The paper aims to develop and validate an analytical framework to assess the service quality and customer satisfaction and hence customer loyalty in banks. We develop a multivariate probit model for the evaluation of service quality and customer satisfaction and an ordered logit model for the rating of customer service of the most frequently used bank. To validate the models, we conduct a survey of 1107 customers from the branches of eight banks in three major cities of the peninsular Malaysia in 2013. The validation was carried out in two stages. First, we analyze the customers’ satisfaction on the core and relational dimension of service quality using multivariate probit model. In the second stage, we analyze the customers’ overall service rating on their most frequently used bank using ordered logit model. The proposed framework can be used by banks to assess service quality and customer satisfaction and identify customers who are likely to switch. This would prove extremely beneficial to banks in the formulation of segment specific customer retention strategies. The framework would also assist regulators to gauge the level of customer confidence in the individual banking institution and intervene as and when necessary to maintain stability in the banking system.