Behavior-Based Customer Segmentation Using K-Means Clustering in Supermarket Retail
Year: 2026
Type: conference-paper
Abstract: This research incorporates customer segmentation as a crucial element in the formulation of data-driven marketing strategies, especially in retail sectors such as supermarkets that face high transaction volumes and diverse consumer shopping patterns. The main objective of the research was to segment supermarket consumers based on actual transaction data and membership data, using a total of 15,631 registered customers. Consumer behavior was engineered through a number of features reflecting purchase intensity, shopping cart structure, variety of products purchased, and transaction recency patterns.In the analysis stage, this research applies and tests two clustering approaches, namely K-Means and hierarchical agglomerative clustering with Ward’s linkage. In the modeling stage, a combination of parameters is searched through grid search and assessment using internal validation metrics to obtain a more suitable cluster configuration. After that, each cluster is interpreted based on the consumer shopping behavior profile formed from the clustering results.This research shows that behavior-based segmentation is able to distinguish groups of customers with different shopping characteristics. The findings suggest that the clustering approach can be used as a basis for designing more targeted and data-driven activation, retention and management strategies for supermarket customers.
Institutions: Binus University, Universitas Ekuitas Indonesia
Language: English
Subfield: Marketing
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7204750110