Mei-Wei Huang, Hao-Wei Yang, Ming-Min Lo, Yung-Tai Tang, Hsin-Hung Wu
Supermarkets need to use a data-driven approach to segment customers based on their purchase transactions to meet different customer needs in this highly competitive retail industry in Taiwan. This empirical study combines clustering techniques and RFM model to analyse member customers’ transaction data from a database of a supermarket in Taiwan within a six-week period. The results showed that 5,410 member customers are grouped into loyal, new, and vulnerable customers. A one-way analysis of variance is performed to show these three groups of customers are statistically different. This research further explores the top 10 best-selling merchandise items in both purchase quantity and total money spent. Loyal customers need to focus on five merchandise items. New customers have eight out of ten best-selling merchandise items appeared in both purchase quantity and total money spent. Supermarket management need to pay more attention to these eight items for new customers in this supermarket. Copyright © 2025 Inderscience Enterprises Ltd.
Department of Business Administration, National Changhua University of Education, Changhua, Taiwan; Department of Marketing and Logistics Management, Chaoyang University of Technology, Taichung City, Taiwan; Department of Finance, Chaoyang University of Technology, Taichung City, Taiwan; Department of International Business, Providence University, Taichung City, Taiwan; Faculty of Education, State University of Malang, East Java, Malang, Indonesia
