Experience mod verification before quote — the workers' comp mod rebuilt from the sources, every difference explained
Every mod checked before quote, each difference shown line by line.
See one case, screen by screen ↓Checking a workers' compensation experience mod against its sources before you quote: the bureau's experience rating worksheet against the prior carriers' loss runs, the final and revised audits, and the ownership record. Experience Mod Verification rebuilds the mod line by line beside the bureau's figure and explains every difference.
How to choose AI solutions for insurance →
Updated
A workers' compensation account is priced on its experience mod. The bureau's worksheet compares the employer's actual losses with the losses expected for its payroll by class code, over three policy years — and it is only as good as the unit statistical reports the carriers sent. A claim coded lost time that never paid a day of indemnity. A reserve from the first report on a claim that has since closed. Installer payroll audited into the sales class. An acquisition nobody reported on an ERM-14.
Rebuilding the worksheet means keying every class line and claim, matching each one to the prior carrier's loss runs, tying payroll to the final and revised audits, and checking the ownership record against Rule 3. Underwriters rarely have the hours, so the mod is taken on trust unless the broker complains. If it is wrong, the account is either lost or under-priced — and the error stays in the mod for three years.
Estimated days for one underwriter to check every mod across a month of 23 quotes — about 2 hours each by hand, about 6 minutes each with the solution.
Six agents read the worksheet, loss runs, audits and ownership record and rebuild the mod; the underwriter decides and signs.
Harlow Ridge Mechanical, a commercial HVAC and plumbing contractor in Marietta, Georgia, is being remarketed from Quillmere Casualty. The bureau says 1.22. The quote is due Friday 16 October. Here is what happens next, screen by screen, in the working solution.
Renee opens Accounts before quote: nine workers' comp quotes in the next three weeks, each with the bureau's mod and the rebuilt one on the same small scale. Five have a difference, two tie, one waits on a bureau ruling and one is still being read. Harlow Ridge shows 1.22 → 1.15 and four differences.
“Medical-only claim coded lost time · stale reserve · payroll class · ERM-14 missing.”
The account opens on the mod rebuild: the bureau's 1.22 and the rebuilt 1.15 on one scale, −0.07, worth $30,002 a year on this quote. A bridge walks from 1.22 to 1.15 one difference at a time, and below it the worksheet is rebuilt line by line — the bureau's figure on the left, the figure from the source on the right.
The worksheet reader recomputes 1.2244 from the bureau's own lines — the arithmetic ties. The loss run reader matches 20 of 20 claims and flags three. The payroll and class matcher ties 15 class lines to three audits, and the ownership checker finds an acquisition with no ERM-14. Then the mod rebuilder tags every claim.
“Question and note drafter is waiting for you — 3 questions and the file note drafted · nothing sent.”
Claim 24-0387, a palm laceration on ductwork, opens beside Quillmere's loss run valued 31 Aug 2026: $17,640 medical, $0 indemnity, 0 days lost, closed 19 Mar 2025. The unit report kept injury type 5. As lost time, all $17,640 counts as primary loss; as medical only it enters at 30% — $5,292 — under the experience rating adjustment.
“Returned to light duty the next shift. A $2,400 indemnity reserve set at first notice was released on 3 Jan 2025; no indemnity was ever paid.” Its tag falls from +0.07 to +0.02 mod.
On Claims, all 20 claims are ranked by their mod impact. Renee puts 24-0387 back to lost time in the what-if: its tag goes to +0.07, the rebuilt mod to 1.20 and the premium to $514,325 a year. Injury type, incurred and a subrogation recovery can all be tested; reset returns it to the loss run.
Claim 23-0216 sits on the worksheet at $31,500 from the first report; it closed on 6 Feb 2026 for $12,800, and the second report was filed after the worksheet was issued. That is −0.04. Claim 24-0077, a knee, reopened on 11 Mar 2026 and is now $15,300 against the worksheet's $8,200 — +0.03, a difference against the employer, shown just the same.
The top four claims carry 61% of the losses in the mod — the argument starts with the longest bars.
For 2024–25 the worksheet carries $418,000 of installer payroll under 8742, outside salespersons. Quillmere's revised final audit of 22 Jul 2026 moved it to 5537, HVAC installation, after Harlow Ridge disputed the first audit. Expected losses rise $4,096; the mod falls 0.01.
“The payroll matcher reports the audit's class; it never re-classes payroll itself. Luis Ortega checked the job descriptions on 9 Oct.”
Harlow Ridge Holdings acquired Pell Creek Plumbing LLC, 100%, on 3 Mar 2025, and its 14 plumbers moved over on 1 June. No ERM-14 is on file. The Rule 3 tests point to combining Pell Creek's experience — an estimated 1.13–1.14 — but that estimate stays out of the rebuilt figure until the bureau rules.
Who decides: “The bureau, on a signed ERM-14.”
To Gavin Hale at Brennick & Hale, the ERM-14 for Pell Creek. To Quillmere's unit statistical reporting, through Gavin, a correction to 24-0387's injury type, confirmation of the 2023–24 second report and a 2024–25 payroll correction. To the bureau, a request for a revised rating for 1 Dec 2026. Renee edits, then sends.
“Nothing is sent until you send it.”
The mod verification note records the four differences, their sources and the open ownership question. Renee picks the recommended decision — quote at the promulgated 1.22 with the mod-revision condition, tell the broker to expect a revised 1.15 — and signs. The difference is over $25,000 a year, so Owen Pryce, her manager, acknowledges before the quote goes to Gavin Hale.
The same account, seen by the five people who carry it — what they said before, and what they see now.
Six agents read the worksheet, loss runs, audits and ownership record and rebuild the mod; the underwriter decides and signs.
Reads the bureau's experience rating worksheet — NCCI or a state bureau's format: policy years and report levels, class lines, claims and the split point, W and B — and checks the printed mod against its own lines.
Reads prior-carrier loss runs in any layout — valuation date, status, medical and indemnity paid and reserved, days lost, reopen dates — and matches each claim to the worksheet.
Ties worksheet payroll to the final and revised audits by class and state, and reports payroll that differs or states missing from the rating.
Reads the submission, state filings and ownership records for changes in the last four years and applies the Rule 3 tests.
Recomputes the mod from the sources with the worksheet's own parameters — medical-only claims at 30% in states with the experience rating adjustment — tags every claim with its mod impact and builds the bridge from the bureau's figure.
Drafts the questions to the broker, prior carrier and bureau, and the mod verification note, each citing the lines it rests on.
Decides and signs. The agents propose; a named person decides.
Anyone on the account can ask in plain words — or have it draft the questions. Every answer cites the worksheet line or loss-run page it rests on.
Why is claim 24-0387 worth +0.07 on the mod?
The worksheet carries it as lost time, so all $17,640 counts as primary loss: $17,640 ÷ $251,301 = +0.07. Quillmere's loss run shows $17,640 medical, no indemnity paid and 0 days lost, closed 19 Mar 2025. As medical only it enters at 30% — $5,292 — and its tag falls to +0.02. That one correction lowers the mod by 0.05.
What would the mod be if every correction is accepted?
1.15, from the bureau's 1.22: claim 24-0387 −0.05, claim 23-0216 −0.04, claim 24-0077 +0.03, payroll 2024–25 −0.01. If the bureau also combines Pell Creek Plumbing, the estimate is 1.13–1.14 — not in the figure until it rules. At 1.15 the standard premium is $492,895, $30,002 a year below 1.22.
Which claims are driving this mod?
Four claims carry most of it: 24-0019, service van rear-ended, $31,200, +0.09; 22-0131, fall from ladder, $27,900, +0.09; 25-0033, crush injury, $24,750, +0.09; 24-0077, knee, $15,300, +0.06. All four are lost time and three are above the $21,500 split point; they tie to the loss runs. The rebuilt mod is still 1.15 — above average — because of them.
Draft the questions for Gavin Hale and the bureau
Three are ready, each citing its line: the ERM-14 for Pell Creek Plumbing to Gavin Hale; unit report corrections to Quillmere through Gavin — injury type on 24-0387, the second report for 2023–24, and the 2024–25 payroll; and a request for a revised rating to the bureau. Nothing is sent until you send it.
13 screens from the working solution, on its sample data. Pick one to see it large.
What counts as a difference, which bureaus, how far back to look and who signs are settings. Changes are versioned and on the audit trail.
| Setting | Default | Choose from |
|---|---|---|
| Mod tolerance — show a difference when it moves the mod at least | 0.01 | 0.01 to 0.05 |
| Claim value tolerance — any claim whose incurred differs by more | $1,000 | $500 · $1,000 · $2,500 |
| Loss-run freshness — ask for new loss runs when older than | 90 days | 60 · 90 · 120 days |
| Medical-only reduction — apply the 70% experience rating adjustment in states that use it | On | On · Off |
| Rating bureaus — worksheets read in each bureau's format | NCCI · NCRB · PCRB · WCIRB · NYCIRB | and 6 more |
| Ownership look-back — ask for an ERM-14 on any change in | 4 years | 3 · 4 · 5 years |
| Refer to the manager when premium moves by more than | $25,000 a year | $10,000 · $25,000 · $50,000 a year |
| Questions to the bureau | Sent by the underwriter of record only | Underwriter |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample account data · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (Business Insurance, workers' comp audit reviews — about one in three mods contains a mistake, by one broker consultant's count; NCCI Experience Rating Plan — the three-year experience period, the formula, the experience rating adjustment and Rule 3 ownership changes). People, companies, carriers and accounts named on this page are fictional — characters and sample data in the working solution.
Checking a workers' compensation experience mod against its sources before you quote: the bureau's experience rating worksheet against the prior carriers' loss runs, the final and revised audits, and the ownership record. Experience Mod Verification rebuilds the mod line by line beside the bureau's figure and explains every difference.
It uses the parameters printed on the worksheet — split point, W, B, expected loss rates and D-ratios — never assumed ones. It computes expected losses from audited payroll and actual primary and excess losses from the loss runs, with medical-only claims at 30% in states with the experience rating adjustment, and builds a bridge from the bureau's figure one difference at a time.
Each claim's share of the mod: its primary loss plus W times its excess, over expected losses plus ballast. It is printed on the claim — “+0.07 mod” — so the underwriter, the broker and the bureau argue about the claims that matter.
No. Quotes use the bureau's promulgated mod. The rebuilt figure is labelled as rebuilt, not official, and recorded in the note; when it differs, the recommended decision is to quote with the mod-revision condition and request the corrections from the prior carrier and the bureau.
Yes. The ownership checker reads the submission, state filings and ownership records for changes in the last four years and applies the Rule 3 tests. It never decides whether experience combines — the bureau does, on a signed ERM-14 — and any estimate is kept out of the rebuilt mod.
Worksheets are read in each bureau's own format: NCCI, NCRB, PCRB, WCIRB, NYCIRB and six more, as set in Settings.
Yes. The agents read, rebuild and draft; the underwriter sends every question and signs the note. When the mod question moves premium by more than $25,000 a year, the underwriting manager acknowledges it before the quote goes out.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your underwriting guidelines, your bureaus and a sample of past worksheets and loss runs — and it goes live once every quality gate has passed. We will walk you through it on your own accounts first.
Submissions ready to price the same day, poor fits declined in minutes.
1-min demoLoss Run Analysis AutomationLoss summaries in minutes, not an hour per account, each claim traced.
1-min demoSchedule of Values ValidationEvery location geocoded and its COPE gaps flagged before the property is priced.
1-min demoWe'll run Experience Mod Verification on a sample of your own worksheets and loss runs.