How Coast Pay built a fraud model around trusted legitimacy signals

Coast needed a fast, consistent way to determine whether an applicant was a real, operating business. TrueBiz became a trusted legitimacy signal within its business onboarding workflow.

AT A GLANCE

Coast provides expense management and corporate cards for trade businesses across the United States. As the business grew, it needed a faster, more consistent way to assess applicant fraud risk.

The problem: gray-area applicants were hard to act on with confidence; first-party fraud, where a real person sets up a real-looking business purely as a front, was difficult to catch through credit and identity checks alone.

The result: an objective legitimacy score built into Coast's own internal fraud model; fewer documents requested from legitimate applicants; a risk team that has stayed lean through a significant spike in application volume.

THE CHALLENGE

Keeping losses low without slowing down growth

As Director of Risk & Analytics at Coast Pay, Delia Nunez's mandate cuts both ways: keep losses low, but don't slow down good spend.

"If keeping losses down was our only goal, I'd decline every application and my job would be done", she says. The real goal is getting in as many good applicants as possible "in a way that's as frictionless as possible, while still keeping losses sensible for our margins". 

Judgment calls, one browser tab at a time

Before TrueBiz, gauging a business's legitimacy meant manual research: checking online customer reviews, LinkedIn presence, and whether a website looked real. "That review would take quite a bit of time, and it wasn't straightforward", Delia says. It took up to five minutes per case, roughly a fifth of a full fraud review, and it didn't always catch what mattered: some sites that looked fine at a glance turned out to be freshly registered domains with empty links behind them. The result depended heavily on individual judgment, and left little confidence on the gray-area cases.

The hardest calls were first-party fraud: someone using their own identity, sometimes with real credit lines and bank statements elsewhere, to set up a business purely as a front. "The whole purpose of setting it up is to eventually defraud whoever is lending to them", Delia says. Because everything else about the applicant checks out, this type of risk is difficult to catch through credit and identity checks alone.

THE SOLUTION

Choosing automation over more hires

Coast's philosophy is to automate wherever possible and stay lean. "We believe in as much infrastructure as possible", Delia says. "If you really require manual review, that's when you bring someone in". TrueBiz fit that approach: "We didn't want to hire more people just to do certain reviews we knew we could get with more objective data".

How TrueBiz fits into Coast's own fraud model

TrueBiz sits inside the custom fraud model Coast has built: "We have our own internal scoring system for fraud", Delia says, "and TrueBiz is an integral part of that system". Every applicant under consideration for onboarding runs through a TrueBiz check as standard practice.

What set TrueBiz apart was the depth of insight and the rigor of its scoring methodology, wrapped in a simple, easy-to-use interface. "Other vendors try to do everything at once, and end up doing a little good in every area, but not great at any one thing", Delia says. "With TrueBiz you get a view on the legitimacy of that website, a score, and the attributes behind it". As the product has grown, with social profiles, reviews and a connected-entities network map added over time, the original score has stayed consistently reliable, which made it easy to fold into Coast's own risk stack.

For Coast, the appeal wasn't just saving time, it was consistency: the same data and scoring methodology regardless of who was doing the review. "With TrueBiz, it's not just the time saved", Delia says. "It's the confidence, the accuracy, and the fact the methodology is precise. It doesn't change per person".

THE IMPACT

Turning gray-area guesswork into confident decisions

Asked to sum up the value in one line: "A simple, objective view on online legitimacy”. 

Before TrueBiz, some cases were hard to call: "It was really difficult to have any confidence on the gray-area cases", Delia says. "Should we ask for bank statements? Should we do more? With TrueBiz, we can say: I'm okay taking that risk, or I'm not. It’s binary".

That confidence runs in both directions. A strong TrueBiz score means Coast can move a legitimate applicant through without extra scrutiny. A weaker score is one of the strongest signals the team has for first-party fraud. "If it looks like a real website but it's a TrueBiz fail, that's a meaningful signal," Delia says. "If I'm looking at first-party fraud specifically, we're definitely looking at TrueBiz".

The team's trust in the score runs deep enough that they rarely look past it. "If it's a TrueBiz pass, it's rare when we question it", Delia says.

More application volume, less paperwork, lean team

Coast has grown considerably since Delia joined nearly four years ago. "The volume we saw in a week is what we now see in a day", she says. Through that growth, the risk team has stayed deliberately small, leaning on tools like TrueBiz rather than adding headcount for reviews that can be handled with objective data.

"I don't think you can automate everything. You'll always need a human in the loop, because trends change", Delia says. "But we've been able to keep the team very, very lean".

That same efficiency shows up for applicants too, as a faster, simpler process. Revenue-verification documents, like bank statements, used to be a fallback whenever a reviewer wasn't sure what they were looking at, adding time for businesses that had nothing to hide. "Once we had confidence with TrueBiz, that became a key reason we didn't need to ask for additional documents", Delia says.

Advice for other risk leaders

Delia’s advice to other risk leaders reviewing technology providers:

“Be curious. Don't take solutions at face value: go dig into the methodology and question the scores. And if you ever think, I wish someone would build something for this, probably someone has".

Want to see what an objective legitimacy score could bring to your own onboarding process? Request a demo.