The rapid adoption of Buy Now, Pay Later (BNPL) services has transformed consumer payment behaviours, presenting both opportunities and challenges for retailers seeking to drive customer retention. This study investigates the psychological drivers underpinning BNPL adoption, with the objective of predicting usage patterns and informing strategic marketing interventions. Drawing on primary data from 350 active BNPL users across retail categories including electronics, apparel, and lifestyle goods, we integrate behavioural constructs—such as impulsivity, perceived financial control, trust in fintech platforms, and gratification delay tolerance—into a predictive modelling framework. Three supervised machine learning algorithms—Logistic Regression, Random Forest, and Extreme Gradient Boosting—were applied to forecast high versus low BNPL usage frequency. Feature importance analysis using SHAP revealed that impulsivity, trust, and perceived affordability were the most significant predictors, while demographic variables played a comparatively minor role. Among the tested models, Random Forest achieved the highest classification accuracy (AUC = 0.89), demonstrating the potential of advanced analytics in consumer finance research.
Sangeeta Sahni Namrata Singh (Mon,) studied this question.