Description: To predict whether a customer is going to default in the future based on his spend and payment patterns. Methodology: Built a Logistic Regression Model using historical data of customers who had defaulted previously Tools Used: R, SAS and SQL server Impact: Early identification of delinquent customers led to an early intervention by the Collections team. This model has helped the collections team to retrieve the money and in the month of august they have collected around 2.5 Billion Rs which has indirectly contributed to decrease the NPA(Non Performing Assets)