One Approval Event. Fourteen Months. A Lot Can Go Wrong in Between.

A colleague of mine got deposed last spring. The plaintiff's attorney had their carrier file open on the table. The approval event — the onboarding screening, the insurance verification, the carrier packet — was timestamped fourteen months before the crash.

The attorney's first question wasn't about the BASIC scores. It wasn't about the insurance certificate. It was: "In the fourteen months between this approval and the incident, how many times did your company re-verify this carrier's safety standing?"

The answer was zero. Nobody had looked again after onboarding.

The attorney pulled out a printed FMCSA compliance snapshot. Seven weeks before the crash, that carrier had received an out-of-service order on a vehicle maintenance violation. My colleague's company had used that carrier on eleven loads since the OOS order was issued. Eleven loads, zero re-checks.

That's what a static approval process costs you.

---

The post-Montgomery v. Caribe Transport II world has changed what it means to "vet a carrier." The Supreme Court's unanimous May 2026 ruling confirmed that brokers can be sued in state court under a negligent-selection theory. The immediate response in the industry was a rush to improve initial vetting — better MC pulls, insurance verification, BASIC score reviews at onboarding.

That response is right, but incomplete. Initial vetting is necessary. It isn't sufficient.

The negligent-selection standard isn't "did you screen this carrier once?" It's "did you exercise reasonable care in the selection of this carrier for this load on this date?" The word selection in that sentence is doing real work. If you selected a carrier fourteen months ago and haven't looked since, you're relying on reasonable care exercised fourteen months ago to defend a decision made yesterday. That's not a strong position.

What an Approval Event Actually Is

Every broker has some version of a carrier approval process. You run the MC pull. You verify insurance against the FMCSA L&I database. You pull the BASIC scores. You get a signed carrier packet with the broker-carrier agreement, check a few ACORD 25 boxes, confirm operating authority. At some point you make a judgment: this carrier meets our standards. Approved.

That judgment — and the data behind it — is an approval event. It's timestamped. The data underlying it was accurate on that date.

But it's not a subscription. The carrier's FMCSA record keeps moving. Their insurance situation keeps changing. Their maintenance program either holds together or it doesn't. Their drivers accumulate inspection history. Their BASIC scores recalculate every month. Every load you tender after that initial approval event is tendered to a carrier you were comfortable with — not necessarily a carrier you are comfortable with.

Most brokers treat the approved list as a permanent state. A carrier gets on it and stays on it until something catastrophic happens — usually until after a loss or a lawsuit forces a review. What they're missing is that the FMCSA keeps running, and so does the clock.

What Changes After You Approve a Carrier

Let me be specific about what can shift between your approval event and your next load tender.

Insurance. A carrier's cargo insurer can file a cancellation notice (BMC-35) with FMCSA with 30 days of advance notice. The policy is still nominally "on file" during that 30-day window — the FMCSA L&I database might show it as current the day before the effective cancellation date. If you tender a load inside that window and cargo is lost, the insurer will decline the claim. Insurance status is not static.

Safety rating. FMCSA can downgrade a carrier from Satisfactory to Conditional following a compliance review. A Conditional rating doesn't automatically revoke authority, but it's a material change in the carrier's risk profile. A Satisfactory rating at onboarding and a Conditional rating at load tender are two different risk decisions. If the rating downgrade happened between those two dates and you didn't catch it, your file reflects the old rating.

BASIC threshold alerts. FMCSA recalculates BASIC scores monthly. A carrier who was clear in all categories at onboarding might push through an alert threshold by month seven. If nobody re-checked, the approval says "clear" while the current FMCSA record says something different.

Out-of-service orders. An OOS order can come down on a specific vehicle, a specific driver, or the carrier's entire operation. A vehicle OOS order for brake violations, for example, gets recorded against the carrier in the Vehicle Maintenance BASIC — but the order itself is distinct from the BASIC score recalculation. An OOS order in the system is a concrete enforcement event, not just a statistical percentile.

Ownership and officer changes. A carrier can change hands — principal officers, controlling owners — without transferring their MC number. The authority number stays the same. The people running the operation may be completely different. FMCSA's MCS-150 is supposed to reflect current officer information, but there can be a lag between a change in control and an updated filing. If the carrier you approved was run by one set of principals and is now run by someone else entirely, your carrier relationship has substantively changed.

None of these are theoretical. All of them happen regularly in the carrier base. The question isn't whether something will change in your approved carriers' records — it's whether you'll know when it does.

The Problem With a Single Document Record

Your carrier file, if you've done the work at onboarding, tells one story. It shows that on a specific date, you collected a specific set of data and made a specific approval decision. That's useful evidence. It's not a complete defense.

Here's the deposition question that makes single-event files vulnerable: "At the time you tendered this load, what was the current status of this carrier's safety record?"

If your answer is "the same as it was at onboarding fourteen months ago, because we didn't look again," you've just told the plaintiff's attorney exactly what they need to know. You had a duty to exercise reasonable care. Reasonable care doesn't have a one-time exemption. You exercised care once, and then you assumed nothing had changed for over a year.

Under 49 CFR § 371.3, brokers are required to keep records of each transaction — carrier name, authority number, rate, shipper — for three years. That's the regulatory floor. It says nothing about ongoing monitoring obligations. But the post-Montgomery negligent-selection standard is a common-law tort standard, not a regulatory checklist. What a reasonable broker would do in 2026 is the question, and a reasonable broker in 2026 has access to real-time FMCSA data. Choosing not to use it is a choice the jury gets to evaluate.

What a Re-filed Carrier Selection Record Changes

The Carrier Selection Record is the document that answers "why did you choose THIS carrier for THIS load on THIS date." Done right, it's not just the carrier's BASIC scores and insurance — it's the affirmative reasoning: this carrier's safety record is current and clear, their authority is active, their insurance is verified against the FMCSA L&I database, and here's what the data looked like when I made this decision.

Most brokers have one of these per carrier, at onboarding. The better answer is to have one per material change.

When FMCSA data changes in a way that's relevant to a carrier's risk profile — insurance carrier change, BASIC threshold movement, authority modification, OOS order, rating change — you should be generating a new decision event. Either: we reviewed the change, found it within our acceptable parameters, and we're clearing the carrier for continued use. Or: we reviewed the change, it falls outside our parameters, and the carrier is suspended pending resolution.

That second record — the re-evaluation — is the one that proves you were actually watching. A plaintiff's attorney can attack a single approval event. It's harder to attack a file that shows you reviewed the carrier on seven separate occasions over fourteen months, each time based on current FMCSA data, and each time made a documented, timestamped decision.

The first record says you were careful once. The sequence of records says you were careful consistently. Those are not equivalent in a deposition.

What DOTScreener Does Here

I built the Continuous Monitoring feature specifically because I saw this gap in how most brokers operate. You do the approval work once. Then you trust that nothing changed. But nothing in the FMCSA system promises you that.

Continuous Monitoring watches your approved carriers for changes in authority status, insurance filings, and safety data. When the system detects a material change — a different insurer on the L&I filing, a new OOS order in the record, a BASIC score crossing an alert threshold — it flags the carrier for your review.

You log in, you see the flag, you look at what changed, and you make a decision. That decision gets recorded with a timestamp. If you clear the carrier, the record shows you reviewed the specific change and found it acceptable. If you suspend them, the record shows that too.

What you end up with, over the life of a carrier relationship, is a documented sequence of evaluations — not a single stale snapshot. Every evaluation is based on data that was current the day you ran it. The record of each evaluation answers the deposition question: "at the time you made that decision, what did you know, and how did you know it?"

That's the file you want in front of a jury. Not a packet from fourteen months ago.

How I Document This

When monitoring flags a carrier for re-review, I log the following:

What changed. Specific data point — "cargo insurer changed from XYZ Specialty to ABC National, effective 30 days from filing date" or "Vehicle Maintenance BASIC crossed alert threshold at 77th percentile" or "active OOS order recorded, vehicle maintenance violation." Not just "flagged for review." The specific thing.

What I checked in response. If an insurer changed, I run the L&I database to confirm the new carrier is a licensed insurer and the filing is clean. If a BASIC crossed alert, I look at the underlying violation codes — specifically, whether the violations are concentrated in the last 90 days (a recent operational problem) or spread across 24 months (a chronic pattern). If an OOS order, I check whether it's vehicle-specific or operation-wide.

What decision I made. Clear for continued use, or suspended pending resolution. If cleared, I note the specific basis — "new cargo insurer verified, clean L&I filing, BIPD minimum confirmed, cargo limit confirmed at $100K." If suspended, I note what resolution would look like — "carrier reinstated when OOS order is lifted and current BASIC data available."

Date and my name on every entry.

That takes five minutes when monitoring flags something. And when a plaintiff's attorney puts the carrier file on the table, I can walk them through it decision by decision.

A 14-month-old approval event doesn't hold up under that kind of scrutiny. A documented re-evaluation history does.

---

— Mason Lavallet

Founder, DOTScreener.com