In areas where a majority of applicants lack traditional credit scores, fintech companies typically base their lending decisions on information they collect directly from the applicants.
Fintech firms find this challenging as the data is frequently insufficient for preventing fraud and filtering out likely defaulters.
In response to this, they are building credit risk assessment models using alternative data, often sourced from third-party data providers.
This case study by Mobilewalla takes a deep dive into how a leading Asian fintech company has significantly improved its credit score model’s performance and found a corresponding decrease in default rates using highly-predictive data sets.
Mobilewalla is a provider of consumer intelligence solutions with deep data science expertise and a global consumer data repository.
The paper explores why it’s important for the fintech company to have access to scalable and highly reliable data as well as how a small improvement in risk modeling can translate into big increases in revenue.
The case study details readily available predefined data features found to be most predictive of default risk and the modeling strategy used to predict defaulters more accurately.
It also shows how these tools resulted in a 5% Gini Index lift and decreased default rates.
As the financial industry continues to evolve at a rapid pace, organisations need to better leverage data and AI to drive innovation and gain competitive advantage.
A recent study by Deloitte found that 30% of financial service firms who they describe as frontrunners are more adept at utilising AI which is helping them increase revenue faster than their competitors.
Explore how you can use third party data to prevent fraud and decrease defaults here.