Pilot programme · US freight
Know who you are hauling for — and be able to show why.
Sources disagree. Authority lapses mid-load. A payment instruction changes the day before pickup. CarrierCheckup resolves the conflict independently, applies your policy, and hands you an evidence packet that still makes sense three months later in a claim.
No self-service sign-up yet: we take pilot customers by hand so the work is measured before it is automated.
The problem
A clean report and a real answer are different things.
Vetting tools return a verdict. When two sources disagree, or a record is simply missing, a verdict has to invent something — and what it invents is usually a confident-looking green.
The expensive cases are not the obvious ones. They are the carrier whose broker authority is pending revocation but not yet revoked; the certificate that is on file for less than the load requires; the phone number that appears on three websites which all copied it from one place.
What other tools do
Missing record → PASS because nothing negative was found.
What we do
Missing record → INCOMPLETE with the exact field named, because no data is not good news.
Six answers, not two
Every check ends in a state you can act on.
Colour is never the only signal, and "we do not know" is a first-class answer rather than a failure to produce one.
Every mandatory claim is confirmed or denied by qualified independent evidence, with no live contradiction.
Mandatory rules are met, with limits you accepted — a cargo class, a coverage ceiling, a date.
Evidence conflicts, a step-up was not completed, or a person has to decide. We say which.
A current, confirmed prohibition applies to the role you are hiring for. Not a low score — a named fact.
A mandatory input is unknown or stale. The check is not finished, and it does not pretend to be.
A required source is down. That is an outage, and it is reported as one instead of quietly lowering the bar.
How it works
Four steps, and you can audit every one.
Identify
We resolve the counterparty by USDOT, MC or name. If two carriers are genuinely plausible, you get both — we never merge on a resemblance.
Gather
Registry facts, authority by kind, insurance required against on file, inspections and out-of-service history — each with the time its source last updated it.
Resolve
Where sources disagree, we contact an independently sourced channel and record what a qualified person actually said — or that nobody answered.
Decide
Your policy runs against the evidence, deterministically. The same snapshot always gives the same answer, and every override is recorded separately.
What you get
An evidence packet, not a screenshot.
Immutable and versioned. Fetch the same version twice and you get the same bytes and the same hash — which is what makes it usable in a dispute months later.
- Every claim with its status and the evidence behind it
- Independence lineage: which root each contact came from
- The policy version and the exact rules that matched
- What we could not establish, named rather than omitted
- Gate decisions: every allow and every refusal to contact
Being straight about it
What CarrierCheckup does not do.
A product that will not name its limits is asking you to find them yourself, at the worst moment.
It does not replace your vetting tool
Keep what you have. We answer the cases it returns as clean but unresolved, and the packet attaches to the load you already have open.
It is not a registrar
We are not FMCSA and we do not decide anyone's authority. We read the registry, record what it said and when it said it.
Answering a call proves nothing by itself
Reaching a number establishes control of a channel. Who the person is, and whether they may speak for the company, is separate evidence.
It is not a fraud score
There is no number to argue with. There is a list of facts, where each came from, and which rule it triggered.
Priced per resolved case, not per lookup.
Lookups and monitoring are included in your quota. You spend an allowance only when a case actually needs someone contacted.
Questions we get asked first
Where does the data come from?
Registry facts come from the FMCSA API. Resolved party and observation data comes through the Doft Intelligence API. Communications — calls and messages — execute through FreightBlast. We never read another product's database directly, and every capability we depend on has to prove it works, that we are licensed to use it, what it costs and who owns it before it can be called.
Do you call carriers on my behalf?
Only when an approved communications policy exists, the provider says the specific contact and purpose are allowed, and the check was made seconds — not hours — before dispatch. During the current pilot stage real automated calling is closed: R0 works on consenting test participants and synthetic fixtures.
What happens when you cannot reach anyone?
The case finishes INCONCLUSIVE with the reason recorded — no independent channel, no answer, or an ambiguous response. We do not lower the threshold to improve a completion rate, and you are not charged a second time for a case that had no accepted attempt.
Is my watchlist visible to anyone else?
No. Public registry records are shared; the fact that you are watching a particular counterparty, and anything you told us privately, is yours. A reference belonging to another tenant answers "not found" rather than "forbidden", because "forbidden" would already tell someone it exists.
How do I get in?
Send a request. We reply by the channel you name, talk through the cases you actually have, quote them, and take the first ones by hand. Accounts are created by us — there is no self-service sign-up during the pilot, on purpose.
Bring us the case your current tool called clean.
Two or three real counterparties is enough to see whether this is worth anything to you.