The Covid-19 not only led to one of the most devastating health crises for the century but also jolted economies globally, which were anyway struggling with mismatched fundamentals. Lockdowns, demand slowdown, job losses, and financial anxiety, dried up cash flows for many businesses as well as individuals. The result was a huge spike in both commercial and retail Non-Performing Assets (NPAs), as borrowers struggled to manage their repayment obligations.

In India, the sheer size and severity of NPAs could be understood from the fact that both the Government and banking regulator RBI are seriously evaluating setting up a ‘Bad Bank’, the one that takes over NPAs from various banks and allows them to maintain clean books. The need for one such bank is pronounced even more when merely in Q3 FY21, major banks wrote Rs 25,500 crore off their books, according to sources.

While policy and regulatory measures will take time to address the issue of NPAs, progress in technologies such as big data, artificial intelligence, and IoT has already started playing a huge role in this direction. This is how it is changing the game:

The new-age approach to default management

Traditionally, banks and NBFCs engage with delinquent customers or potentially ‘risky’ customers over phone calls, emails, or through personal visits. The methods of collection are often perceived as intrusive and inefficient. However, breaking the tradition is now the need of the hour. Therefore, financial institutions are increasingly investing in many areas powered by technology:

Identification of potential defaulters: A loan is as good as the creditworthiness of the borrower. The more data lenders have about a potential borrower, the better they can evaluate his creditworthiness. Banks can get a better insight into borrowers by combining intelligence from their deposit accounts and payment patterns. It allows banks to foresee the potential of default and take preventive measures accordingly. For instance, solutions powered by Machine Learning (ML) and Natural Language Processing (NLP) can help in analyzing an individual borrower’s digital footprints and project if there are any adverse financial situations underway. For instance, the savings account of the person may have a sudden decline of inflow and his digital activity correlates it with a job loss. In such a case, lenders may have greater default risk.

Personalised collection strategy: Technology makes it possible for banks and NBFCs to personalize their collection strategy according to the traits of individual borrowers. This is done by evaluating data points from past repayment strategies that helped recover dues from borrowers with similar profiles. Using these insights, banks can build a predictive model based on personas. Such models need to be continually updated and refined so that they keep working hard for the bank and constantly improving yields. Big data, artificial intelligence, and machine learning will play an important role in identifying critical factors and variables for improving these models and thus in tweaking existing strategies to keep processes effective.

The above examples only discuss the tip of the iceberg and there is a lot more potential to improve lending, collection, and recovery processes through technology. Banks are working with their technology partners to implement these new-age technologies for non-intrusive yet effective methods of interaction with delinquent customers. The attempt is not merely to bring down the NPAs but also to improve the customers’ experience.

The author is CEO and Co-founder of Datacultr

DISCLAIMER: Views expressed are the authors' own, and Outlook Money does not necessarily subscribe to them. Outlook Money shall not be responsible for any damage caused to any person/organisation directly or indirectly.