What problems was Fair For You trying to solve?
As a part of continued new lending decisioning improvements Fair For You was looking for a way to reduce the number of affordability-based declines by replacing and improving on ONS-based expenditure estimates.
Their goals were to:
- Increase their approval rates without compromising on risk
- Reduce the time spent on manual underwriting
- Streamline their operational processes for maintaining their affordability models
They needed more accurate affordability assessments to responsibly extend credit to a broader customer base while maintaining their risk appetite.
How did Infact help to solve these problems?
Infact's Affordability Engine was integrated into the lender's existing decisioning process, providing:
- Granular, individual-level expenditure predictions
- Enhanced affordability assessments using applicant-declared income and other relevant data points to personalise expenditure estimations for each applicant
- Comprehensive coverage with predictions for 100% of applicants
- Seamless integration with existing credit risk models following a short period of dual processing
How is the lender benefiting?
- Fair For You achieved a 16% increase in approval rates, materially reducing affordability-based declines
- Maintained their risk profile as part of larger decisioning improvements
- 48% reduction in manual underwriting time
- 75% reduction in time spent maintaining affordability model
Regulatory compliance
The solution supported the lender in meeting regulatory requirements for affordability assessments, a critical consideration in the non-prime sector.
How are the lender's customers benefiting?
Increased access to credit
More applicants are being approved for loans, with a 16% increase in approval rates. This means more borrowers can access the credit they need.
Faster decisions
With a 48% reduction in refer rate, borrowers receive decisions more quickly, improving their overall experience.
Responsible lending
Despite the increase in approvals, the default rate remained slightly lower than the incumbent model, adding an additional layer of insight to help protect vulnerable customers from over-extending.
.svg)
.svg)
%201.svg)
