Machine Learning Supercharges Banks’ Efforts To Fight Debit Fraud

Consumers have more heavily leaned on debit during the pandemic, with the economic downturn making shoppers more cautious than ever about the prospect of taking on credit card debt. A recent study even estimates that shoppers could ultimately shift $100 billion worth of annual spending from credit cards to debit cards. Debit solutions draw on funds consumers already have in their bank accounts, and while this makes them reassuring to debt-wary consumers, it can have implications if these details are snatched by fraudsters. While fraud affects less than 1 percent of all card purchases, consumers who do lose funds from their bank accounts must go through lengthy and often stressful processes to get their money back.

The December Next-Gen Debit Tracker® examines how card issuers are working to sharpen their fraud-fighting tools and leverage innovative, machine learning (ML)-based strategies and technologies to keep shoppers safe.

Around The Next-Gen Debit World

Bad actors have ramped up their attacks against debit card holders in India, where such scams are reported to have risen 75 percent during the pandemic. Officials have struggled to stop or even detect these crimes, and fraudsters are making the task especially difficult by leveraging various schemes. One popular scam sees fraudsters pretending to be government officials and alleging that consumers need to hand over payments data to receive relief funds.

Getting security right also means balancing customers’ security and privacy concerns. Location tracking allows financial institutions (FIs) to detect red flags such as users making mobile purchases from one location while claiming to be in another, but getting customers on board with the technology can be challenging. PYMNTS’ research recently found that simply explaining the technology’s benefits to consumers as well as how and why the information would be used could reduce their objections.

Safeguarding debit card payments is increasingly important, with more consumers shopping online during the pandemic. A recent report found that nearly three-quarters of consumers planned to use digital payments during the holiday season, including leveraging payments that use their debit card details. Debit card use has also been on the rise during the pandemic, jumping 9 percent from February to November.

Find more on these and the rest of the headlines in the Tracker.

Leveraging Machine Learning, Rules-Based Analysis To Fight Debit Fraud

ML is a powerful, flexible tool in the fight against cybercriminals who attempt to compromise debit payments. This advanced learning technology can be especially useful against card-not-present (CNP) fraud originating from scammers overseas, but it works best as part of a multilayered approach, according to Karen Boyer and Frank Wheelahan, executives at People’s United Bank. Boyer and Wheelahan explain in this month’s Feature Story how pairing ML with rules-based analysis can help FIs catch new fraud trends early.

Read the full story in the Tracker.

Deep Dive: How Machine Learning-Powered Solutions Power Up Fraud-Fighting

Many consumers are limiting or modifying their in-store shopping habits during the pandemic, meaning they are looking to pay remotely online or use contactless methods when purchasing in stores. These behavioral changes mean that shoppers are not entering their PINs, forcing FIs to develop new ways to filter legitimate customers from fraudsters. The Deep Dive examines how FIs are using ML and other tools to analyze customers’ behaviors and identify genuine cardholders.

Find the Deep Dive in the Tracker.

About The Tracker

The Next-Gen Debit Tracker®, a PYMNTS and PULSE collaboration, examines how leveraging machine learning can help banks fight debit fraud as more commerce moves online.