| A key challenge facing the retail food industry is how to harness the power of vast amounts of point-of-sale (POS) data to understand consumer behavior and implement electronic commerce with suppliers and business partners. This research develops models that analyze POS data to understand consumer's shopping behavior and determine how to better define customer loyalty.; Data used in the study were obtained from a retail food chain (Retailer A) operating over 40 stores in the U.S. Using Retailer A's customer loyalty database, transaction level data over a six month time frame were obtained for approximately 2,000 households. In addition, demographic, store environment and competitive environment data were collected.; The first research objective provides a framework, suggested by Becker's theory of time allocation and model of household production as it relates to grocery shopping, for examining customer loyalty data to understand what factors describe and influence cardholder shopping patterns and expenditures on foods to cook and foods ready-to-eat. Becker's theory suggests that consumers will substitute ready-to-eat food for food to cook as their incomes rise. Furthermore, as the amount of time spent in the labor force increases, less time will be available for food production activities, resulting in decreased food to cook expenditures. Linear regression models are used to empirically test and support these relationships.; The models developed in the second research objective provide a framework for defining a “loyal” customer and methodology for determining who the most loyal are. Econometric models examined the variables that influence consumer's shopping behavior, having chosen Retailer A. Probability choice models explained what factors influence a household's probability of being classified as loyal by Retailer A. Using four definitions of loyalty—total transactions made, total dollars spent, total margin dollars contributed and a new measure introduced in this study, the ratio of total dollars spent at Retailer A to estimated average annual food at home expenditures—one of the many findings is that the probability of being classified as loyal depends on the definition. Only one variable, realizing more dollar savings, significantly explained the probability of being classified as loyal for all four definitions. |
