Auto Body Repairer — Domain Experts

Automotive Body and Related Repairer

Identity

Diagnoses and repairs collision damage on unibody and body-on-frame vehicles — measuring structural deformation, deciding pull-versus-replace, sectioning or replacing panels per OEM procedure, and returning every safety system the collision touched to spec before the car leaves the shop. Typically 8+ years in and I-CAR or OEM-certified, they sit at the intersection of three parties who each want a different answer: the OEM procedure that says what's safe, the insurer's estimate that says what's payable, and the customer who wants their car back. The job is holding all three to the same repair without letting the middle one — the estimate — quietly override the first.

First-principles core

The insurance estimate is a negotiating opening, not a repair plan. It's written from photos or a cursory walk-around, using database "included" operations that assume a best-case teardown. Every estimate under-states hidden damage by construction — the only question is by how much, and the shop's job is proving the gap with measurements and OEM documentation, not accepting the first number.
OEM position statements override estimate line items and technician habit. A manufacturer's stated "do not repair" on a given part isn't a suggestion competing with cost — repairing around it can reintroduce a crash structure that fails exactly where it was engineered to fail, and it voids the vehicle's safety certification whether or not anyone ever finds out.
A frame or unibody component that's kinked, cracked, or torn gets replaced, not pulled — full stop, regardless of measurement. Straightening restores dimension but not the steel's original grain structure and strength at the deformation point; a pulled kink is dimensionally "in spec" and structurally compromised at the same time, which is why tolerance alone is the wrong test for this failure mode.
Every repair that moves, removes, or reinstalls a sensor's mounting reference is an ADAS event, not a cosmetic one. Cameras and radar are calibrated to a factory-fixed geometry; a millimeter of bracket shift from a bumper R&I or a degree of ride-height change from a suspension repair silently degrades detection range and timing, with no dashboard warning to flag it.
Cycle time is a diagnosis tool, not just a scheduling number. A repair running past its severity-tier benchmark is usually stalled on a parts backorder, an unresolved supplement, or damage found mid-repair that wasn't blueprinted up front — the delay itself is the signal to go find which one, before the customer calls asking why.

Mental models & heuristics

When no OEM tolerance is published, default to I-CAR's ±5mm UPCR fallback — but treat OEM-published tolerance (commonly ±3mm factory build spec) as the real number whenever it exists. Never anchor to the tolerance for a part with an OEM tolerance you haven't looked up.
When measuring, read the deviation as the sum of both endpoints' error, not either alone. A point reading +3mm off datum and its mate reading −3mm is 6mm of actual span deviation, not "both within 3mm" — this is the single most common measurement-report misread.
When a structural steel component's strength class is unknown, treat it as non-sectionable until the OEM procedure says otherwise. UHSS and boron-alloyed (hot-stamped, ~1,500 MPa) steel sectioned outside an explicit OEM cut-line loses its designed crush behavior; "we've sectioned this grade before on another model" is not a procedure.
When a bumper cover, windshield, or glass-mounted sensor is removed for any reason, assume a calibration is owed until the OEM procedure rules it out — not the reverse. The industry-wide miss is assuming "we didn't touch the sensor" is the same as "the sensor's reference geometry didn't move."
When a supplement crosses roughly 25-30% of the original estimate total, expect insurer desk-review friction and pre-build the file accordingly — measurement printouts, OEM procedure pages, and dated teardown photos, not just an amended line-item list. Below that threshold, most adjusters approve well-documented supplements without escalation; above it, undocumented supplements get partial denials by default.
Static vs. dynamic calibration is a location decision, not a preference. Static calibration needs a level, controlled-lighting bay with exact target placement to OEM-specified distances; dynamic needs a specific road drive at speed with specific lane markings. A shop without static-calibration space isn't skipping a step by subletting it — it's the only correct call.
Total-loss math runs off actual cash value and repair-cost percentage, not gut feel. When repair-cost-to-ACV crosses the state total-loss threshold (commonly 70-80%, varies by state formula), the file moves to the insurer's total-loss desk regardless of whether the shop believes the car is fixable — don't sink diagnostic time into a repair path the numbers already killed.

Decision framework

Blueprint before estimating anything final. Teardown to bare structure on any repair touching a structural or safety-system area before treating the insurer's initial estimate as more than a placeholder — a supplement built on a photo-only estimate is adversarial by default; one built on a blueprint is just math.
Measure before deciding pull, sculpt, section, or replace. Pull the OEM tolerance if published, the ±5mm I-CAR fallback if not, and classify the damage as elastic (pull candidate), kinked/torn (replacement mandatory), or ambiguous (get a second measurement point and a supervisor sign-off).
Check every affected part against its OEM position statement before writing the repair line, not after the estimate is approved — a sectioning plan built before checking the statement gets re-argued with the insurer twice instead of once.
Flag every ADAS-relevant operation on the initial estimate, even if the insurer's database doesn't prompt for it. List the specific calibration (which sensor, static or dynamic) as its own line with the OEM's triggering procedure cited, not folded into "R&I bumper."
Write the supplement as a documentation packet, not a bigger number. Each new line pairs a photo or measurement reading with the OEM or database source that justifies it; unsupported line items are the ones that get partially denied and reopen the whole negotiation.
Pull and log DTCs before and after the repair, on every vehicle with ADAS or electronic stability control, so a pre-existing fault code can't be mistaken for repair-caused damage in either direction.
Verify against a post-repair test — printed alignment specs, calibration confirmation report, or a documented test drive — before returning the vehicle. A repair that looks complete on the shop floor and hasn't been verified against its OEM-required test is not finished, it's unconfirmed.

Tools & methods

Electronic 3D measuring systems (bench or tower-mounted, e.g., laser or ultrasonic) referenced against OEM datum specs, not a tram gauge alone for anything structural.
I-CAR Repairability Technical Support (RTS) portal and OEM repair-procedure portals (Honda Service Express, GM TIS2Web equivalents, etc.) pulled per-VIN, not per-model-year assumption — trim and build-date variants change position statements.
Estimating platforms (Mitchell, CCC ONE, Audatex/Solera Qapter) for the original and supplemental writes; treated as a starting database, not ground truth on labor time or included operations.
ADAS calibration equipment — static target frames and OEM-specified floor targets for camera/radar, or a subletted calibration specialist where in-house static bay space or equipment isn't available.
Scan tool / OEM diagnostic software for pre- and post-repair DTC pulls, distinct from the aftermarket generic scanner used for basic code reads.

Communication style

To the customer: leads with what was found during teardown versus what the estimate assumed, in plain terms ("the estimate assumed the rail was just dented — it's actually torn, which means replacement, not straightening"), and states the delay and cost delta before it becomes a surprise invoice. To the insurance adjuster: leads with the specific measurement reading or OEM procedure citation, not an argument about fairness — "rail deviation is 7mm against a 3mm OEM tolerance, here's the printout" moves a claim; "this needs more money" doesn't. To other technicians: direct and procedural, citing the position-statement page number or the measurement point that drove the call, since the next person to touch the car needs the same reference, not the conclusion alone.

Common failure modes

Estimating off the photo instead of the teardown, then treating every hidden-damage finding as an argument to have rather than an expected step in the process.
Sectioning a part because it's dimensionally repairable while skipping the strength-class and position-statement check — dimension and structural integrity are different questions.
Treating ADAS calibration as optional when the customer is cost-sensitive — it's not a customer choice once the OEM procedure requires it; presenting it as negotiable creates liability the shop, not the customer, will own.
The overcorrection: after one comeback for a missed calibration, calibrating everything regardless of whether the OEM procedure actually requires it for that repair — billing (and delaying) for calibrations the vehicle doesn't need burns the trust the caution was meant to protect.
Writing a supplement as one lump sum increase instead of itemized, sourced lines — lump-sum supplements read as padding to an adjuster and get partial-denied by default, which then requires the exact itemization that should have shipped the first time.
Letting cycle time drift without diagnosing why — a car sitting past its severity-tier benchmark for "parts" often means a parts order was never actually placed, or a supplement has been sitting unapproved for a week.

Worked example

Situation. 2021 Honda CR-V, rear-ended at moderate speed. Insurer's initial (photo-based) estimate: $4,150 — rear bumper cover replace, rear reinforcement bar replace, trunk lid repair and blend, paint materials on two panels. Vehicle has a rear-view camera and rear radar-based blind-spot/cross-traffic system mounted through the bumper and tailgate area. No calibration line on the initial estimate.

Teardown and measurement. Bumper cover and reinforcement bar removed; blueprint reveals the rear frame rail (a mild/high-strength steel section on this platform, not boron UHSS — confirmed against the Honda position statement before considering sectioning) is deformed at the pinch-weld area. 3D measurement against Honda's published datum: right-side point reads +4mm off datum, left-side mate point reads −3mm — a 7mm span deviation against Honda's published ±3mm tolerance for that point pair. No kinking, cracking, or tearing observed — deformation is within elastic range, so per the first-principles rule this is a pull-and-verify candidate, not mandatory replacement. The blind-spot radar's mounting bracket, welded to the rail near the deformation, is confirmed bent and requires replacement regardless of the pull outcome, since a bent sensor bracket cannot hold OEM aim tolerance.

Estimate reconciliation (supplement build):

|---|---|---|---|

| Bumper cover replace | $620 | — | confirmed, no change |

| Reinforcement bar replace | $340 | — | confirmed, no change |

| Trunk lid repair & blend | $980 | — | confirmed, no change |

| Rear radar + camera calibration (static + dynamic per Honda procedure) | not on original | +$650 | OEM procedure requires calibration after rear structural repair and bracket replacement; omitted entirely from photo-based estimate |

| Subtotal | $4,150 | +$2,180 | |

Reasoning that overturns the naive read. A generalist reading "52% supplement" would expect insurer pushback on the whole number. The actual friction point is narrower: $980 (rail) and $210 (bracket) are supported by dated teardown photos and the measurement printout — those get approved on the first pass in most claims workflows. The $650 calibration line is the one the insurer's estimating database never prompted for, because "R&I rear bumper" doesn't auto-flag a calibration requirement in most platforms — this is exactly the SCRS Database Enhancement Gateway pattern of a systemically missing not-included operation. Submitting it with the Honda procedure page citing calibration as a required post-repair step, rather than as a bare labor line, is what converts it from a negotiable add to a covered OEM requirement.

Supplement submission (as delivered):

> Supplement request — Claim #[xxxx], 2021 Honda CR-V

> Teardown complete. Three items not visible at initial photo estimate:

> 1. Rear frame rail deformation — 7mm measured deviation vs. Honda's published 3mm datum tolerance (measurement printout attached). No kinking/tearing — repair path is pull and reinforce per Honda structural procedure §[ref], not replacement. +$980.

> 2. Radar sensor mounting bracket — bent at weld point, confirmed at teardown (photos attached), fails OEM aim tolerance if reused. +$210.

> 3. Rear radar + camera calibration — required per Honda position statement following rear structural repair and sensor bracket replacement (procedure page attached); static and dynamic calibration both required on this trim. +$650.

> Quarter panel blend added to paint materials line as a direct consequence of item 1 (adjacent panel reopened). +$340.

> Revised total: $6,330. Vehicle will not be released without calibration confirmation report per Honda procedure — this is not a discretionary add.

Going deeper

references/playbook.md — blueprinting/teardown sequence, the frame pull-vs-replace measurement procedure, supplement documentation packet, ADAS calibration decision process, and cycle-time benchmarks by severity.
references/red-flags.md — smell tests on estimates, measurements, and calibration gaps, with the first question and the data to pull.
references/vocabulary.md — terms generalists (and some insurance adjusters) misuse, with the practitioner distinction that matters.

Sources

I-CAR — Uniform Procedures for Collision Repair (UPCR), Repairability Technical Support (RTS) portal, and published position-statement summaries (e.g., "Don't Section Ultra-High-Strength Steel," "Always Follow Vehicle Maker Procedures").
Honda/Acura Body Repair News and OEM position statements on ultra-high-strength and boron steel (front door ring, rear frame rail) — including the 2015 Acura MDX door-ring crash-test demonstration of incorrect sectioning.
Stellantis/Ram OEM repair-procedure documentation on non-sectionable inner reinforcement components.
Society of Collision Repair Specialists (SCRS) — Database Enhancement Gateway (DEG), the industry mechanism for reporting missing or incorrect "not-included" operations in Mitchell/CCC/Audatex estimating databases.
CCC Intelligent Solutions — annual Crash Course industry report data on supplement frequency (>60% of repairs) and typical original-to-final estimate gaps.
asTech and OEM calibration-requirement guidance on static vs. dynamic ADAS calibration triggers (windshield, bumper, structural, suspension, and alignment work).
I-CAR Gold Class shop recognition and OEM-specific certification programs (e.g., Honda ProFirst) as the credentialing context referenced in the mental models section.
No direct auto-body-repairer practitioner has reviewed this file yet — flag corrections or gaps via PR.

View SKILL.md source on GitHub · maturity: draft

Jurisdiction: US (baseline)