Chun-Ju Chang, Ming-Min Lo, Hao-Wei Yang, Hsin-Hung Wu
In the retail industry, cluster analysis has been commonly used in analysing customer behaviours. This study utilises the database of customer transaction data from a supermarket in Taiwan in the first half of 2021. Twenty-two merchandise items are defined as the most frequently purchased products by member customers. A two-stage cluster analysis is employed. The results show that 8,073 member customers are grouped into ten clusters, with three clusters representing male customers, while the rest are female customers. Cluster 8 with a group of 51 female customers with all age groups is found to be the core customers for this supermarket. Cluster 10 consisting of female customers whose ages are 25 and above is the essential customers. In contrast to female customers, cluster 3 covers all age groups and is the most essential male customers for this supermarket and can be viewed as the core customers as well. The supermarket management can develop different marketing strategies for each cluster to meet member customer needs. Copyright © 2026 Inderscience Enterprises Ltd.
Department of Business Administration, National Changhua University of Education, Changhua, Taiwan; Department of Finance; Department of Marketing and Logistics Management, Chaoyang University of Technology, Taichung City, Taiwan; State University of Malang, Faculty of Education, East Java, Malang, Indonesia
