The prevailing challenging nature of directly measuring customer satisfaction in retail banking encourages this research to propose a novel metric that leverages customer behavior and demographics to directly measure customer experience (CX) in banks, replacing traditional surveys. Previous studies acknowledge the importance of customer satisfaction (CSAT) but mention challenges in accurately measuring it. The framework includes two models: the customer satisfaction prediction model, which analyzes transactional data to predict a customer satisfaction score, and the customer churn propensity model, which identifies those customers at risk of leaving the bank. By combining these two models, the framework calculates a net satisfaction score that represents a customer's overall CX. This score eliminates the bias from self-reported surveys and allows banks to evaluate marketing campaigns, proactively analyze customer sentiment, and prioritize interactions based on experience scores. The accuracy of this framework is validated through high-performing XGBoost models. Future research could explore quantifying customers’ digital inclination for further enhancements in CX measurement. Overall, this framework enables data-driven strategies for managing customers by quantifying their experiences accurately.