Table of contents
Even before the pandemic hit, India’s bad debt ratio was the worst amongst the world's top ten economies. Cut to 2021, with Rs. 10 lakh crore worth of stressed assets, projected NPAs for the current financial year are set to beat a two-decade high. Of course, the pandemic did not help matters - with lockdowns that limited people’s ability to repay and banks’ ability to collect.
This unearthed the glaring pitfalls in India’s lending processes, where digitization has been a patchwork at best. Efforts to streamline underwriting, identity verification or decisioning are influenced by the need to cut down customer wait time. Technologies such as artificial intelligence, big data analytics or social listening tools are deployed to accelerate time to disbursal. However, post disbursal, the scene remains largely archaic with a decades-old approach to collections or loan restructuring.
Typically, banks liaise with a collections agency and outsource responsibilities to proprietors who act as collection agents. These are essentially unregulated entities who tend to flout rules of conduct, resort to harassment and risk reputational damage for the lender.
This is why the Reserve Bank of India (RBI) released a report recommending regulatory, technological and consumer protection frameworks for loan recovery. These recommendations put in place guardrails for banks and non-banking financial corporations (NBFCs) backing a digital lender, effectively making them guardians to stop the mushrooming digital lending landscape from going rogue.
It all boils down to this - to rein in bad debt, lenders need to overhaul collection processes. Starting from AI-driven, real-time customer segmentation to defining a collections-focused risk-predicting model and loss-forecasting strategies, lenders will have to take digitization to the anteriors of the loan management lifecycle.
We discuss how.
Incumbent collection practices are foolhardy
Collection efforts have been driven by rudimentary forms of customer segmentation, such as, number of days delinquent or frequent past defaults. Such segmentation harbours inherent biases about customers’ repayment abilities and reduces them into “early-stage” and “late-stage” delinquents (or some variation thereof).
This model skewedly focuses on delinquency buckets that do not add any value to the lender. There are customers who would pay up after 15 days delinquent with minimal contact, and there could be customers who would default on an upcoming payment despite reminders.
Moreover, the days-delinquent strategies are labour-intensive and assign templatized messages to each category, often proving to be counterproductive. A McKinsey research found out that 20% of customers said they have withheld a planned payment because they had an upsetting call from a collector.
Instead, lenders should focus on tailoring their segmentation criteria for every borrower according to loan amount and tenure, along with their financial behaviour during the credit cycle. Here are alternate approaches to segmentation that are more likely to elicit payments -
Segment of one
Move away from a broad-based, static approach to segmentation that puts customers together in a black box based on past events. Instead, create ‘segments of one’, studying each borrower’s real-time financial behaviour near the due date. Machine learning models can help lenders glean insights to determine the probability of a default on a case-by-case basis (each borrower’s ability and intent to pay in real time).
Value-at-risk (VAR) segmentation
VAR - a measure of the potential and probability for losses to occur - helps lenders prioritize collections from high-risk, high-balance customers and refocus their communication strategies.
Lenders need to assess personal customer data and analytics to identify subjective experiences such as lack of liquidity or cash flow restraints that deter customers from timely repayment.They can create sub-groups within their cohort of at-risk borrowers and alter repayment plans or contact strategies on a single-bucket customer case.
Redundant data insights
Current collection risk assessment models continue to resemble those used at the time of underwriting. However, by the time a loan is disbursed till the first repayment date, chances are that the customer’s financial behaviour may have changed.
Lenders assess a longer history of credit hygiene during underwriting, but by the time of collections, this data becomes redundant. What becomes pertinent at this point is the customer’s activity during the credit cycle - the utilization of the loan and their spending behaviour. Unfortunately, most lenders lack the required visibility into a customer’s financial records post disbursement, creating a vacuum that puts the lender in a blind spot regarding the success rate of eNACH representation.
Moreover, traditional NBFCs credit scores from credit bureaus which are only updated bi-monthly. This lag contributes to the vacuum, where a lender has no insight if the borrower had defaulted on other loans in a month’s time and can in no way take remedial measures to ensure repayment on time. Inversely, if a delinquent borrower had made good on his/her past dues, the lender can only know of the changed financial behaviour after a lag of 60 days, meanwhile adopting aggressive collections practices that may potentially border on harassment.
Inefficient contact and debt relief methods
Attempts to contact customers are expensive and inconvenient, other than being downright aggressive and harassing. This practice continues despite consumer protection guardrails like the central bank’s Code of Recovery.
Contact methods remain ineffective as banks are simply not using customers’ preferred channels which are more likely to deliver the desired outcome. Lenders continue to contact customers through mail and phone calls during the later stages of delinquency despite their likelihood of responding better to digital nudges.
Templated communication is futile if customer satisfaction is high on lenders’ agendas. They must contact customers through preferred channels to elicit payments more successfully.
(Sourced from McKinsey)
Currently, contact strategies target two-dimensional risk profiles derived from the “days delinquent” segmentation approach. But instead of doing the talking, banks must heed their customer’s demands.
Here’s how effective communications can help increase recoveries -
Inbound recovery operations
Instead of reaching out to customers themselves, banks can deploy inbound recovery operations. They can design pre approved solutions for their debt, giving borrowers options to choose from. For instance, they can offer a reduction in the interest rate or no repayment for a few months. This would prompt contact from the debtor’s end.
Tailored debt restructuring
A combination of VAR and behavioral segmentation provides greater visibility into the customer’s financial circumstances. Better communication can help lenders draw up debt restructuring and relief plans according to the customer’s liquidity.
How FinBox CollectX can fix collections
Designing a contact strategy, optimizing collection resources or predicting a likely default should now be a priority.
FinBox can help lenders optimize collection processes with advanced behavior-based segmentation tools. Designed to take more accurate cues from debtor behaviour, FinBox CollectX - our intelligent, risk prioritization engine - is a state-of-the-art early warning system that makes collections streamlined and sure-shot.
The collection prioritization engine analyzes borrower financial activity in real time. It takes stock of customers’ spending behaviour, quality of credit and its utilization. Data assessed by FinBox CollectX helps lenders preempt a default by giving them a lowdown on customers’ financial health as early as one week prior to the due date.
FinBox’s technology stack works on rule-based workflow automation best equipped to process data for both underwriting and collection. The engine segregates scenarios based on real-time events and helps lenders assess the likelihood of a default based on the current financial situation of a borrower.
FinBox CollectX helps lenders manage delinquencies profitably by prioritizing accounts on agreed-upon criteria (be it days delinquent or value at risk), so lenders can optimize collection efforts based on severity. Ultimately, this has enabled lenders to curb the delinquent accounts by 10%, controlling high roll rates and increasing collections.
Boost eNACH success rates
FinBox CollectX has yielded 30% improvement in eNACH success rates of already defaulted accounts. The engine gives lenders much-needed visibility into borrower behaviour such as spending patterns and credit utilization during the credit cycle, departing from the industry practice of using underwriting-like models to arrive at the possibility of another default. Based on this, lenders know if an eNACH representation will be successful or if a restructured payment plan needs to be worked out to offer early settlement.
Tailor contact strategies
Intelligence sourced from FinBox CollectX helps lenders allocate resources for communication as per the severity of the case. Lenders can tailor text messages, emails and notifications for each borrower, ensuring higher customer satisfaction and improved collection efficiency.
India’s leading digital lender improves eNACH success rate by 30% with industry-beating customer satisfaction scores
FinBox CollectX helped India’s leading digital lender, offering buy-now-pay-later services, to consumers to bring down default rates significantly. With its forward-looking collections prioritization, FinBox CollectX improved collections efficiency by 50%, boosting eNACH success rates by 30% for over 10,000 already delinquent accounts. Our recommendations for tailored contact strategies for each borrower segment helped the lender maintain its industry-leading net promoter score, higher than Amazon and Uber.
Find out how FinBox CollectX can help you clean up your loan books. Click here.