The work aims to address the critical issue of customer attrition and improve their strategies to hold the customers by providing more facilities in the banking sector through the application of machine learning techniques. In this study bank customer churn or attrition pre-diction by using a comprehensive data analysis. The main objective is to forecast whether customers are going to leave the bank soon. This research aims to develop a predictive model that can accurately identify the risk of customer churn in banking sectors by analyzing large customer data. The dataset contains detailed information on customer personal data, relational data with the bank, transaction details, and credit profiles. This project includes data preprocessing, handling missing values, and ensuring data consistency which are the vital steps to achieve the goal. Compared predictive performance of some machine learning algorithms support vector machine (SVM) and Decision Tree here. Focused on evaluation metrics: Precision, Recall, and F1-score each model’s effectiveness and importance. The findings from this study focused on the customer churn and insight of the banks to their customer’s retentions. Banks can improve their strategies by finding the issues for customer’s churn and also can improve their loyalty toward customers and this can boost their profit. This study identifies the importance of decision-making by analyzing the data to minimize customer attrition.