How the CarrierMark Score is calculated
CarrierMark grades a motor carrier two ways, regulatory compliance and litigation exposure, from a versioned, CFR-cited rule library. This is the published method behind every grade, so an insurer, broker, or carrier can see exactly what a score means and how it was derived.
Two scores, never collapsed
Every carrier carries two independent scores. They are never averaged or merged into a single number, because they answer different questions, and the gap between them is the most important signal CarrierMark produces.
Whether the carrier would pass a new-entrant safety audit or compliance review under 49 C.F.R. Part 385. Driven by the regulatory severity of open findings.
If there is an accident, how strongly the file supports a negligent hiring / retention / entrustment / supervision case, and whether the pattern reaches gross-negligence or punitive territory. Driven by the litigation weight of open findings.
The inputs: a versioned, CFR-cited rule library
Both scores are computed by a rule engine over the carrier's data. Each rule is data, not code, and carries a specific CFR citation and an effective date, so any past grade is reproducible against the rule library that was in effect at the time.
v1 covers the driver and fleet surfaces with the highest litigation risk:
- 49 C.F.R. Part 391Driver qualification: applications, MVRs, medical certificates, road tests, prior-employer investigation, annual reviews.
- 49 C.F.R. Part 382Drug & alcohol program and the FMCSA Clearinghouse: pre-employment full queries and annual limited queries.
- 49 C.F.R. Part 396Inspection, repair & maintenance: pre/post-trip DVIRs, the periodic annual inspection, and out-of-service defect resolution.
- 49 C.F.R. Part 395Hours of service and ELDs: violation summaries and unassigned-driving reconciliation.
Each rule resolves to one of four declarative checks:
Compliance Status: how it grades
Open findings are weighted by their regulatory severity, how directly the gap maps to a citation in a DOT audit, and rolled into a 0–100 status, then a letter grade and a pass/at-risk/fail band.
Litigation Exposure: the second score
This is the part no other compliance tool produces. Every gap is mapped to the negligence theory a plaintiff would build on it, the element of that theory it helps prove, and a base weight that encodes a trucking-defense lawyer's judgment about how badly it plays to a jury.
- Base weight: the defense-lawyer judgment for the requirement and the theory it supports (hiring, qualification, retention, supervision, maintenance, entrustment, spoliation).
- Causal match: a gap connected to the alleged cause of the crash (a fatigue theory meeting an HOS gap) weighs far more than an unrelated one.
- Incident severity & venue: a paperwork gap is latent until a crash gives it weight; the jurisdiction's tilt then amplifies or dampens every theory.
Per-subject exposure rolls into a band, and a punitive / gross-negligence flag trips on the patterns courts treat as conscious indifference: spoliation, a known lapse left uncorrected, or a pattern of serious gaps.
Verify before trust
No extracted value updates a scored compliance table until it clears a verification gate. Every grade traces back to a specific document on file: an application, a medical certificate, an MVR, an inspection report. Not an unconfirmed guess. Low-confidence or rule-failing fields are flagged for one-tap human confirmation first.
Versioned & reproducible
The rule library is versioned with effective dates, and an append-only audit log records every change to the underlying file. A grade computed today can be reproduced exactly against the library version in effect on its date, which is what makes the score defensible if it is ever questioned in an audit or a courtroom.
What the score is, and isn't
The CarrierMark Score is a risk-screening measure, not legal advice and not a guarantee of any audit or trial outcome. The litigation-exposure analysis is the kind of work product a defense lawyer prepares; carriers and their counsel should treat an exposure report accordingly. A grade reflects the carrier's file at the moment it was generated and changes as the file does.