Fair Isaac: Home Credit, finalist for an award for AI-based optimization

The use of AI-based optimization by financial services companies is booming across Europe. One of the first beneficiaries of this technology is Home Credit & Finance Bank in Russia, which has just been named finalist for the Banking Tech 2021 Awards, in the Best Use of AI category. Using FICO’s AI-powered optimization, Home Credit was able to:

  • Increase turnover by 24%

  • Increase the average loan by 5%

  • Exceed 20% loan volume growth targets, while maintaining the same default and conversion rates

  • Maintain the overall conversion rate and the anticipated default rate

  • Improve pricing with more individual pricing instead of risk-based pricing

  • Use what-if analysis to deeply understand trade-offs, including volume, interest, and participation

Meeting new competitive and regulatory challenges

Home Credit, a group that lends to underserved clients with little or no credit history in nine markets in Europe, Russia, India and Asia, has increased lending volumes and profitability while maintaining the same default rate, despite increased competition in the market and regulations that impacted the capital available for lending.

In 2019, all banks in Russia found themselves in new conditions. Increased competition in the market and new requirements from the Central Bank resulted in lower profits. Home Credit needed a solution to manage this situation while keeping the risks at the same level. The solution had to be able to manage different types of data and business constraints to answer the question “What is the most profitable credit option that the bank can offer to this particular customer?”

Previously, the answers to these questions were obtained through a sequence of price pilots and manual analysis. All of this work was time consuming and the results did not always match the bank’s needs.

Application of AI-based optimization

FICO® Decision Optimizer provided Home Credit with a data-driven approach to define the most profitable loan option based on actual data, instead of assumptions. The final optimization model incorporates more than 50 raw characteristics and predictions from four machine learning models; Home Credit Group Risk, Use, Underwriting and Underwriting Models. The heart of optimization is an equation that gives an expected lifetime benefit for each specific contract.

Find out more about the Home Credit approach in this series of short videos with Petr Kapoun, former Home Credit CRO in Russia.

Congratulations to the Crédit Domicile team!

Disclaimer

Fair Isaac Corporation published this content on 04 October 2021 and is solely responsible for the information it contains. Distributed by Public, unedited and unmodified, on 04 October 2021 14:10:23 UTC.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *