How to Fight Fraud Without Adding Friction

Fraud has always followed the money, but in recent years, technology has dramatically accelerated the sophistication and scale of fraud tactics. 

Every step the financial industry takes towards digital offerings that enhance security and outcomes, bad actors take similar steps in advancing their fraud tactics and reach. It’s clear that data is the most powerful weapon financial institutions have for fighting back. But, data alone stands little chance of combating fraud. 

Financial institutions need to leverage enhanced and connected consumer-permissioned data to turn the tide and stay ahead of fraudsters. This was the main theme of a recent panel at the Money Experience Summit unpacking how data can be used to prevent fraud without adding friction for legitimate customers.

Here are the key takeaways from the discussion. 

Knowing Who (or What) Is at the Door

One of the panel’s central questions was how to distinguish between legitimate users, malicious actors, and growing numbers of automated agents. Karan Gandhi, Senior Director of Credit Strategy at Best Egg, emphasized the importance of persistent, device-level intelligence. He noted that biometrics and other device-level identity authentication methods present a formidable challenge to anyone or anything trying to pose as a known and trusted customer. 

But while this problem can be lessened at the device level, this isn’t a single data point problem. IP addresses, device fingerprints, behavioral patterns, and authentication history all matter. Combating the future of fraud requires multiple signals taken together. 

Behavioral Signals: How Humans Give Themselves Away

One of the most compelling insights from the panel was how behavioral data can reveal whether an interaction is human or automated.

“If you had to apply for a loan in my name, it would take time to type it,” Gandhi explained. “But, if you type your own name, you’ll do it at a certain speed.” 

Platforms can analyze typing cadence, mouse movement, copy-paste behavior, and time-on-task to detect anomalies. These subtle signals are extremely difficult for bad actors, and even AI agents, to perfectly replicate. 

Utilizing data channels to monitor and examine these “biometric behaviors” is a perfect example of how fraud can be combated at scale. This kind of intelligence helps institutions spot fraud without constantly interrupting legitimate users.

This also goes beyond logins and identity verifications to help manage fraud and risk. Financial institutions and fintechs can also look at spending behaviors, transactions, and more to spot anomalies if an identity or account has already been compromised. 

Achieving Protection Without Adding Friction 

One of the most important topics the panel discussed was how to balance security against friction for legitimate users. Essentially this means not trading the user's experience for greater security. The vault that is too good keeps out the owner too.  

One main point was the need to move away from one-time authentication towards continuously-updated risk profiles. Brian Bender, general manager of partner solutions at Alloy, described this as building “a dynamic system for how to monitor the risk profile of customers over time.” 

The goal should be to not degrade the user experience unless something actually looks wrong. When behavior shifts or patterns match known fraud signals, then it’s time to step up verification.

That same logic applies not just to customers, but to vendors and AI agents as well. Nicole Lauredan, Head of Product Partnerships at Stripe, emphasized the importance of ecosystem-level vigilance against fraud. By ensuring that third-party integrations and data providers are held to the same fraud and risk standards as internal systems, fraudsters have fewer gaps to wriggle through.  

Data Is the Tool. People Make it Powerful.

Access to data is not what separates effective fraud prevention from reactive response. It comes down to how that data is used. 

Fraud prevention succeeds when data is placed in the right hands and applied with context, discipline, and judgment. It requires teams that understand which signals matter, how they fit together, and when to intervene without disrupting legitimate customers. 

When used thoughtfully, data allows institutions to move from static defenses to adaptive strategies that evolve alongside fraud itself. Check out the replay of the full discussion to hear more key insights.