Freight fraud doesn't play fair, and the tools it hides behind keep getting better. This was a big week for sharpening how you catch it without slowing down the honest carriers you actually want to onboard. Here's what's new.
Flag exactly what you mean to, signal by signal
One on/off switch was never enough. A VPN on a carrier's connection isn't the same threat as a location that jumps two states mid-session, or a browser built specifically to defeat detection. But a single catch-all control treated them all alike, so you either waved through real fraud or held up good carriers.
Now you tune each signal on its own. VPN, proxy, and location-spoofing are three independent controls, and the carrier is told the exact thing to turn off, so a location-spoofer never gets told to disable a VPN they never had. Automated-browser activity, device velocity, repeat failed attempts, operating-authority status, changed registered contacts, recent authority filings, and equipment- or authority-for-sale listings each get their own setting too. For every one: allow it, send it to review, or route it to your fallback handoff. Set your risk appetite signal by signal, and change it whenever your fraud picture shifts. No code, no deploy, no support ticket.
We also added a control for anti-detect browsers, purpose-built evasion software with no honest use in carrier onboarding. It now gets flagged and routed to your fallback by default instead of sailing through untouched. That's roughly 1 in every 95 sessions you were never seeing.
A faster carrier check that fewer drivers abandon
Every extra second on a verification screen is a carrier who walks away. So we went after the wait.
Your carriers now reach the first screen about 1.3 to 1.5 seconds faster, because we prefetch what the widget needs before the carrier even taps to begin. The bigger win is what happens when the government carrier records slow down during busy hours, through no fault of your carrier. Instead of making a legitimate carrier sit and wait, we serve a recent, trusted copy the instant the live lookup runs long. That trims roughly two seconds off the slower end of lookups and fixes stalls that were hitting about 1 in 8 checks.
None of this trades away protection. Every identity and authority signal (revoked, inactive, out-of-service) still fires unchanged. Slowness alone can never fail a first-time carrier. We just stopped making honest carriers pay for slow infrastructure, so more checks finish and fewer onboardings get abandoned.
More of your legitimate carriers actually get verified
Every good carrier you turn away is a load you can't cover. So we closed the gaps that used to strand them at the verification screen, with no configuration required on your end.
-
No phone on the FMCSA record? When the email checks out, the carrier verifies on email alone and finishes, instead of dropping to manual review. The email path only opens when the email is genuinely reachable and the session carries no risk signal, a risk flag can never remove a requirement, and you always see when a check finished on email alone. Nothing quietly passes as more than it was.
-
MC-started checks echo the exact MC you launched with (including the newer randomized MCs that carry no MC on the FMCSA record), so your MC-keyed system always ties our answer back to your original request.
See where your carriers drop off
Your daily and weekly email digests now open with your whole funnel as one picture: everyone who showed up splits cleanly into Verified, Caught, and Didn't finish, with warning-sign carriers called out right inside the Verified tile, the number that tells you which policy to tighten.
"Didn't finish" is honest now, too. You can see who left before ever asking for a code versus who got a code and never entered it. Two very different stories: the first is often an impersonator hitting the wall, the second is frequently a stale FMCSA number that couldn't reach a real carrier. In 30 days of live traffic, 35 of 43 carriers who didn't finish never even requested a code.
And if you'd rather watch it all in the tools you already use: drop your GA4 Measurement ID into the console and your entire verification funnel flows into your own Google Analytics. Every embed you run, no code changes, and zero PII ever leaves. Only non-identifying fields like DOT, tier, status, outcome, and timing.
Just shipped: we don't trust the record about what your phone can do
The FMCSA record hands us a phone number and calls it a "cell," but that label isn't reliable. Plenty of numbers filed as a cell are actually landlines or VOIP lines that can't receive a text at all. Taking the record at its word meant a carrier could tap "text me a code" and sit waiting on an SMS that was never going to arrive.
So we stopped taking its word for it. Now we check the real line type ourselves, in the moment, and only offer a text when a text can actually land. A landline gets a voice call instead of a dead button; mobile and VOIP numbers still offer both. It happens at session start, so there's zero added wait. Just fewer failed sends and carriers who used to hit a wall now getting through.
Try it
If you're already running Verify, these changes are live. Open your latest digest and take a look at where your carriers drop off. If you're not, we'd like to show you what your real carrier traffic looks like once the fraudsters can't hide in it. Reach out and let's talk, and we'll get you set up in an afternoon.