Propensity to Buy Model

Problem

Banking Institutions wanted to know their existing potential customers that can buy their term deposit product. Knowing that term deposits allow banks to hold onto a deposit for a specific amount of time, banks can invest in higher gain financial products to make profit. In addition, banks also have a better chance to persuade term deposit clients into buying other products such as funds or insurance to further increase their revenues. 

Solution

Understanding the data and the need of the bank was the first part of the solution provided. After a deep exploration of the variables, we identified the important data & variables for building the predictive model. From our business experience, loan details, customer profile data, credit in default, term deposit subscription, and customer response data are used to be able to predict the customer’s most needed purchase. Different types of ML algorithms were implemented to get the best model that understands the correlation between the variables and makes the most accurate prediction.

Results

The model gave a prediction of 90% recall and 85% by testing the customer purchase for the next year, by giving a pass from the customer for the model to go into production.