Based on its list of corporate customers, banks classify customers with high demand for loan packages (corporate customers with financial problems). Then provide this list of customers to bank branches in the system for branches to use sales staff via telesales channel to call, invite, introduce, and make appointments to advise on the bank's loan packages. The use of machine learning models will help banks select the most potential customers, thereby creating efficiency in their business processes. The data is taken from the General Department of Taxation, variables related to the financial index of enterprises in the last 2 years. Models such as Random forest, Gradient Boosting, and XGBoost can be used. The output of the model is to support the CRM (Customer Relationship Management) team to call and offer credit and non-credit products to corporate customers.