Small business owner or lending fraudster?

On paper, they can look almost the same. But in reality, they’re night and day. One is an honest, hard-working small business entrepreneur who isn’t going to let occasional setbacks determine their long-term success or value. This borrower can be a good bet for your portfolio. They might be the next customer whose business takes off and stays with you for years. The other is a criminal whose only goal in life is to trick you out of capital.

Spotting the difference can require a level of insight beyond what snapshots of data provide during a typical small business loan underwriting process.

The reason is twofold:

• Fraud detection usually falls into two camps: either consumer-focused or commercial-focused. Small businesses live in a gray zone between the two.
• Fraudsters are evolving and getting more and more sophisticated in exploiting the gray zone, blending identities and obscuring connections hence hiding the shadows of its data. They’ve even started using AI to help them.

We’ve seen this dynamic developing for quite a few years now. About 10 years ago, the data we were accumulating was growing as fast as our client base. But our attempts to query that data for evidence of fraud did not do a better job than our old manual and reactive fraud detection tactics. The dots just didn’t connect.

So we took a more holistic view of the data and what we discovered changed everything for our fraud identification capabilities. Instead of looking at attributes and entities in isolation, we looked at the entire ecosystem. The true complexity of small business financing can include a jumble of corporate entities, SSNs, EINs, DBAs, bank accounts, addresses, phone numbers, payment histories, principals, etc. 

As we started to connect the pieces across both individual and business attributes, patterns emerged over time. The overlaps of data told stories we had missed before. Because they had context not previously illuminated in traditional underwriting research.

We discovered why fraudsters loved to game the small business lending system. They could easily obscure their motives in a shell game of identities. But, by doing so, they also betrayed distinctive patterns. If you knew what to look for.

So… we created a system that was more intelligent in layering data on itself to identify patterns specifically related to the complexities of small business financing. Then we started using it in our internal review of applications. Sure enough, the flags started appearing.

We named this system Lynx.

Lynx doesn’t rely on static checks or binary flags. It examines every attribute, across both the individual and the business, and looks for data integrity, consistency, and patterns—especially the kind that fraudsters try to hide.

We started using it internally. And the results spoke for themselves:

• We flagged bad actors faster.
• We separated persistent entrepreneurs from serial fraudsters.
• We made more confident lending decisions.
• And most importantly—we unlocked capital for the good actors who deserved it.

But we knew this was just the beginning. We were off to a great start. But we quickly realized that we needed something very important to take it to the next level. More data. Third Party data and real time intelligence. And much more.

If you would like to hear more about how Lynx can strengthen the anti-fraud protection systems in your small business lending, please reach out to enterprisesales@rapidfinance.com.

Lynx gives you a clearer view of small business loan applicants.