Deviations from Matching in Consumer Choice

Abstract

Consumer researchers have established that most buyers of fast-moving consumer goods such as packed foods practice multi-brand purchasing. Analyses of such products show that most consumers tend to purchase a variety of brands within a product category, selecting among a small “repertoire” of brands rather than being exclusively loyal to a single brand (Ehrenberg, 1988). Research generally shows that in stationary conditions (i.e., the absence of any marked short-term trend in sales) (a) only a few consumers acquire a given brand on consecutive shopping occasions; (b) most consumers buy several different brands, selecting them apparently randomly from a subset or “repertoire” of known, tried and tested brands. At the brand level; (c) each brand attracts only a small percentage of 100%-loyal consumers; (d) brands within a product category tend to differ broadly with respect to their penetration levels but tend to be more similar in terms of their average purchasing frequency; and (e) brands with smaller penetration levels (or market shares) also tend to show smaller average buying frequencies and smaller percentages of 100%-loyal consumers (i.e., the effect known as “Double Jeopardy”). These patterns have been demonstrated for a variety of product categories, from food and drinks to aviation fuel, from personal care products to pharmaceutical prescriptions, for patterns of shopping trips and selection of store chains (Ehrenberg, 1988; Uncles et al., 1995; Goodhardt et al., 1984).