This study addresses the need for behavior-based customer segmentation in the retail sector by introducing a novel methodological framework that combines multidimensional factor analysis with machine learning. The framework also yields strategic insights with direct implications for retail marketing, campaign management, and customer relationship development. Adopting a data-driven approach, the study uncovers behavioral patterns among supermarket customers in Greece. Using factor and clustering methods, six distinct shopper profiles were identified based on purchasing habits, store preferences, promotional responsiveness, and affinity for private label products. A predictive model was then developed to classify unknown customers into these profiles. The results provide practical tools for targeted marketing and customer engagement. Key business implications include enhanced personalization, improved customer loyalty strategies, and more efficient campaign planning, demonstrating the strategic value of integrating behavioral analytics into retail decision-making.