Electronic retailing generates rich transaction data, yet converting these records into actionable customer groups remains challenging.This study develops a data driven segmentation approach by integrating recency-frequency-monetary (RFM) modelling with K-means clustering.Using a retail marketing dataset of 2,240 customers, data were cleaned and standardised, and internal validation was used to select the clustering solution.The results identify four distinct customer segments with clear differences in purchase recency, buying intensity, and spending contribution.Segment interpretation is strengthened by incorporating customer lifetime value and loyalty indicators, highlighting high value customers at risk of inactivity and low value customers with growth or reactivation potential.The findings demonstrate that RFM based clustering supports targeted electronic marketing actions, customer relationship management optimisation, and more efficient allocation of promotional resources in retail distribution.
Duong et al. (Thu,) studied this question.