent them from being known. The main focus of our paper is on the other spaces. 2 This story is based on the article written by Glenda D. Webber (1994): ``Genetically engineered fruits and vegetables'' that is available on the Internet at the following address: http://biotech.iastate.edu/ biotech_info_series/bio8.html and several other press articles.

References Amit, R. and Schoemaker, P.J.H. (1993), ``Strategic assets and organizational rent'', Strategic Management Journal, Vol. 14 No. 1, pp. 33-46. Amran, M, and Kulatilaka, N. (1999), Real Options: Managing Strategic Investment in an Uncertain World, Harvard Business School Press, Boston, MA. Barney, J.D. (1991), ``Firm resource and sustainable competitive advantage'', Journal of Business, Vol. 17 No. 1, pp. 99-120. Bender, K. and Westgren, R. (2001), ``Constructing the market(s) for genetically modified and non-modified crops'', American Behavioral Scientist. Capon, N. and Glazer, R. (1987), ``Marketing and technology: a strategic coalignment'', Journal of Marketing, Vol. 51, July, pp. 1-14. Consumer Reports (1999), ``Seeds of change'', September, pp. 41-6. Cooper, R.G. (1990), ``Stage-gate systems: a new tool for managing new products'', Business Horizons, Vol. 33 No. 3, pp. 44-54. Cooper, R.G. (1994), Winning at New Products: Accelerating the Process from Idea to Launch, 2nd ed., Addison-Wesley, Reading, MA. Crouch, M.L. (1998), ``How the Terminator terminates: an explanation for the non-scientist of a remarkable patent for killing second generation seeds of crop plants'', occasional paper, The Edmonds Institute, Edmonds, Washington, DC, available at: www.bio.indiana.edu/people/terminator.html.

62

A hierarchical framework of new products development

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 48-63

Maria Tereza Alexandre et al.

Meyer, M.H. and Lehnard, A.P. (1997), The Power of Product Platforms, The Free Press, New York, NY. Miller, J.G. (1978), Living Systems, McGraw-Hill, New York, NY. Myers, J.H. (1976), ``Benefit structure analysis: a new tool for product planning'', Journal of Marketing, Vol. 40 October, pp. 23-32. Nelson, C. (2001), ``Risk perception, behavior and consumer response to genetically modified organisms'', American Behavioral Scientist. Pelc, K.I. (1996), ``Knowledge mapping: a tool for management of technology'', in Gaynor, G.H. (Ed.), Handbook of Technology Management, McGrawHill, New York, NY, pp. 13.1-13.19. Rural Advancement Foundation International (RAFI) (1999), available at: www.rafi.ca Reisner, A. (2001), ``Social movement organizations' reaction to genetical engineering in agriculture'', American Behavioral Scientist. Sheth, J.N., Newman, B.I. and Gross, B.L. (1991), Consumption Values and Market Choices: Theory and Applications, South-Western Publishing Co., Cincinnati, OH. Srivastava, R.K., Shocker, A.D. and Day, G.S. (1978) ``An exploratory study of the influences of usage situation on perceptions of product markets'', in Hunt, H.K. (Ed.), Advances in Consumer Research, Vol. 5, Association for Consumer Research, Ann Arbor, MI, pp. 32-8. Sudharshan, D. (1995), Marketing Strategy: Relationships, Offerings, Timing and Resource Allocation, PrenticeHall, Englewood-Cliffs, NJ. Tauber, E. M. (1975), ``HIT: heuristic ideation technique ± a systematic procedure for new product search'', Journal of Marketing, Vol. 39, January, pp. 67-70. Time (1999), ``Of corn and butterflies'', May 31, pp. 80-1.

Trigeorgis, L. (1996), Real Options: Managerial Flexibility and Strategy in Resource Allocation, MIT Press, Cambridge, MA. Ulrich, K.T. and Eppinger, S.D. (1995), Product Design and Development, McGraw-Hill, New York, NY. United States Senate (1999), a report on the Joint Economic Committee's biotechnology summit, available at: www.senate.gov/~jec/bio_report.htm Urban, G.L. and Hauser, J.R. (1993), Design and Marketing of New Products, Prentice-Hall, Englewood-Cliffs, NJ. Van Wyk, R. (1996), ``Technology analysis: a foundation for technological expertise'', in Gaynor, G.H. (Ed.), Handbook of Technology Management, McGraw-Hill, New York, NY. Webber, G.D. (1994), ``Genetically engineered fruits and vegetables'', Biotechnology Information Series, available at: www.biotech.iastate.edu/ biotech_info_series/bio8.html Zwicky, F. (1969), Discovery, Invention, Research: Through the Morphological Approach, Macmillan, New York, NY.

Further reading Kendall, H.W., Beachy, R., Eisner, T., Gould, F., Herdt, R., Raven, P.H., Schell, J.S. and Swaminathan, M.S. (1997), Bioengineering of Crops: Report of the World Bank Panel on Transgenic Crops, The World Bank, Washington, DC. Kung, S. and Wu, R. (1993), Transgenic Plants, Volume 2: Present Status and Social and Economic Impacts, Academic Press, San Diego, CA. Standard and Poor's (1999), Biotechnology Industry Survey, 1 April.

63

Introduction

Building organisational culture that stimulates creativity and innovation

Post-industrial organisations today are knowledge-based organisations and their success and survival depend on creativity, innovation, discovery and inventiveness. An effective reaction to these demands leads not only to changes, in individuals and their behaviour, but also to innovative changes in organisations to ensure their existence (Read, 1996). It appears that the rate of change is accelerating rapidly as new knowledge, idea generation and global diffusion increase (Chan Kim and Mauborgne, 1999; Senge et al., 1999). Creativity and innovation have a role to play in this change process for survival. The result is that organisations and leaders try to create an institutional framework in which creativity and innovation will be accepted as basic cultural norms in the midst of technological and other changes. Authors like Ahmed (1998), Martell (1989), Pheysey (1993), Robbins (1996) and Schuster (1986) have emphasised the importance of organisational culture in this context. Organisational culture appears to have an influence on the degree to which creativity and innovation are stimulated in an organisation.

E.C. Martins and F. Terblanche

The authors E.C. Martins is Management Consultant in Organisational Diagnostics, Glenvista, Johannesburg, South Africa. F. Terblanche is a Senior Lecturer in the Department of Information Science, University of South Africa, Pretoria, South Africa. Keywords Organisational culture, Creativity, Innovation, Attitudes Abstract The purpose of this article is to present, by means of a model, the determinants of organisational culture which influence creativity and innovation. A literature study showed that a model, based on the open systems theory and the work of Schein, can offer a holistic approach in describing organisational culture. The relationship between creativity, innovation and culture is discussed in this context. Against the background of this model, the determinants of organisational culture were identified. The determinants are strategy, structure, support mechanisms, behaviour that encourages innovation, and open communication. The influence of each determinant on creativity and innovation is discussed. Values, norms and beliefs that play a role in creativity and innovation can either support or inhibit creativity and innovation depending on how they influence individual and group behaviour. This is also explained in the article.

Research problem In some organisations, action is taken to stimulate creativity and innovation. The right steps may have been taken, such as involving personnel in decision making, recruiting and appointing personnel with creativity characteristics, setting standards for work performance and giving regular feedback, but creativity and innovation are hampered in some way. The culture of an organisation may be a contributing factor in the extent to which creativity and innovation occur in an organisation (Johnson, 1996; Judge et al., 1997; Pienaar, 1994; Shaughnessy, 1988; Tesluk et al., 1997; Tushman and O'Reilly, 1997). The current organisational culture and the demands of creativity and innovation may lead to a conflict situation. This leads to the question: What determinants of organisational culture have an influence on stimulating and promoting organisational culture in organisations? This central research question is subdivided into the following more specific research questions:

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 1 . 2003 . pp. 64-74 # MCB UP Limited . ISSN 1460-1060 DOI 10.1108/14601060310456337

64

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche .

.

.

.

.

.

What is the role of organisational culture in organisations? How can the dimensions of organisational culture be described? What is understood by creativity and innovation in organisations? What is the relationship between creativity, innovation and organisational culture? What determinants of organisational culture have an influence on creativity and innovation in organisations? How can a culture supportive of creativity and innovation be built?

behaviour, norms, values, philosophy, rules of the game and feelings all form part of organisational culture (Hellriegel et al., 1998; Smit and Cronje, 1992). Organisational culture forms an integral part of the general functioning of an organisation. A strong culture provides shared values that ensure that everyone in the organisation is on the same track (Robbins, 1996). The role that organisational culture plays in an organisation can be divided into the functions of organisational culture and the influence that organisational culture has on the different processes in the organisation. Furnham and Gunter (1993) summarise the functions of organisational culture as internal integration and coordination. Based on a literature study of the functions of organisational culture, internal integration can be described as the socialising of new members in the organisation, creating the boundaries of the organisation, the feeling of identity among personnel and commitment to the organisation. The coordinating function refers to creating a competitive edge, making sense of the environment in terms of acceptable behaviour and social system stability (which is the social glue that binds the organisation together) (Martins, 2000). Organisational culture offers a shared system of meanings, which forms the basis of communication and mutual understanding. If the organisational culture does not fulfil these functions in a satisfactory way, the culture may significantly reduce the efficiency of an organisation (Furnham and Gunter, 1993). Organisations use different resources and processes to guide behaviour and change. Organisational culture complements rational managerial tools by playing an indirect role in influencing behaviour. Culture epitomises the expressive character of organisations: it is communicated through symbolism, feelings, the meaning behind language, behaviours, physical settings and artifacts. Rational tools and processes like strategic direction, goals, tasks, technology, structure, communication, decision making, cooperation and interpersonal relationships are designed to do things. The expressive practice of culture is more a reflection of a way of saying things (Coffey et al., 1994). An example is the role that organisational culture plays in the mission and goal statements. Organisational culture fills the gaps between what is formally announced and what actually takes place. It is

The purpose of this article is to present, by means of a model, the determinants of organisational culture which influence the degree of creativity and innovation in an organisation.

Method A literature study, which was descriptive in nature, was undertaken. The aim is to describe the phenomena as accurately as possible. Literature in the managerial sciences is used to describe organisational culture, creativity and innovation in organisations. The demands that creativity and innovation place on the culture of an organisation are derived from the literature study.

Organisational culture defined and its role in organisations Organisational culture is defined in many different ways in the literature. Perhaps the most commonly known definition is ``the way we do things around here'' (Lundy and Cowling, 1996). In this research organisational culture is defined as the deeply seated (often subconscious) values and beliefs shared by personnel in an organisation. Organisational culture is manifested in the typical characteristics of the organisation. It therefore refers to a set of basic assumptions that worked so well in the past that they are accepted as valid assumptions within the organisation. These assumptions are maintained in the continuous process of human interaction (which manifests itself in attitudes and behaviour), in other words as the right way in which things are done or problems should be understood in the organisation. The components of routine 65

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

Against this background and the work of Schein (1985), Martins (1987) developed a model to describe organisational culture based on the typical ideal organisation and the importance of leadership in creating an ideal organisational culture. Martins' model is based on the interaction between the organisational sub-systems (goals and values, structural, managerial, technological and psychosociological sub-systems), the two survival functions, namely the external environment (social, industrial and corporate culture) and the internal systems (artifacts, values and basic assumptions), and the dimensions of culture. The dimensions of culture encompass the following (Martins, 1987, 1997): . Mission and vision (determines personnel's understanding of the vision, mission and values of the organisation and how these can be transformed into measurable individual and team goals and objectives). . External environment (determines the degree of focus on external and internal customers and also employees' perception of the effectiveness of community involvement). . Means to achieve objectives (determines the way in which organisational structure and support mechanisms contribute to the effectiveness of the organisation). . Image of the organisation (focuses on the image of the organisation to the outside world and whether it is a sought-after employer). . Management processes (focuses on the way in which management processes take place in the organisation. It includes aspects such as decision making, formulating goals, innovation processes, control processes and communication). . Employee needs and objectives (focuses on the integration of employees' needs and objectives with those of the organisation as perceived by employees/personnel). . Interpersonal relationships (focuses on the relationship between managers and personnel and on the management of conflict). . Leadership (focuses on specific areas that strengthen leadership, as perceived by personnel).

the direction indicator that keeps strategy on track (Martins, 2000).

Model to describe organisational culture in organisations Several models have been developed to describe the relationships between phenomena and variables of organisational culture. Some examples are the model of organisational culture as part of organisation reality developed by Sathe (1985), which focuses on the influence of leadership, organisation systems and personnel on the actual and expected behaviour patterns, the effectiveness thereof for the organisation and the level of personnel satisfaction brought about by these behaviour patterns. The criticism of this model is that it does not examine the influence of external factors on the organisational culture. Schein's (1985) model depicts the levels of organisational culture, namely artifacts, values and basic assumptions and their interaction. Schein's model is criticised for not addressing the active role of assumptions and beliefs in forming and changing organisational culture (Hatch, 1993). Some researchers see organisational culture in organisations against the background of the systems theory developed by Ludwig von Bertalanffy (1950) and adapted by several authors such as Katz and Kahn who initially applied the systems theory to organisations in 1966 (French and Bell, 1995), Kast and Rosenzweig (1985) and Kreitner and Kinicki (1992) for application in the organisational development field. The systems approach offers a holistic approach, but also emphasises the interdependence between the different sub-systems and elements in an organisation, which is regarded as an open system (French and Bell, 1995). The organisation system model explains the interaction between the organisational subsystems (goals, structure, management, technology and psycho-sociology). This complex interaction, which takes place on different levels, between individuals and groups within the organisation, and with other organisations and the external environment, can be seen as the primary determinant of behaviour in the workplace. The patterns of interaction between people, roles, technology and the external environment represent a complex environment which influences behaviour in organisations.

This model is a comprehensive model which encompasses all aspects of an organisation upon which organisational culture can have an influence, and vice versa. This model can therefore be used to describe organisational 66

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

procedures, new to the relevant unit of adoption, designed to significantly benefit the individual, the group, organization or wider society''. It appears that the context in which a new idea, product, service or activity is implemented determines whether it can be regarded as an innovation within that specific context (Martins, 2000). Innovation is often associated with change (Drucker (1985) cited in West and Farr, 1990; Robbins, 1996; Hellriegel et al., 1998). Innovation is regarded as something new which leads to change. However, change cannot always be regarded as innovation since it does not always involve new ideas or does not always lead to improvement in an organisation (CIMA Study Text, 1996; West and Farr, 1990). An example of change that cannot be regarded as an innovation is changing office hours in an exceptionally hot summer season. In the research under discussion innovation can be defined as the implementation of a new and possibly problem-solving idea, practice or material artifact (e.g. a product) which is regarded as new by the relevant unit of adoption and through which change is brought about (Martins, 2000). The concepts of creativity and innovation in the context of this research (determining which determinants of organisational culture influence creativity and innovation) can be illustrated as in Figure 1. According to Figure 1 creativity and innovation can be regarded as overlapping constructs between two stages of the creative process, namely idea generating and implementation.

culture in an organisation and thus be used as background to determine which determinants of organisational culture influence the degree of creativity and innovation in organisations.

Creativity and innovation in organisations The concepts of creativity and innovation are often used interchangeably in the literature. Consequently, it is important to analyse these concepts in the context of this research. Some definitions of creativity focus on the nature of thought processes and intellectual activity used to generate new insights or solutions to problems. Other definitions focus on the personal characteristics and intellectual abilities of individuals, and still others focus on the product with regard to the different qualities and outcomes of creative attempts (Arad et al., 1997; Udwadia, 1990). Creativity as a context-specific evaluation can vary from one group, one organisation and one culture to another and it can also change over time. Evaluating creativity should therefore be considered at the level of a person, organisation, industry, profession and wider (Ford, 1995). In the research under discussion the context of creativity is at the level of the organisation, and the concept of creativity can be defined as the generation of new and useful/ valuable ideas for products, services, processes and procedures by individuals or groups in a specific organisational context. Definitions of innovation found in the literature vary according to the level of analysis which is used. The more macro the approach (e.g. social, cultural), the more varied the definitions seem to be (West and Farr, 1990). Some definitions are general and broad, while others focus on specific innovations like the implementation of an idea for a new product or service. In an organisational environment, examples of innovation are the implementation of ideas for restructuring, or saving of costs, improved communication, new technology for production processes, new organisational structures and new personnel plans or programmes (Kanter (1983) cited in West and Farr, 1990; Robbins, 1996). West and Farr (1990) define innovation as follows: ``the intentional introduction and application within a role, group or organization of ideas, processes, products or

Relationship of creativity and innovation with organisational culture Organisational culture seems to be a critical factor in the success of any organisation. Successful organisations have the capacity to absorb innovation into the organisational culture and management processes (Syrett and Lammiman, 1997; Tushman and O'Reilly, 1997). According to Tushman and O'Reilly (1997) organisational culture lies at the heart of organisation innovation. The basic elements of organisational culture (shared values, beliefs and behaviour expected of members of an organisation) influence creativity and innovation in two ways: 67

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

perceive what is considered valuable and how they should act in the workplace.

Figure 1 Defining creativity and innovation

Organisational culture affects the extent to which creative solutions are encouraged, supported and implemented. A culture supportive of creativity encourages innovative ways of representing problems and finding solutions, regards creativity as both desirable and normal and favours innovators as models to be emulated (Lock and Kirkpatrick, 1995). Against the background of the systems approach which sees organisations as open systems consisting of different sub-systems interacting with one another, Martins (2000) explains the relationship between organisational culture, creativity and innovation as follows. Certain environmental circumstances, strategic approaches, the values and actions of top management, organisational structure and technological cycles can be associated in the following ways with organisational cultures that support creativity and innovation: . External environment (e.g. economy and competitiveness encourage continual changes in products, technology and customer preferences) (Kanter (1988) cited in Tesluk et al., 1997). . Reaction to critical incidents outside and within the organisation, which is reflected in the strategy (e.g. innovation strategy) of the organisation (Robbins, 1997; Schein (1990) cited in Tesluk et al., 1997). . Managers' values and beliefs (e.g. free exchange of information, open questioning, support for change, diversity of beliefs) (Amabile, 1988; Kanter, 1988; King and Anderson (1990) and Woodman et al. (1993) in Tesluk et al., 1997). . The structure of the organisation, which in turn allows management to reach organisational goals (e.g. flexible structure characterised by decentralisation, shared decision making, low to moderate use of formal rules and regulations, broadly defined job responsibilities and flexible authority structure with fewer levels in the hierarchy) (Hellriegel et al., 1998). . Technology, which includes knowledge of individuals and availability of facilities (e.g. computers, Internet) to support the creative and innovative process (Shattow, 1996).

(1) Through socialisation processes in organisations, individuals learn what behaviour is acceptable and how activities should function. Norms develop and are accepted and shared by individuals. In accordance with shared norms, individuals will make assumptions about whether creative and innovative behaviour forms part of the way in which the organisation operates (Chatman (1991) and Louis (1980) both cited in Tesluk et al., 1997). (2) The basic values, assumptions and beliefs become enacted in established forms of behaviours and activity and are reflected as structures, policy, practices, management practices and procedures. These structures and so on impact directly on creativity in the workplace, for example, by providing resource support to pursue the development of new ideas (Tesluk et al., 1997). In this way individuals in organisations come to

The assumptions of personnel in the organisation on how to act and behave within 68

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

which creativity and innovation take place in the organisation. This influence can be divided into five determinants of organisational culture. Each of these determinants is discussed briefly to describe their influence in promoting or hindering creativity and innovation.

the sub-systems context, as explained above, will have an impact on the degree of creativity and innovation in the organisation (Martins, 2000). Based on the explanation of the relationship between organisational culture, creativity and innovation, the question now arises as to which specific determinants of organisational culture have an influence on the degree to which creativity and innovation are encouraged and stimulated in the organisation.

Determinants of organisational culture that support creativity and innovation Based on a literature study it was found that there is little agreement on the type of organisational culture needed to improve creativity and innovation. There also seems to be a paradox in the sense that organisational culture can stimulate or hinder creativity and innovation (Glor, 1997; Tushman and O'Reilly, 1997). Several researchers (Ahmed, 1998; Filipczak, 1997; Judge et al., 1997; NyÈstrom, 1990; O'Reilly, 1989; Pinchot and Pinchot, 1996; Tesluk et al., 1997) have worked on identifying values, norms and assumptions involved in promoting and implementing creativity and innovation. Very few empirical studies, and especially quantitative research, seem to have been done to support the findings of researchers, but several values, norms and beliefs have been identified by researchers such as Judge et al. (1997), NyÈstrom (1990) and O'Reilly (1989) in their empirical research. In order to synthesise the cultural values and norms that influence creativity and innovation, as found in the literature, the following integrated interactive model was created (Martins, 2000). In studying the influence of organisational culture on creativity and innovation, it became clear that the dimensions of Martins' model a of organisational culture (1987, 1997) have a direct bearing on the influence of organisational culture on creativity and innovation. Consequently this model was used as a starting-point in developing a model of the determinants of organisational culture that influence creativity and innovation. Although the newly developed model may illustrate only part of the phenomenon, it offers a startingpoint for improved understanding. The model (Figure 2) shows that the dimensions that describe organisational culture have an influence on the degree to 69

Strategy An innovation strategy is a strategy that promotes the development and implementation of new products and services (Robbins, 1996). Covey (1993) claims that the origin of creativity and innovation lies in a shared vision and mission, which are focused on the future. Furthermore, the vision and mission of a creative and innovative organisation are also customer- and market-oriented, focusing on solving customers' problems among other things (CIMA Study Text, 1996). An example of a vision that emphasises creative and innovative behaviour is: ``Our company will innovate endlessly to create new and valuable products and services and to improve our methods of producing them'' (Lock and Kirkpatrick, 1995). It is also important that employees should understand the vision and mission (which support creativity and innovation) and the gap between the current situation and the vision and mission to be able to act creatively and innovatively. Judge et al. (1997) describe successful innovation as chaos within guidelines; in other words top management prescribes a set of strategic goals, but allows personnel great freedom within the context of the goals. Organisational goals and objectives reflect the priorities and values of organisations and as a result may promote or hinder innovation (Arad et al., 1997). Hall (cited in Arad et al., 1997) found that personal and organisational goals that emphasise quality rather than effectiveness improve the levels of innovation. It appears that reflecting the value of purposefulness in the goals and objectives of organisations has an influence on creativity and innovation. Arad et al. (1997) mentions that, apart from a few research studies, sufficient research about the effects of organisational and individual goals and objectives has not yet been done. Structure Organisational culture has an influence on the organisational structure and operational

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

Figure 2 Influence of organisational culture on creativity and innovation

creativity and innovation. On the other hand, values like rigidity, control, predictability, stability and order (mostly associated with hierarchical structures) will hinder creativity and innovation (Arad et al., 1997). It is especially the values of flexibility as opposed to rigidity, and freedom as opposed to control, that are emphasised in the literature. A high level of responsibility and adaptability also accompanies an organisational structure that allows for flexibility. Examples of flexibility in organisations are to make use of a job rotation programme or to do away with formal and rigid job descriptions. Freedom as a core value in stimulating creativity and innovation is manifested in

systems in an organisation (Armstrong, 1995). The structure seems to emphasise certain values which have an influence on the promotion or restriction of creativity and innovation in organisations. In the innovation literature, much has been written about the structural characteristics of organisations and according to Arad et al. (1997) and the CIMA Study Text (1996) a flat structure, autonomy and work teams will promote innovation, whereas specialisation, formalisation, standardisation and centralisation will inhibit innovation. As regards the influence of organisational culture on a structure that supports creativity and innovation, values like flexibility, freedom and cooperative teamwork will promote 70

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

autonomy, empowerment and decision making. This implies that personnel are free to achieve their goals in an automatic and creative way within guidelines (described as ``chaos within guidelines'' by Judge et al. (1997)). Personnel therefore have the freedom to do their work and determine procedures as they see fit within the guidelines provided. Management should also believe in personnel and encourage them to be more creative by allowing them more freedom, in other words empowering them instead of controlling them (Judge et al., 1997, p. 76). The literature study revealed that the degree to which employees have freedom and authority to participate in decision making in solving problems determines the level of empowerment, which is positively related to the level of creativity and innovation in an organisation (Arad et al., 1997, p. 4). The speed of decision making can also promote or inhibit creativity and innovation. Tushman and O'Reilly (1997, p. 117) claim that cultural norms which lead to quick decision making (e.g. that speed is important and that the work rate is fast) should promote the implementation of innovation. Co-operative teams are identified by some authors as having an influence on the degree to which creativity and innovation take place in organisations. Well-established work teams which allow for diversity and individual talents that complement one another should promote creativity and innovation (Arad et al., 1997; Mumford et al., 1997). Cross-functional teams which encourage social and technical interaction between developers and implementers can improve and promote creativity and innovation. Another important aspect is that team members should be able to trust and respect one another, understand one another's perspectives and style of functioning, solve differences of opinion, communicate effectively, be open to new ideas and question new ideas. Such effective teamwork is partly based on team members' skills and abilities and partly on the shared values within the group (e.g. values about shared trust and solving differences) (Shattow, 1996; Tushman and O'Reilly, 1997).

technology and creative people, are mechanisms that play this role. Behaviour that is rewarded reflects the values of an organisation. If creative behaviour is rewarded, it will become the general, dominant way of behaving (Arad et al., 1997). The problem is that many organisations hope that personnel will think more creatively and take risks, but they are rewarded for well-proven, trusted methods and fault-free work. Personnel should also be rewarded for risk taking, experimenting and generating ideas. Intrinsic rewards like increased autonomy and improved opportunities for personal and professional growth may support the innovation process (Shattow, 1996; Amabile and Gryskiewicz (1987) and Kanter (1983) cited in Arad et al., 1997). It is also important to reward individuals as well as teams (Tushman and O'Reilly, 1997). Management should be sensitive to which methods of reward and recognition will inspire personnel in their specific organisation to be more creative and innovative (Tushman and O'Reilly, 1997). An organisational culture that promotes creativity and innovation should allow employees time to think creatively and experiment (Shattow, 1996). In organisations where creativity and innovation are encouraged, personnel are, for example, allowed to spend 15 percent of their time on generating new ideas and working on their favourite projects. Emphasis on productivity and downsizing, which leads to more pressure on employees to work harder, is not conducive to creativity and innovation in organisation (Filipczak, 1997). Information technology as a support mechanism is an important resource for successful innovation (Shattow, 1996). In organisations where it is part of the culture to use computer technology such as the Internet and intranet to communicate and exchange ideas, the chances of creativity and innovation taking place are improved (Bresnahan, 1997; Khalil, 1996). Recruitment, selection and appointment and maintaining employees are an important part of promoting the culture of and specifically creativity and innovation in an organisation. The values and beliefs of management are reflected in the type of people that are appointed. Apart from personality traits like intelligence, knowledge, risk taking, inquisitiveness and energy, a value like diversity

Support mechanisms Support mechanisms should be present in the culture of an organisation to create an environment that will promote creativity and innovation. The literature study revealed that rewards and recognition and the availability of resources, namely time, information 71

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

balance should be reached in the degree to which risk taking is allowed. This can be achieved by spelling out expected results, assigning the responsibility of monitoring and measuring risk taking to someone in the organisation, creating a tolerant atmosphere in which mistakes are accepted as part of taking the initiative, regarding mistakes as learning experiences, and assuming that there is a fair chance of risks being successful. Research by NyÈstrom (1990) indicates that the most creative and innovative departments in an organisation regard competitiveness as an important aspect of their culture. According to Read (1996, p. 226), competitiveness in organisations has shifted to the creation and assimilation of knowledge. In creating a culture of competitiveness managers should reach out to internal and external knowledge, encourage debating of ideas, create an environment in which constructive conflict will lead to information flow, support projects based on information flow and actively manage the choice of organisational design. Support for change is a value that will influence creativity and innovation positively (Arad et al., 1997; Eyton, 1996; Glor, 1997; Johnson, 1996; Tushman and O'Reilly, 1997). Managers can create a culture that supports change by looking for new and improved ways of working, creating a vision that emphasises change and revealing a positive attitude towards change (Arad et al., 1997; Tushman and O'Reilly, 1997). An example of a culture in which change is supported is to expect personnel, when stating their annual objectives for the year, to indicate how they intend changing their work methods. Tolerance of conflict and handling conflict constructively are values that support creative and innovative behaviour in organisations (Mumford et al., 1997; Robbins, 1997; Judge et al., 1997). When there is conflict between different ideas, perceptions and ways in which information is processed and evaluated, the process of handling conflict should be handled constructively to promote creativity and innovation. Understanding different individual thinking styles and training personnel in the process of constructive confrontation will create a culture supportive of creativity and innovation.

is of utmost importance in the appointment of creative and innovative people. Appointing people of diverse backgrounds should lead to richer ideas and processes that should stimulate creativity and innovation (Bresnahan, 1997; Gardenswartz and Rowe, 1998). Behaviour that encourages innovation Values and norms that encourage innovation manifest themselves in specific behavioural forms that promote or inhibit creativity and innovation. The way in which mistakes are handled in organisations will determine whether personnel feel free to act creatively and innovatively. Mistakes can be ignored, covered up, used to punish someone or perceived as a learning opportunity (Brodtrick, 1997). Tolerance of mistakes is an essential element in the development of an organisational culture that promotes creativity and innovation. Successful organisations reward success and acknowledge or celebrate failures, for example, by creating opportunities to openly discuss and learn from mistakes (Ryan, 1996; Tushman and O'Reilly, 1997). An organisational culture in which personnel are encouraged to generate new ideas, without being harmed, and where the focus is on what is supported instead of on what is not viable, should encourage creativity and innovation (Filipczak, 1997). Fair evaluation of ideas will also support and encourage creativity (Amabile, 1995). Several authors (Arad et al., 1997; Lock and Kirkpatrick, 1995; Samaha, 1996) indicate that an organisational culture that supports a continuous learning orientation should encourage creativity and innovation. By focusing on being inquisitive, encouraging personnel to talk to one another (e.g. to clients within and outside the organisation to learn from them), keeping knowledge and skills up to date and learning creative thinking skills, a learning culture can be created and maintained. Taking risks and experimenting are behaviours that are associated with creativity and innovation. A culture in which too many management controls are applied will inhibit risk taking and consequently creativity and innovation (Judge et al., 1997). The assumption that risks may be taken as long as they do not harm the organisation will not encourage personnel to be creative and innovative by experimenting and taking risks (Filipczack, 1997, p. 37). It is imprrtant that a

Communication An organisational culture that supports open and transparent communication, based on 72

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

It may be concluded that there is a need for empirical research to support the theoretical research findings on the organisational culture determinants that support creativity and innovation in organisations.

trust, will have a positive influence on promoting creativity and innovation (Barret, 1997; Robbins, 1996). Teaching personnel that disagreement is acceptable, since it offers the opportunity to expose paradoxes, conflict and dilemmas, can promote openness in communication. At the same time personnel must feel emotionally safe to be able to act creatively and innovatively and should therefore be able to trust one another, which in turn is promoted by open communication. An open-door communication policy, including open communication between individuals, teams and departments to gain new perspectives, is therefore necessary to create a culture supportive of creativity and innovation (Filipczak, 1997; Frohman and Pascarella, 1990; Samaha, 1996).

References Ahmed, P.K. (1998), ``Culture and climate for innovation'', European Journal of Innovation Management, Vol. 1 No. 1, pp. 30-43. Amabile, T.M. (1995), ``Discovering the unknowable, managing the unmanageable'', in Ford, C.M. and Gioia, D.A. (Eds), Creative Action in Organizations: Ivory Tower Visions & Real World Voices, Sage, London, pp. 77-81. Arad, S., Hanson, M.A. and Schneider, R.J. (1997), ``A framework for the study of relationships between organizational characteristics and organizational innovation'', The Journal of Creative Behavior, Vol. 31 No. 1, pp. 42-58. Armstrong, M. (1995), A Handbook of Personnel Management Practice, 5th ed., Kogan Page, London. Barret, R. (1997), ``Liberating the corporate soul'', HR Focus, Vol. 74 No. 4, pp. 15-16. Bresnahan, J. (1997), ``The elusive muse'', CIO, Vol. 11 No. 2, pp. 50-6. Brodtrick, O. (1997), ``Innovation as reconciliation of competing values'', Optimum, Vol. 27 No. 2, pp. 1-4. Chan Kim, W. and Mauborgne, R. (1999), ``Strategy, value innovation, and the knowledge economy'', Sloan Management Review, Spring, pp. 41-54. CIMA Study Text (1996), Organisational Management and Development, Stage 3, paper 11, 3rd ed., BPP Publishing, London. Coffey, R.E., Cook, C.W. and Hunsaker, P.L. (1994), Management and Organizational Behavior, Irwin, Burr Ridge, IL. Covey, S.R. (1993), ``Innovation at four levels'', Executive Excellence, Vol. 10 No. 9, pp. 3-5. Eyton, R. (1996), ``Making innovation fly'', Business Quarterly, Vol. 61 No. 1, pp. 59-64. Filipczak, B. (1997), ``It takes all kinds: creativity in the workforce'', Training, Vol. 34 No. 5, pp. 32-40. Ford, C.M. (1995), ``Creativity is a mystery: clues from the investigators' notebooks'', in Ford, C.M. and Gioia, D.A. (Eds), Creative Action in Organizations: Ivory Tower Visions & Real World Voices, Sage, London, pp. 12-52. French, W.L. and Bell, C.H. (1995), Organizational Development: Behavioral Science Interventions for Organization Improvement, 5th ed., Prentice-Hall International, Englewood Cliffs, NJ. Frohman, M. and Pascarella, P. (1990), ``Achieving purpose-driven innovation'', Industry Week, No. 239, 19 March, pp. 20-4. Furnham, A. and Gunter, B. (1993), Corporate Assessment: Auditing a Company's Personality, Routledge, London. Gardenswartz, L. and Rowe, A. (1998), ``Why diversity matters'', HR Focus, Vol. 75 No. 7, pp. S1-S3. Glor, E.D. (1997), ``Encouraging public sector innovation'', Optimum, Vol. 27 No. 2, pp. 41-8.

Conclusions In attempting to create a culture supportive of creativity and innovation, it has been found that one of the best approaches to describe organisational culture is based on the open systems approach. This conclusion is based on the fact that it offers a holistic approach that allows the investigation of the interdependence, interaction and interrelationship of the different sub-systems and elements of organisational culture in an organisation. The patterns of interaction between people, roles, technology and the external environment represent a very complex environment. Under these circumstances creativity and innovation can be influenced by several variables. It appears that creativity and innovation will flourish only under the right circumstances in an organisation. The values, norms and beliefs that play a role in creativity and innovation in organisations can either support or inhibit creativity and innovation, depending on how they influence the behaviour of individuals and groups. The model designed in this research highlights the determinants that play a role in promoting creativity and innovation. The way in which these determinants, namely strategy, organisational structure, support mechanisms, behaviour that encourages innovation and communication, operate will either support or inhibit creativity and innovation. It is clear that these determinants overlap and interact with one another, which supports the open systems approach that was followed. 73

Building organisational culture that stimulates creativity and innovation

European Journal of Innovation Management Volume 6 . Number 1 . 2003 . 64-74

E.C. Martins and F. Terblanche

Hatch, M.J. (1993), ``The dynamics of organizational culture'', Academy of Management Review, Vol. 18 No. 4, pp. 657-93. Hellriegel, D., Slocum, J.W. and Woodman, R.W. (1998), Organizational Behavior, 8th ed., South-Western College, Cincinnati, OH. Johnson, M.M. (1996), ``Finding creativity in a technical organization'', Research Technology Management, Vol. 3a No. 5, pp. 9-11. Judge, W.Q., Fryxell, G.E. and Dooley, R.S. (1997), ``The new task of R&D management: creating goaldirected communities for innovation'', California Management Review, Vol. 39 No. 3, pp. 72-85. Kast, F.E. and Rosenzweig, J.E. (1985), Organization and Management: A Systems and Contingency Approach, 4th ed., McGraw-Hill, New York, NY. Khalil, O.E.M. (1996), ``Innovative work environments: the role of information technology and systems'', SAM Advanced Management Journal, Vol. 61 No. 3, pp. 32-6. Kreitner, R. and Kinicki, A. (1992), Organizational Behavior, 2nd ed., Irwin, Homewood, IL. Lock, E.A. and Kirkpatrick, S.A. (1995), ``Promoting creativity in organizations'', in Ford, C.M. and Gioia, D.A. (Eds), Creative Action in Organizations: Ivory Tower Visions & Real World Voices, Sage, London, pp. 115-20. Lundy, O. and Cowling, A. (1996), Strategic Human Resource Management, Routledge, London. Martell, C. (1989), ``Achieving high performance in library work'', Library Trends, Vol. 38 No. 1, pp. 73-91. Martins, E.C. (2000), ``Die invloed van organisasiekultuur op kreatiwiteit en innovasie in `n universiteitbiblioteek/ The influence of organisational culture on creativity and innovation in a university library'', MInf dissertation, University of South Africa, Pretoria. Martins, N. (1987), ``Organisasiekultuur in `n finansieÈle instelling/Organisational culture in a financial institution'', DPhil thesis, University of Pretoria, Pretoria. Martins, N. (1997), ``Elandsrand Gold-mine: organisational culture survey'', Johannesburg (report, unpublished). Mumford, M.D., Whetzel, D.L. and Reiter-Palman, R. (1997), ``Thinking creatively at work: organization influences on creative problem solving'', The Journal of Creative Behavior, Vol. 31 No. 1, pp. 7-17. NyÈstrom, H. (1990), ``Organizational innovation'', in West, M.A. and Farr, J.L. (Eds), Innovation and Creativity at Work: Psychological and Organizational Strategies, Wiley, Chichester, pp. 143-61. O'Reilly, C. (1989), ``Corporations, culture and commitment: motivation and social control in organizations'', California Management Review, Summer, pp. 9-25. Pheysey, D.C. (1993), Organisational Cultures: Types and Transformations, Routledge, London. Pienaar, H. (1994), ``Die kreatiewe en innoverende universiteits biblioteek/The creative and innovative university library'', DPhil thesis, University of Pretoria, Pretoria. Pinchot, E. and Pinchot, G. (1996), ``Five drivers for innovation'', Executive Excellence, Vol. 13 No. 1, pp. 9-10. Read, W.H. (1996), ``Managing the knowledge-based organization: five principles every manager can

74

use'', Technology Analysis and Strategic Management, Vol. 8 No. 3, pp. 223-32. Robbins, S.P. (1996), Organizational Behavior: Concepts, Controversies, Applications, 7th ed., Prentice-Hall, Englewood Cliffs, NJ. Robbins, S.P. (1997), Essentials of Organizational Behavior, 5th ed., Prentice-Hall, Upper Saddle River, NJ. Ryan, M.J. (1996), ``Driving out fear'', Health-care Forum Journal, Vol. 39 No. 4, pp. 28-32. Samaha, H.E. (1996), ``Overcoming the TQM barrier to innovation'', HR Magazine, Vol. 41 No. 6, pp. 145-9. Sathe, V. (1985), Culture and Related Corporate Realities: Text, Cases and Readings on Organizational Entry, Establishment, and Change, Irwin, Homewood, IL. Schein, E.H. (1985), Organizational Culture and Leadership, Jossey-Bass, San Francisco, CA. Schuster, F.E. (1986), The Proven Connection between People and Profits, Wiley, New York, NY. Senge, P., Kleiner, A., Roberts, C., Ross, R., Roth, G. and Smith, B. (1999), The Dance of Change: The Challenges of Sustaining Information in Learning Organizations, A Fifth Discipline Resource, Nicholas Brearley, London. Shattow, M. (1996), ``Out of the blue'', Electric Perspectives, Vol. 21 No. 3, pp. 44-54. Shaughnessy, T.W. (1988), ``Organizational culture in libraries: some management perspectives'', Journal of Library Administration, Vol. 9 No. 3, pp. 5-10. Smit, P.J. and Cronje, G.J. de J. (1992), Management Principles, Juta, Kenwyn. Syrett, M. and Lammiman, J. (1997), ``The art of conjuring ideas'', Director, Vol. 50 No. 9, pp. 48-54. Tesluk, P.E., Faar, J.L. and Klein, S.R. (1997), ``Influences of organizational culture and climate on individual creativity'', The Journal of Creative Behavior, Vol. 31 No. 1, pp. 21-41. Tushman, M.L. and O'Reilly, C.A. III (1997), Winning through Innovation: A Practical Guide to Leading Organizational Change and Renewal, Harvard Business School Press, Boston, MA. Udwadia, F.E. (1990), ``Creativity and innovation in organizations: two models and managerial implications'', Technological Forecasting and Social Change: An International Journal, Vol. 38 No. 1, pp. 65-80. Von Bertalanffy, L. (1950), ``The theory of open systems in physics and biology'', Science, Vol. 111, pp. 23-9. West, M.A. and Farr, J.L. (1990), ``Innovation at work'', in West, M.A. and Farr, J.L. (Eds), Innovation and Creativity at Work: Psychological and Organizational Strategies, Wiley, Chichester, pp. 3-13.

Further reading Burke, M.E. (1994), ``Creativity circles in information management'', Librarian Career Development, Vol. 2 No. 2, pp. 8-12. Schoenfeldt, L.F. and Jansen, K.J. (1997), ``Methodological requirements for studying creativity in organizations'', The Journal of Creative Behavior, Vol. 31 No. 1, pp. 73-90.