identifying customer segments in e-commerce: a data-driven framework using transactional patterns

This study proposes a structured and interpretable approach to customer segmentation by enhancing the classical rfm (recency, frequency, monetary) model. the research addresses the common limitation of traditional rfm-based methods, which often overlook behavioral diversity and lack flexibility for practical, data-driven decision-making. the main objective is to develop a segmentation framework that provides actionable insights based on transparent, rule-based logic rather than opaque clustering algorithms. using transactional data from a leading e-commerce platform in the netherlands, the methodology applies quartile-based scoring to rfm indicators and maps customers into six distinct behavioral segments. visual analytics are employed to support interpretation, enabling businesses to tailor engagement strategies for each group. the results demonstrate improved segment differentiation, managerial interpretability, and relevance to real-world applications. the study also highlights the model’s adaptability to other customer-centric industries, with future research directions focused on incorporating machine learning and behavioral enrichment for customer lifetime value (clv) prediction.