Abstract

This study proposes a novel AI-driven framework for modeling Customer Return Rate Evolution (CRRE) by integrating intelligent personalization and next-best-action (NBA) marketing strategies within advanced Customer Relationship Management (CRM) systems. While traditional CRM approaches primarily rely on static segmentation and historical analysis, this research introduces a dynamic, predictive paradigm leveraging machine learning, natural language processing, and behavioral analytics to forecast customer revisit behavior and optimize engagement strategies in real time.The proposed model combines customer lifetime value (CLV), sentiment intelligence, and adaptive recommendation engines to capture the temporal evolution of return rates across different customer segments. Empirical evaluation on multi-source datasets demonstrates that AI-powered personalization significantly enhances customer retention, engagement, and revenue performance, confirming that hyper-personalization and predictive analytics play a critical role in shaping customer loyalty .Furthermore, the integration of next-best-action strategies enables firms to deliver context aware and timely interventions, improving decision-making efficiency and customer experience. The findings highlight that AI-driven CRM systems outperform traditional approaches by providing adaptive, data-driven insights and real-time optimization capabilities.This research contributes to the literature by introducing the concept of Customer Return Rate Evolution as a dynamic performance metric and by proposing a unified framework that bridges predictive analytics, personalization, and marketing decision intelligence. Managerially, the study provides actionable insights for organizations seeking to enhance retention strategies and achieve sustainable competitive advantage in increasingly data-driven markets.