An online B2C lender has been looking to expand their operations but has seen a concerning rise in the number of defaults on the loans they have already extended. Their risk models are built around traditional financial records — bank statements and FICO scores — but don’t take into account external factors. They have seen their default rate climb from the industry average 20% to nearly 30%, a worrying trend that has caused them to lose significantly month-over-month.
Our customer understood that the only way to remain competitive and still extend a good number of loans was to find a new way to score their borrowers’ default risk. The company connected its internal data to Explorium to enrich it with alternative financial and people data that could provide greater visibility into consumers’ financial health and activity. By enriching their data with thousands of external sources, the lender was able to generate the following features:
Once their risk models were re-trained using the new datasets, the company managed to reduce their default rate from its high of nearly 30% to 21%. More importantly, they have been able to better vet potential borrowers and extend much safer loans that are more likely to be repaid. The result is a revenue boost of nearly 4% in the six months following the adoption of their new risk model alone.