Loan underwriting is time-consuming and errors are expensive. Most loan applications require a skilled underwriter. There are hundreds of “hard” and “soft” rules which take years to learn. There is a backlog of loan applications, which means a loss of revenue.


Based on the existing loan history, the team trained a predictive model that automates loan decisions. A binary classification model was used with loan characteristics (income, credit score, debt, etc.) as input. The output was an approved/declined score.


  • Financial Sevices

Tools Used

  • Data Science Experience (DSX)

Data Science Techniques

  • Binary classification



  • Decrease of loan failure rates
  • Consistent, repeatable decision-making
  • Revenue increase by approving more loans

Need help with a similar case?

Contact us to find out how we can help!