The telecommunications industry is particularly competitive and faces high churn rates, making customer retention a critical challenge. Despite a vast body of research, especially in recent years, studies on churn determinants remain fragmented, lacking a comprehensive synthesis. This study conducts a System-atic Literature Review to consolidate existing knowledge, identifying key churn predictors such as contractual, financial, and behavioral factors. It reviews 50 articles on the topic between 2019 and 2024 from the Web of Science (WoS) database. The review highlights advances in machine learning models, particu-larly random forest, XGBoost, and neural networks, while also uncovering gaps in explainability, dataset generalizability, and the integration of real-time data. Additionally, the study maps research trends, methodological limitations, and emerging approaches, offering a structured foundation for future work. By ad-dressing these gaps, this review guides the development of more robust, inter-pretable, and industry-relevant churn prediction models.

This work was supported by Fundação para a Ciência e a Tecnologia, grant UIDB/00315/2020 (DOI: 10.54499/UIDB/ 00315/2020) and by MSc in Business Analytics from Iscte-Business School.

Churn,Prediction,Telecom