Early Warning System For Credit Card/Loan Default Payment Behaviour
Identifying risk clusters of specific customer lifestyle– income segments likely to default in credit repayment
Axis Bank is the third largest private-sector bank in India, offering a comprehensive range of financial products. The bank has its head office in Mumbai, Maharashtra. It has 4000 branches as of 25th March 2019. It has 12,705 ATMs and 3,548 cash recyclers spread across the country as of 31st December 2018 and ten international offices. The bank employed over 55,000 people and had a market capitalization of ₹1.31 trillion (US$18 billion) as of March 31st, 2018.
It sells financial services to large and mid-size corporations, SMEs, and retail businesses.
The Challenge
When underwriting a new customer for a credit card or loan, the bank runs a credit score verification process for the applicant with the Indian Credit Bureau (TransUnion CIBIL). The problem is that nearly 30-40% of the applying customers, especially personal loans, do not have a credit score in India.
Without a credit score for the loan/credit card applicant, Axis Bank was looking for detailed and reliable consumer profile information to build into their approval models for significantly downsizing default ratios. Note: Among the Indian population of 1.4 billion, only 400 million individuals have a presence on credit bureaus today. As many as 191 million Indians above the age of 15 are still without a bank account (Source: CIBIL, 2019)
The KENTRIX Solution
KENTRIX profied Axis Bank customers into consumer lifestyle affinity segments integrating income and product purchase affinities. Further, the history of loan/credit card non-repaying/default customers was profiled to identify a “default consumer profile” as a template for the bank.
Based on the customer profiling delivered, the bank identified that their risks for defaulting in repayment lie 4 to 5 times higher in specific consumer lifestyle segments with the integrated income, age, and family status profiles. Axis Bank integrated the KENTRIX data on customer income and lifestyle affinity segmentation into their Risk Models, which helped the Bank significantly bring down its default rate.
