Mobile

Dec 13, 2019

How Mobile Usage Can Influence Access To Credit

Subscribe to our newsletter

This interview originally appeared in Outlook Money India.

Technology has revolutionised the lending business. From borrowing and repayment history being used to evaluate creditworthiness of a potential customer, lenders are increasingly using new data points to arrive at the decision to lend these days, including mobile phone based data. Creditworthiness, which essentially means the ability and willingness to repay a loan, is the foundation of success of any lending business be it large banks, NBFCs or fintech companies. Michele Tucci, Chief Product Officer, Credolab, explains how mobile device based metadata is being used to redefine the way creditworthiness is assessed in conversation with Vishav. Edited excerpts.

How is mobile-device metadata connected to creditworthiness?
Creditworthiness can be assessed in two ways: ability to repay and willingness to repay. The first one is usually based on information rather transactional such as income, rent, utility bills and similar. The second one is behavioral in nature. We analyse how a consumer uses his mobile phone. Machine Learning algorithms are able to detect any delinquent behavioral pattern among about one million possible ones. The process to find such a correlation uses a very solid statistical approach that has been audited even by a professor of statistics at Yale University.


What are some of the advantages of this approach? How can it redefine the industry?
The most important advantage of this approach, and also the most impactful on the industry, is the ability to assess the creditworthiness of consumers that today have no credit score - for instance with CIBIL - or have a 'thin file', a credit report with not enough information or credit history for a financial institution to make an informed credit decision. We deliver the score in real-time, hence helping the financial institutions design digital loan or credit card application journeys that deliver value in real-time.


What is the need for this kind of credit scoring? Do finance companies really need this considering there already are other credit scoring mechanisms?
Finance companies today have different pain points to solve. Some need to increase market share, some need to decrease cost of risk, some need to digitalise their processes. Our credit scoring mechanisms are designed to help finance companies achieve their objectives with as little as possible impact on their day-to-day operations and IT investments.


How are you minimising risk using this approach?
We help finance companies decrease cost of risk by improving the predictive power of their credit risk models. Our approach offers finance companies the ability to assess a completely new dimension of a loan or credit card application. Our solution works well for new-to-credit customers as well as those with 'thick credit files'.


What are some of the privacy risks and how can they be addressed?
All our solutions are designed to work without personal information. Our mobile apps and mobile SDKs access only metadata (defined as data about other data), not personal data. Our credit scoring system is also based exclusively on metadata.