آیا همیشه بیشتر بهتر است؟اکتشاف اثرات متفاوت عملکرد ادغام بر کارایی در توسعه محصول جدید

The objective of this paper is to explore the relationship between performance in new product development (hereinafter NPD) and functional integration under different conditions of project uncertainty. Functional integration is conceived as a two-dimensional concept, encompassing a behavioural — collaboration — and a structural — interaction — dimension. This study is based on 92 questionnaires looking at NPD activities in more than 40 British and Dutch companies from various industrial sectors. The results suggest that the nature of the relationship between integration and performance is contingent upon the project stage and the degree of novelty in the new product. Integration in the initial stages of the project assumes a prominent role in the quality of the end product, whereas in later stages it is more associated with time to market than with costs and end product quality. Results further show that the dimension collaboration of integration may be more relevant under circumstances of high new product innovativeness than when minor variations are introduced in a new product. The paper ends with a discussion of the use of universal approaches to NPD management.

Numerous studies have shown that functional integration is a critical aspect of today's new product development (NPD) activities (Gupta et al., 1985, Gupta et al., 1986, Souder and Moenaert, 1992 and Song et al., 1997). As far back as 1970, Allen observes that the frequency, the structure of communication networks, and the nature of communicational mechanisms differ sharply between low and high performers in R&D organisations (Allen, 1970). More recently, Calantone et al. (1995) reviewed 500 articles and books on NPD, innovation, and general management, and suggest that the quality of the marketing–R&D interface is one of the current concerns in NPD research. Also in Europe the interface management is regarded as a very important issue for current research (Brockoff and Pearson, 2000). Similarly, Griffin and Hauser (1996) elucidate that communication between marketing and R&D is related to success, regardless of the focus on services or products, and on consumer or industrial markets. This relationship is especially critical when the degree of uncertainty in the environment is high. The central question raised by research on integration is that NPD is a process that requires the capability to obtain, process and interpret large amounts of market, technical, financial and other information, in order to develop product ideas and evaluate their technical soundness, manufacturability and economic feasibility. This requires organisations and individuals to be able to overcome internal differences and barriers built during the process of differentiation (Lawrence and Lorsch, 1967, Dougherty, 1992 and Griffin and Hauser, 1996), and implement an NPD process based on collaboration of structurally separated, yet interdependent, functional units. Conceiving of the innovation process this way means that along with diversity of information and tasks to be performed, there is also diversity of behaviours and attitudes, as projects usually require the contribution of many people with different roles in the process. The problem of integrating tasks and activities becomes as well one of integrating attitudes and behaviours. The objective of the present paper is to explore the relationship between performance in NPD and integration of tasks and activities, on one hand, and of behaviours, on the other. Furthermore, it aims to explore the nature of this relationship under different conditions of project uncertainty. Previous research has examined the bi-dimensional nature of functional integration (e.g. Kahn, 1996), but this has been done at a departmental level. This paper complements existing literature by focusing on the marketing–R&D integration at the project level of analysis. It starts by reviewing definitions of integration and the effect of integration practices on performance. It then presents the method and results of the empirical study and it concludes with discussion and implications for managers and researchers involved in NPD activities.

Despite the low values found for both the correlation and regression analysis, these are comparable to those reported by other authors. For example, the standardised path coefficient estimates between cooperation and performance in the study of Song et al. (1997) vary between 0.22 and 0.38. Pinto and Pinto (1990) report an adjusted R2 of 0.29 between cooperation and task outcomes. And the regression coefficients described by Kahn (1996) vary between 0.01 and 0.14. If on one hand these low values — yet statistically significant — indicate that other variables not included in this study help to explain the variance of performance measures — as indeed suggested by Brown and Eisenhardt (1995) —, they also show that functional integration is associated with time, cost, and quality. Overall, the results of this investigation confirm those of several authors that have looked at the relationship between functional integration and process and product performance (e.g. Dougherty, 1992, Pinto et al., 1993, Moenaert et al., 1994 and Griffin and Hauser, 1996). However, the findings also suggest that integration is a multidimensional construct, which relates differently to outcomes in NPD. Firstly, interaction between marketing and R&D in the initial stages of the NPD process seems to assume a prominent role in the quality of the end product, whereas interaction in later stages appears to be more associated with time to market than with costs and quality of the end product. These results give partial support to Verganti (1997) and Song et al. (1998): in both studies it was found that integration in the concept generation stage contributes to better product effectiveness and product efficiency. Secondly, although collaboration has been advocated as having a more important impact of performance than interaction (Kahn, 1996), the current research has shown opposite results. A possible reason for this disparity might be related to the different operational definitions used in this study and in Kahn's work. For example, Kahn (1996) uses mechanisms of communication for measuring interaction, whereas the current study used NPD activities that are likely to require integration between functional units. Finally, this research provides evidence that the degree of product innovativeness plays an important role in understanding the benefits of intra-organisational collaboration during NPD. As shown, collaboration may be more relevant under circumstances of high new product innovativeness than when minor variations are introduced in a new product. As such, these findings suggest that a universal approach to the management of NPD may not be possible. As argued by Kamoche and Cunha (2000), more than insisting on the benefits of a one best way, researchers should analyse how different contingencies recommend the adoption of different NPD configurations. Moreover, it is anticipated that the traditional sequential models (e.g. Cooper, 1988 and Cooper, 1990) may be useful for ‘routine innovations’, but not for the development of new products with a higher degree of innovativeness. The data further shows that interaction may be beneficial for less innovative new products, while collaboration may be necessary while developing highly new products. This can be explained by the type of activities involved in both kinds of projects: more structure can be used in the first case; less structure and a need for sense making require collaboration in the latter. Therefore, this research suggests that the challenges posed by different types of products may be fundamental not only for understanding the paths for NPD, but also for designing NPD activities which adapt to the type of product. With regard to this, existing product typologies (e.g. Pinto and Covin, 1989 and Coombs et al., 1998) could be used to increase knowledge in the field. Recent research (e.g. Brown and Eisenhardt, 1995 and Brown and Eisenhardt, 1997) proposes that unconventional development practices such as extensive communication, loose structuring, and fluid job descriptions, are fundamental ingredients in NPD projects, which confirms that more collaboration may be necessary when a new path must be discovered, while more interaction may be recommended for developing minor innovations. These results fundamentally reflect the divergence between exploration and exploitation, and confirm that more is better if such a divergence is considered.