Introduction to the Special Issue on Market Segmentation

Intern. J. of Research in Marketing 19 (2002) 181 – 183 www.elsevier.com/locate/ijresmar Editorial Introduction to the Special Issue on Market Segmentation In market segmentation, one distinguishes homogeneous groups of customers who can be targeted in the same manner because they have similar needs and preferences. In 1956, Smith defined: ‘‘Market segmentation involves viewing a heterogeneous market as a number of smaller homogeneous markets, in response to differing preferences, attributable to the desires of customers for more precise satisfactions of their varying wants.’’ This being an accurate definition to date, one of its most appealing aspects is that it presents segmentation as a conceptual model of the way a manager wishes to view a market. Even if it is a powerful concept, it is still an empirical question as to how well it describes the situation for a particular product or service to provide input to managerial decisions; there are alternatives to segmentation, in particular one-to-one marketing in one extreme and mass marketing in the other. The opportunity to market one-to-one leads potentially, but not necessarily, to greater profitability: oneto-one marketing does not preclude segmentation. When implementing one-to-one strategies, firms currently first develop a limited number of marketing mixes targeted to market segments and then personalize some of their components to each member of these target segments. The available new information technology enables this customization of the marketing mix. Although many companies have developed new business and increased their profits with one-toone marketing, its usage as an implementation tactic does not preclude market segmentation as a general strategy to approach a market. Recently, segmentation and subsequent customization has become very effective in industries where customer retention is a primary goal, so that firms can identify, profile, target and reach segments using their own customer transaction databases. A similar argument can be made for mass marketing. Many companies have successfully implemented mass marketing strategies by targeting consumer populations across the globe with some standardized components of the marketing mix (e.g., product), but with customized implementations of the other components, such as communications and distribution. Nevertheless, taking mass marketing as a starting point for strategy involves risk. Not all companies can afford to market their products globally because of the substantial initial investments required to produce and market at such a global scale or because markets may be heterogeneous at a global scale. Consumers in different countries often have more in common with each other than with other consumers in the same country. Many current global marketing strategies are successful because they identify and tailor to the needs and wants of segments of consumers that cut across national boundaries. Nowadays, companies that operate globally therefore identify and target cross-national segments, developing a global marketing mix where possible and tailoring (part of) its components to cross national market segments. Thus, companies have started to recognize, investigate, and exploit various possible levels of aggregation of their markets, considering their marketing strategies and implementing marketing instruments on a continuum that ranges from aggregate (mass marketing) to disaggregate (one-to-one marketing). Anywhere in between these two extremes lies the identification and targeting of market segments. Distinguishing segments will be profitable whenever demand is heterogeneous, but economies of scale in production, logistics or marketing can be realized. The strategic goals of the firm then determine the requirements for segmentation bases and segmentation methods to be uncovered by marketing research. 0167-8116/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 8 11 6 ( 0 2 ) 0 0 0 7 5 - 7 182 Editorial The distinction of one-to-one marketing versus segmentation is typically linked to the question of whether consumer heterogeneity is better described by a continuous or by a discrete distribution of consumer preferences. The former is associated with finite mixture models, the latter with hierarchical Bayes’ methods (Wedel et al., 1999). Recently, studies have shown that even under conditions that should theoretically favor one of the approaches, the other does surprisingly well, both in terms of recovery of the true parameters and in terms of forecasting hold out observations (Andrews, Ansari, & Currim, 2002), so that neither seems to empirically outperform the other. It has been argued that the underlying assumption of a limited number of segments of individuals that are perfectly homogeneous within segments in finite mixture models is too restrictive (cf. Allenby & Rossi, 1999). Market segmentation would lead to an overly restrictive partition of the continuous distribution into homogeneous segments, while assuming a continuous mixing distribution allows individual level estimates of model parameters to be easily obtained, which is particularly useful to support one-to-one marketing approaches. However, an important issue in the discussion of a continuous versus a discrete distribution of heterogeneity is managerial relevance. In applying models to segmentation problems, one should recognize that every model is at best a workable approximation of reality. One cannot claim that segments really exist or that the distributional form of unobserved heterogeneity (usually assumed to be normal) is known. After all, market segments are not real entities naturally occurring in the marketplace, but groupings created by managers to help them develop strategies that better meet consumer needs at the highest expected profit for the firm. Segmentation has proven to be a very useful concept to managers, even when the final stage of the implementation of the strategy involves one-to-one marketing. Models that approximate market heterogeneity by a number of unobserved segments, in particular mixture models, offer managerial appeal in many applications (see Wedel & Kamakura, 2000, for a review). Managers seem comfortable with the idea of market segments, and current state-of-the-art (mixture) models do a good job of identifying useful groups. However, to enable one-to-one, micro- or direct marketing applications, a continuous approximation of customer heterogeneity such as provided by hierarchical Bayes’ models may be appropriate, possibly in combination with unobserved segments to enable one to target individual customers in the customization stage of strategy. The focus on methodology in the academic literature may mislead one to believe that segmentation is essentially a marketing research problem. Quite to the contrary, market segmentation strategy does not entail a mere market condition to be identified. Segments are not homogeneous groupings of customers naturally occurring in the marketplace, but are determined by the marketing manager’s strategic view of the market. Smith’s (1956) original definition clearly states that. The managers’ perspective determines the way homogeneous groups of potential customers are to be identified by marketing research. The strategic purposes of segmentation determine the bases and methods used in market research; different segments may be identified in the same population of customers in different segmentation studies with different purposes (e.g., new product development, pricing or defining direct marketing targets). Although much progress has been made in the area of models for segmentation, much remains to be done in the conceptualization of strategic market segmentation and in the integration of marketing research and strategy. Many of the issues raised by Wind (1978) remain valid, and building on that, we stated two years ago (Wedel & Kamakura, 2000) that research should be undertaken with the following objectives: (1) The development of better theoretical underpinning of heterogeneity, with the purpose of identifying variables to be included in models and of assisting researchers in appropriate model specification. International market segmentation is an important area to be further explored in that respect. (2) Development of models that integrate segmentation, one-to-one marketing, targeting and positioning and enable empirical validation of the segmentation concept through model comparisons. Monte Carlo comparisons are needed to identify the conditions under which models and estimation methods provide adequate representations of the complex market conditions facing managers. (3) Empirical testing of the predictive validity of segment solutions and the study of the stability of segments over time. An understanding of the dynamic 183 Editorial nature of preferences and market segment composition is essential for strategies focused on the evolution rather than the proliferation of products and businesses. We believe that the articles collected in the current special issue address several of these issues. Steenkamp and Ter Hofstede provide an extensive overview of past work and an outlook on future work in international market segmentation, an important but under researched area. Heilman and Bowman provide a mixture model for segmenting customers based on their behavior in several categories. The article by DeSarbo, Degeratu, Ahearne and Saxton extends finite mixture to the new domain of market share models. Brangule-Vlagsma, Pieters and Wedel show how mixture models can be applied to segment consumers based on their value systems, and investigate the stability of these segments over time. Boone proposes the use of neural networks for market segmentation, and compares them with various other methods. Bock and Uncles provide a framework for the classification of segmentation variables that may assist thinking about segmentation issues and serve as a checklist for selecting segmentation bases. The work presented in this issue underlines the importance of market segmentation as an academic research area. We thank the authors for their interest in our special issue, the reviewers for their time in evaluating the manuscripts and Jan-Benedict Steenkamp and Hubert Gatignon as past and current editors of the International Journal of Research in Marketing for their support. References Allenby, G. M., & Rossi, P. E. (1999). Marketing models of heterogeneity. Journal of Econometrics, 89, 57 – 78. Andrews, R. L., Ansari, A., & Currim, I. S. (2002). Hierarchical Bayes versus finite mixture conjoint analysis models: a comparison of fit, prediction, and partworth recovery. Journal of Marketing Research, 39, 87 – 98 (2002, May). Smith, W. (1956). Product differentiation and market segmentation as alternative marketing strategies. Journal of Marketing, 21, 3 – 8. Wedel, M., & Kamakura, W. A. (2000). Market Segmentation: Conceptual and Methodological Foundations. Dordrecht: Kluwer. Wedel, M., Kamakura, W. A., Arora, N., Bemmaor, A., Chiang, J., Elrod, T., Johnson, R., Lenk, P., Neslin, S., & Poulsen, C. S. (1999). Discrete and continuous representation of heterogeneity. Marketing Letters, 10, 217 – 230. Wind, Y. (1978). Issues and advances in segmentation research. Journal of Marketing Research, 15, 317 – 337. Michel Wedel * Faculty of Economics, University of Groningen, P.O. Box 800, 9700 AV Groningen, The Netherlands School of Business, University of Michigan, Ann Arbor, MI 48109-1234, USA E-mail addresses: [email protected], [email protected] Wagner Kamakura Duke University, Durham, NC, USA * Corresponding author. Faculty of Economics, University of Groningen, P.O. Box 800, 9700 AV Groningen, The Netherlands. Tel.: +31-50-3633735; fax: +31-50-3633720.