Marketing Letters 5:1, (1994): 19-30 © 1994 Kluwer Academic Publishers, Manufactured in the Netherlands. Cross-Category Variation in Customer Satisfaction and Retention EUGENE W. ANDERSON* Assistant Professor of Marketing, School of Business Administration, The University of Michigan, 701 Tappan Street, Ann Arbor, M148104, (313) 763-1566, fax (313) 763-5688 Key words: customer satisfaction, repurchase likelihood, goods vs. services A bstract Perceived quality, expectations, customer satisfaction, and effect of customer satisfaction on re- purchase likelihood are found to be higher for products than for services, but repurchase likelihood for products is lower. Retailers have the highest repurchase likelihood and score lowest on the other variables. A set of relevant category characteristics is used to further understand variation in both the levels of these variables and their relationships. QuMity, expectations, satisfaction, and satisfaction's effect on repurchase are higher - and repurchase likelihood is lower - when compe- tition, differentiation, involvement, or experience is high and when switching eosts, difficulty of standardization, or ease of evaluating quality is low. 1. Introduction Customer satisfaction is widely used in evaluating business performance both in- ternally and externally. Internally, customer satisfaction is used to monitor per- formance, allocate resources, and compensate employees. Externally, customer satisfaction provides information to a wide fange of interest groups, including customers, competitors, investors, and public policy makers. These parties may use customer satisfaction to assess a firm's quality, degree of vutnerability to com- petition, value of intangible customer assets, and contribution to general eco- nomic welfare. However, is it reasonable to directly compare customer satisfaction across cat- egories? For example, what if customer satisfaction is relatively easy to achieve or relatively more important in one category compared to another? Understanding such differences has implications for firms making decisions concerning more than one category (such as evaluating performance of category managers oo de- ciding where to invest in improving quality), for investors and policy makers eval- uating the relative performance of firms and industries, and eren for customers *The author gratefully acknowledges the data provided through the funding of the Swedish Post Office and the support of the National Quality Research Center at the University of Michigan Business School. This research has benefitted from the comments of Claes Fornell, Michael D. Johnson, Donald R. Lehmann, Mary Sullivan, and participants in the Customer Satisfaction Work- shop at the University of Michigan Business School.