Abstract

Predicting Customer Lifetime Value (CLTV) is crucial for effective marketing resource allocation and strategic customer relationship management. This paper proposes a novel hybrid approach that integrates traditional Recency, Frequency, and Monetary (RFM) analysis with advanced machine learning techniques to enhance the accuracy and robustness of CLTV predictions. We develop and evaluate several machine learning models, including regression algorithms and classification models for churn prediction, and compare their performance against traditional RFM-based methods. The proposed hybrid model leverages the strengths of both approaches, using RFM scores as features within the machine learning models. Empirical results, derived from a real-world transactional dataset, demonstrate that the hybrid model significantly outperforms both traditional RFM analysis and individual machine learning models in predicting CLTV, leading to improved marketing ROI and customer retention strategies. Furthermore, the paper provides insights into the key factors driving customer lifetime value and offers practical recommendations for businesses to optimize their customer engagement strategies.