Lopo L. Rego, Neil A. Morgan, & Claes Fornell Reexamining the Market Share– Customer Satisfaction Relationship Market share and customer satisfaction are often used to assess marketing performance. Despite the widespread assumption of a positive relationship between these two variables, the limited extant empirical literature on the subject indicates either a negative or a nonsignificant relationship. The authors reexamine this relationship over a longer time period than has previously been possible in a representative sample of U.S. consumer markets and find a consistently significant negative market share–customer satisfaction relationship. This is because customer satisfaction is generally not predictive of firms’ future market share, but market share is a strong negative predictor of firms’ future customer satisfaction. In follow-up analyses, the authors find that a firm’s customer satisfaction can predict its future market share when it is benchmarked against that of its nearest rival and customer switching costs are low. In examining why the market share–future customer satisfaction relationship is generally negative, they find strong support for preference heterogeneity as a key mediator in this relationship. They also show that marketing more brands moderates the negative effect of preference heterogeneity on future customer satisfaction. Thus, larger brand portfolios offer a strategy solution for the general market share–satisfaction trade-off. Keywords: customer satisfaction, market share, marketing performance, empirical generalizations, brand portfolio Lopo L. Rego is Associate Professor of Marketing (e-mail: lrego@indiana. edu), and Neil A. Morgan is PetSmart Distinguished Professor of Market- ing (e-mail: namorgan@indiana.edu), Kelley School of Business, Indiana University. Claes Fornell is Donald C. Cook Distinguished Professor (Emeritus) of Business Administration, Stephen M. Ross School of Busi- ness, University of Michigan (e-mail: cfornell@umich.edu). The authors gratefully acknowledge insightful comments and suggestions from semi- nar participants at Arizona State University, Cardiff University, Case West- ern Reserve University, Indiana University, and Purdue University. They also thank the National Quality Research Center at the University of Michigan for access to the American Customer Satisfaction Index data- base. Ajay Kohli served as area editor for this article. © 2013, American Marketing Association ISSN: 0022-2429 (print), 1547-7185 (electronic) Journal of Marketing Vol. 77 (September 2013), 1–20 1 A s performance outcomes that are intimately con- nected with the firm’s marketing activities, market share and customer satisfaction are central constructs in marketing theory and practice. They are also often viewed as interrelated (e.g., Anderson, Fornell, and Lehman 1994): managers commonly believe that enhancing customer satis- faction is an appropriate strategy for improving market share (e.g., Morgan, Anderson, and Mittal 2005). For exam- ple, the widely cited “service–profit chain” logic suggests that improved customer satisfaction should lead to both enhanced retention of a firm’s existing customers and posi- tive reputation effects that will attract new customers; there- fore, it should be positively related to the firm’s future mar- ket share (e.g., Kamakura et al. 2002). However, the limited empirical evidence to date suggests that firms’ market share and customer satisfaction either are unconnected or have a negative relationship (Anderson, Fornell, and Lehman 1994; Griffin and Hauser 1993). As a result, Fornell (1995) posits a nonpositive association between market share and customer satisfaction as an empirical generalization. Here, we empirically reassess Fornell’s (1995) proposi- tion and answer three primary questions. First, what is the relationship between market share and customer satisfac- tion over time? It is important to examine a longer time series of data because time-varying effects can significantly influence any relationship involving firms and their market- and customer-level performance. Yet the few prior studies of this relationship had access to only one (Fornell 1995) or two (Anderson, Fornell, and Lehmann 1994) years of cus- tomer satisfaction data. Here, using American Customer Satisfaction Index (ACSI) data over a 13-year period, our analyses establish that market share and customer satisfac- tion have a significant and stable negative association over time. This strengthens Fornell’s original nonpositive empiri- cal generalization to negative and shows that although most managers assume that these two key aspects of marketing performance are positively related, under most conditions, this is not the case, and the reverse is true. This finding has important implications for firms’ goal setting and perfor- mance measurement. Second, what is the nature of the relationship between market share and customer satisfaction? Our analyses reveal that the overall negative association between these two variables is a result of a generally weak and insignifi- cant impact of a firm’s current customer satisfaction on its future market share as well as a strong and significant nega- tive impact of a firm’s current market share on its future customer satisfaction. We examine two factors that may influence the effect of current satisfaction on future market share: the firm’s customer satisfaction relative to its nearest rival and customer switching costs. We find that a firm’s customer satisfaction positively predicts its future market share when it is computed relative to that of its nearest rival and customer switching costs are low. This finding has important implications for firms’ customer feedback system