Leveraging artificial intelligence for predictive customer churn modeling in telecommunications: a framework for enhanced customer relationship management

Abstract Customer churn remains a critical challenge in the telecommunications industry, impacting profitability and long-term customer value. This study proposes an Artificial Intelligence (AI)-driven framework integrated within Customer Relationship Management (CRM) systems to proactively identify and retain high-risk customers. Using a Random Forest classifier on a publicly available telecom dataset ( N = 2,668), the model achieved an accuracy of 95.13% and an AUC of 0.89 . Techniques such as SMOTE and class weighting were applied to address class imbalance (14.6% churn). Comparative experiments with XGBoost , SVM , and ANN confirmed the robustness of the proposed model. Feature importance analysis revealed that total day minutes, total day charge, and customer service calls were the most influential predictors. The study contributes by linking explainable AI insights to CRM operationalization, providing actionable strategies for proactive customer engagement and retention.