Abstract

In the dynamic and competitive domain of financial technology, particularly within the online payment sector, the phenomenon of customer churn presents a significant challenge, necessitating advanced predictive strategies for sustainable customer relationship management. This study, centered on a leading payment provider in Turkiye, employs survival analysis to discern key churn determinants. Three distinct models were developed, each integrating variables such as demographics, payment history, and usage patterns, but with varying definitions of churn. Empirical findings from the study highlight the paramount importance of variables, such as changes in commission rates, refund rates, payment counts, merchant types, and sector affiliations in predicting customer churn. A significant insight of the research is the identification of a heightened churn risk post a 25-month customer-company engagement duration. Among the models, the one characterizing churn as a lapse in activity over a one-month interval exhibited superior efficacy. This study underscores the utility of survival analysis in the domain of churn prediction in online payment platforms, offering pivotal insights for strategizing enhanced customer retention approaches.

Download

Share

COinS