Preparing for the Next Evolution of Buy Now, Pay Later...And Beyond

Recent holiday shopping season data showed U.S. shoppers spent $16.6 billion dollars using BNPL plans, with BNPL volume increasing 14% year-over-year.

A survey by the Federal Reserve Bank of New York showed 64% of respondents had been offered a BNPL loan, and 29% of that group had used it as a payment method within the past year. BNPL is also growing in adoption with older consumers, not just Gen Z and millennials. According to Insider Intelligence, about 29% of Gen X leveraged BNPL in 2023, a number that is expected jump to over 40% by 2027. The share of Baby Boomer BNPL users is expected to grow from 13% to 17.9% in 2027.

One of the major reasons BNPL’s popularity has grown so rapidly is its broad accessibility. Consumers don’t have to pay interest, and some providers don’t even charge late fees for missed payments. The risk checkpoints that have historically acted as barriers for the credit unserved or underserved no longer exist. And without the data and risk technology in place to fill in those gaps, delinquency can and will happen. And it’s happening at a rate of 2.39%, up from 1.83% in 2020, according to a recent report by The Consumer Financial Protection Bureau (CFPB).

Consider the larger economic context fueling BNPL usage: Developed and emerging markets are grappling with cost-of-living crises; inflation is high, and delinquencies across consumer lending are the highest they’ve been in over a decade.

What can BNPL providers do to mitigate risk in these highly volatile times?

What’s needed is a sustainable strategy that maximizes process/cost efficiency and portfolio performance with artificial intelligence (AI) and machine learning (ML). But defining the sustainable strategy is just the start – enabling technology to support these strategies and adapt as needed is crucial.

Focusing on the Data that Matters

An advanced risk decisioning foundation helps ensure BNPL providers aren’t losing revenue to default and fraud, while stemming revenue loss by avoiding unnecessary data calls and bloated processes. Eliminating excess data ensures every data point is earning its place and adding value to decisioning.

Use of AI/ML enables analysis of decisioning processes. Are there redundant data sources? Are they the right sources for your target market? Is each data point coming at the right step in the process? In short, how effective is your data? With embedded reporting, organizations can continuously optimize their strategies to keep pace with the needs of the business.

Machine learning can be used to test tactics and help analyze strategies. Will removing a data point suspected to be redundant change the outcome of decisions? Will calling on a data source later in the process reduce the cost of that process? How do these tweaks affect different segments? How can organizations complete processes faster?

These efficiencies are key to long-lasting success by reducing cost on the backend and minimizing both customer default and churn rates. Running more efficient processes that can increase application volume and more accurately define affordability, provides the ability to serve more customers who are more likely to pay on time. Iterate as frequently as necessary to ensure accurate, efficient models are in place to back lending strategies.

Evaluating BNPL Product Performance  

When processes are optimized for efficiency and cost, the next focus should be maximizing portfolio performance and increasing customer lifetime value (CLV). This involves leveraging data analytics to gain insights into borrower behaviors, trends, and overall portfolio performance. This data supports informed decisions on credit limits, pricing, and risk management strategies.

Another key tactic is to separate product lines to explore their performance individually based on AI/ML-powered performance analytics. The BNPL product for e-commerce may be performing well but the one for auto repairs may need an adjustment in risk appetite or application processing to perform at its peak. Organizations may also need to expand or reduce customer segments.

Global providers will likely have different strategies and processes already in place. Use AI/ML to help optimize different customer bases to ensure targeting the right customers in the right areas. Are there certain segments defaulting at higher rates? Are there low risk segments that are being overlooked? Are there high loan value segments worth leaning more into risk to capture? AI/ML helps to optimize lending strategies based on performance data and identified areas for improvement.

Finally, it is important to incorporate iterative performance analytics to determine the effectiveness of customer-centric approaches. By doing this, organizations can:

• Predict how building in new segments will affect your risk or test out the viability of new business lines.
• Measure how CLV changes dynamic credit limits are imposed based on customer behavior or if more touchpoints are added to the customer journey.
• Regularly monitor portfolio performance and key metrics, such as default rates, repayment rates, and customer satisfaction.

A Solid Risk Decisioning Foundation is Key to Sustainable BNPL Growth

Accurate assessment of credit risk is the most effective path to minimize business risk. The best BNPL solutions stand on solid risk decisioning foundations that fight fraud, determine creditworthiness accurately, and still support financial inclusion.

When building a risk decisioning foundation, focus on the following:

• Return complex decisions: BNPL products do not rely on the traditional methods of determining creditworthiness. That’s why it’s vital to analyze and action multiple types of data in multi-step processes that ramp up from broad to distinct. Look across identity, fraud, and credit for the fullest picture of risk – however, consider the picture may look different for someone who is credit served with a prime or super-prime credit score than an underserved or unserved borrower. A complex decisioning engine can assign different weight to different factors.
• Easily integrate data and analytics: Returning complex decisions requires the right data. The absence of robust bureau data can make that a challenge. Seek out a unified decisioning engine with alternative data and complex analytics to power the accurate decisioning to form a solid risk mitigation foundation.

Next-Gen BNPL Requires Next-Gen Technology

BNPL has changed the world and it’s time to grow and mature with the industry by expanding into new markets to promote financial inclusion and achieve profitability. The time is right to build or optimize a BNPL product that can compete now and in the future.

This necessitates advanced credit risk decisioning. By choosing technology that can provide both highly accurate credit risk decisioning with exceptional customer experience and accessibility, BNPL providers will be ready for the next evolution of BNPL and beyond.