How Much Money Roofers Leave on Insurance Claims (And Where It Actually Leaks)

Ask ten restoration roofers how much money they leave on insurance claims and you'll get ten shrugs. Most owners can tell you their close rate and their average job size to the dollar, but they cannot tell you how much approved scope they never billed, how much recoverable depreciation a homeowner never released, or how many supplements died in an inbox because nobody followed up after the third email. The leak is invisible precisely because it never shows up as a loss. It shows up as a job that closed at $14,200 when it should have closed at $18,600, and everyone moves on to the next door.

The honest answer, from years of watching restoration files move from kitchen table to depreciation check, is that a typical contractor running insurance work leaves somewhere between 10% and 30% of legitimately earned restoration revenue on the table. Not fraud. Not padding. Not anything a careful adjuster would object to. Just earned, documentable, recoverable money that falls through cracks in the process: scope that was on the roof but not on the estimate, depreciation that was approved but never collected, supplements that were valid but never resubmitted, and code items that applied but were never cited.

This is a breakdown of exactly where that money goes, with real dollar math on each leak, and the documentation discipline that closes it. It stays strictly on the contractor's side of the line: documenting your own inspection, your own scope, your own field evidence, and routing everything insurer-facing through human review. The homeowner files the claim and the insurer decides what it owes. Your job is to make sure that decision is based on a complete, accurate, page-cited record of what the roof actually needs.

The shape of the leak: why earned revenue disappears

Restoration revenue leaks for a structural reason. A roofing claim is not one transaction. It's a sequence of a dozen handoffs, each one a place where information can be lost: the inspection, the photo set, the contractor's estimate, the carrier's estimate, the scope reconciliation, the first supplement, the approval, the build, the final invoice, the certificate of completion, the depreciation release, and the final payment. Every handoff is a leak point. Money doesn't vanish all at once. It bleeds a few hundred dollars at a time across the whole chain.

Three forces make it worse. First, the people who do the inspecting are usually not the people who do the billing, so field knowledge never reaches the invoice. The sales rep saw the bent drip edge but the office never knew to bill it. Second, restoration roofers are volume businesses, and volume punishes follow-up. When you're signing four jobs a week, a supplement that needs a fourth email after 21 days simply does not get the fourth email. Third, the documentation that would justify the missing money was never captured in a usable form. The rep has eleven blurry photos on a personal phone and no idea which one shows the damaged pipe boot.

None of these are character flaws. They're process gaps. And because they're process gaps, they're fixable with process, not heroics. Let's walk the chain leak by leak and put a number on each one.

Leak #1: Scope that was on the roof but never on the estimate

The single biggest leak is the simplest: line items that were physically present and damaged on the roof but never made it onto the contractor's estimate or the carrier's estimate. These aren't exotic. They're the accessories and details that everyone knows a roof has but nobody itemizes under pressure.

The usual suspects, in rough order of how often they go missing:

• Drip edge and rake edge metal. Frequently omitted, frequently required by code. On a 30-square roof you might be looking at 280-340 linear feet of eave and rake.
• Pipe jacks / vent boots. A house has three to six of them. They crack with age and hail. They're cheap individually and easy to forget collectively.
• Ridge vent vs. box vents. If the existing ventilation was ridge vent and the estimate prices three turtle vents, you've under-scoped both labor and material.
• Step flashing and counter flashing at walls and chimneys, often hidden under siding and assumed rather than measured.
• Ice and water shield in valleys and at eaves where code or manufacturer specs require it.
• Starter strip as a separate line rather than buried in the field shingle count.
• Detach and reset of solar, satellite, gutters, or HVAC that sits on or near the roof plane.
• Steep and high charges when pitch exceeds 7/12 or the eave height exceeds one story, both of which carry legitimate labor multipliers.

Here is what the math looks like on a single average residential reroof. Numbers are illustrative ranges, not quotes, and vary by region, pricing database version, and carrier:

Add those up and a single job routinely carries $1,500 to $3,500 of legitimately present scope that's at risk of going unbilled. On a roof that approves at $16,000, that's roughly 10-20% of the job sitting in items most people consider too small to chase. They're not too small. They're the difference between a 38% gross margin and a 30% one.

Why scope goes missing, and the fix

Scope goes missing because the inspection and the estimate are disconnected. The fix is a closed-loop field-to-estimate discipline: every inspection produces a structured record of what's on the roof, and the estimate is reconciled against that record before it goes out, not after.

The practical version is a standardized roof inventory checklist that the rep completes on every inspection, photo-backed, item by item: count the penetrations, measure the eave and rake, note the ventilation type, flag the pitch, photograph the flashing. Then the office reconciles the carrier estimate line-for-line against that inventory. Anything on the inventory that's not on the estimate becomes a documented, evidence-linked scope item to raise for review. This is the contractor documenting the contractor's own scope. It is squarely on the allowed side of the line.

Here's the inventory in checklist form, the version a rep can run in ten minutes on the roof:

• Field area and pitch. Total squares, predominant pitch, any sections over 7/12, eave height (single vs. multi-story).
• Penetrations. Count every plumbing vent, count every B-vent / furnace flue, count every electrical mast, photograph each boot or collar.
• Ventilation. Type (ridge vent, box/turtle, power, gable), linear feet of ridge vent if present, count of static vents.
• Edge metal. Linear feet of eave, linear feet of rake, existing drip edge present or absent, gutter apron present or absent.
• Flashing. Step flashing at every wall, counter flashing at masonry, chimney crickets, headwall and sidewall conditions, skylight flashing kits.
• Valleys. Linear feet, open vs. closed, ice-and-water present.
• Accessories to detach and reset. Solar, satellite dishes, antennas, gutters, lightning protection, HVAC line sets crossing the plane.
• Decking condition notes. Visible sag, prior layers, soft spots, anything likely to surface on tear-off.

A worked example shows why the checklist pays. Take a real-feeling job: a 1,900-square-foot ranch, 28 squares, 6/12 pitch, with four plumbing vents, one furnace flue, 110 linear feet of ridge vent, and step flashing at a 14-foot wall. The carrier's first estimate prices the field shingles, underlayment, tear-off, and a flat ridge cap, and stops there. Reconciled against the inventory, the office surfaces: drip edge at eave and rake (about 300 LF, not on the estimate), four pipe boots (not on the estimate), ridge vent replacement rather than ridge cap over a sealed ridge (a material and labor difference), step flashing at the wall (assumed, not itemized), and ice-and-water at the eaves where the jurisdiction requires it. That's five documented items, each backed by a labeled photo and a count from the inventory, worth a combined $2,100 of present-but-unbilled scope on a job that approved at $14,800. None of it is invented. All of it was on the roof. The only thing that changed is that someone wrote it down at the source and reconciled it before the file closed.

The handoff that makes the checklist stick

A checklist only works if the field data reaches the office in a usable form. The failure mode is a rep who fills out a paper checklist that lives in a truck and a photo set on a personal phone that nobody else can see. The discipline that closes the gap is simple: the inventory and the photos attach to the job file the moment the inspection ends, before the rep leaves the driveway, and the office reconciles within 24 hours while the roof is still fresh in everyone's memory. The longer the gap between inspection and reconciliation, the more scope evaporates, because nobody can later remember whether that boot was cracked or just old.

Leak #2: Recoverable depreciation that's approved but never collected

This is the quietest, ugliest leak, and on a per-dollar basis it may be the largest. To understand it you have to understand how a replacement-cost claim pays out.

Most restoration claims are written on a replacement cost value (RCV) basis with recoverable depreciation. The carrier calculates the full replacement cost of the roof, then subtracts depreciation (for age and wear) to arrive at the actual cash value (ACV). The carrier typically pays the ACV first, minus the deductible. The depreciation is held back and becomes payable, recoverable, once the work is actually completed and documented to the carrier's satisfaction. That depreciation release is real money the homeowner is owed and that funds the contractor's invoice.

Here's the leak. Recoverable depreciation can be a substantial fraction of the total claim, sometimes 20-35% of RCV on an older roof. And it is only released when the contractor submits proper completion documentation: a final invoice matching the approved scope, a certificate of completion, and often dated completion photos. If that paperwork is sloppy, late, or never sent, the depreciation can go unreleased. Many policies cap the window for claiming recoverable depreciation, often 180 days to two years from the date of loss depending on the policy and state. Miss the window and the money is simply gone.

Let's put numbers on it. Take a claim with an RCV of $20,000, a $2,000 deductible, and 25% depreciation:

If the completion paperwork never goes in correctly, that $5,000 stays with the carrier. The contractor builds a $20,000 roof and collects $13,000 plus whatever the homeowner pays out of pocket. That's a 25% haircut on a job that was fully approved. It's the worst kind of leak because the money was already approved. Nothing needed to be argued. It just needed paperwork.

How depreciation gets stranded

Depreciation gets stranded for boring, mechanical reasons:

1. No completion packet exists. Nobody assembled the final invoice, certificate of completion, and completion photos into a single submittable package.
2. The final invoice doesn't match the approved scope. If the invoice line items don't reconcile to the carrier's approved estimate, the release stalls in review.
3. The homeowner is the bottleneck. On many policies the depreciation check is issued to the insured, and the contractor has to coordinate with the homeowner to endorse and forward it. If that relationship went cold, the check sits.
4. Nobody tracked the deadline. Without a date-of-loss clock and a release deadline on every open file, expirations sneak up.

The fix is a depreciation release tracker: every RCV claim gets a record of the date of loss, the recoverable depreciation amount, the policy's claim window, and a status from "work complete" to "docs submitted" to "check received." No file closes until the depreciation is collected or formally written off with a reason. Treat unreleased depreciation as accounts receivable, because that's exactly what it is.

The completion packet that releases the money

The single artifact that releases recoverable depreciation is the completion packet, and most stranded depreciation traces to a packet that was never assembled or never matched the approved scope. A clean packet has, at minimum:

1. A final invoice that reconciles line-for-line to the carrier's approved scope. If the carrier approved 32 line items totaling $20,000 and your invoice shows a lump sum or a different line structure, the reviewer can't tie it out and the release stalls. Mirror the approved estimate's structure.
2. A signed certificate of completion dated to the actual completion date.
3. Dated completion photos showing the finished roof, the new accessories installed, and ideally a few wide shots that establish the whole roof was done. Tie the photos to the same labeling system you used at inspection.
4. Any change documentation for scope that was added during the build (supplements, hidden conditions), so the invoice total is fully explained.

When the packet is complete and self-explaining, the release is a formality. When it's missing a piece, it becomes a back-and-forth that a busy coordinator may never finish.

A worked depreciation example, including the homeowner bottleneck

Return to the $20,000 RCV claim with $5,000 of recoverable depreciation. The work is done in week six. Production marks the job complete and moves on. Here's the timeline where the money strands:

• Week 6: roof complete. No completion packet assembled, because production's job ends at "complete" and the office never got a trigger.
• Week 9: the office notices the receivable is short by $5,000 and assembles a packet, but the final invoice is a lump sum that doesn't match the carrier's 30-line estimate.
• Week 11: the carrier asks for an itemized invoice. The coordinator, mid-storm-season, sets it aside.
• Week 20: the depreciation check is issued to the homeowner (per the policy), who has since moved on and doesn't realize it needs to be endorsed to the contractor.
• Month 14: nobody followed up, the claim window closed, and $5,000 of fully approved revenue is gone.

Every step in that timeline is a tracker failure, not a coverage problem. With a release tracker, week 6 generates a packet task the day production closes the job, the invoice mirrors the approved estimate from the start, and the homeowner-check handoff is a tracked step with its own owner and clock. The same $5,000 lands in the bank by week 8.

Leak #3: Dead supplements

A supplement is a request to add or revise approved scope after the initial estimate, supported by documentation of conditions that the original estimate missed: a code requirement, a hidden condition exposed during tear-off, a measurement correction, a pricing database update. Supplements are normal, legitimate, and expected. Carriers process them every day.

Supplements die for one reason above all others: nobody follows up. A supplement gets submitted, the carrier requests additional documentation or simply doesn't respond, and the supplement coordinator is buried under thirty other files. The fourth follow-up never happens. The supplement quietly dies, and with it goes $800, $1,500, $3,000 of valid scope.

The common dead-supplement patterns:

• The unsupported supplement. Submitted with a number but no photo, no measurement, no code citation. Easy for the carrier to set aside, and easy for the contractor to forget.
• The orphaned supplement. Submitted, acknowledged, pending more info, and then dropped because the person who submitted it changed roles or got overwhelmed.
• The code supplement nobody cited. A local code requires a second layer of ice and water shield or full deck re-nailing, and it genuinely applies, but the estimate never cited the code section, so it reads like an upsell instead of a requirement.
• The pricing-update supplement. Xactimate price lists update monthly. A claim approved on an old list may legitimately re-price higher on a current list, but only if someone catches it and documents it.

The number here is real. Restoration shops that don't run disciplined supplement tracking commonly leave 5-15% of their total restoration revenue in dead and unsubmitted supplements. On a shop doing $4M in restoration volume, that's $200,000 to $600,000 a year evaporating in follow-ups that never happened.

A supplement workflow that doesn't leak

The difference between shops that collect supplements and shops that don't is almost never knowledge. It's process. A supplement that survives has four things and a clock:

1. A trigger. A documented reason it exists: code section, hidden condition photo, measurement correction, or current price list.
2. Evidence attached. A page-cited document or a dated, labeled photo that shows the condition. "Photo 14: cracked pipe boot, north slope" beats forty unlabeled images.
3. A reconciliation. A line-for-line comparison of the contractor's documented scope against the carrier's estimate, so the gap is obvious and specific.
4. A human-reviewed submission. Drafted internally, reviewed by a qualified person before anything goes to the carrier, and worded as a request for factual review, not a demand.
5. A clock and an owner. Every open supplement has a next-action date and a named person. If the carrier hasn't responded in the agreed window, the system surfaces it for the next touch.

Notice the framing on point 4. You are submitting factual documentation of conditions you observed and asking for review. You are not telling the carrier what it owes, not interpreting the policy, not negotiating a settlement. The contractor documents; the insurer decides.

The language that keeps supplements factual and on the right side of the line

How a supplement is worded matters as much as what's in it, both for approval odds and for staying inside the contractor's lane. The same condition can be written two ways:

The left column interprets coverage, asserts entitlement, and threatens. Those are the moves that belong to a licensed public adjuster or attorney, not a contractor. The right column does only three things: states a documented fact, attaches the evidence, and requests review. That's the entire contractor playbook, and it's also, not coincidentally, what actually gets supplements approved. Adjusters move faster on a labeled photo and a code section than on an argument.

A supplement aging board

The operational heart of supplement recovery is an aging board, the same way a collections desk ages receivables. Every open supplement sits in a status with a clock:

• Drafted (evidence attached, awaiting internal review)
• In review (qualified reviewer checking framing and completeness)
• Submitted (sent to carrier, clock started)
• Awaiting carrier (next-touch date set; if no response by the date, it surfaces)
• Info requested (carrier asked for something; owner and due date assigned)
• Approved (moves to invoice and build)
• Closed-won / closed-lost (with a documented reason on every loss)

The board's job is to make sure nothing sits in "awaiting carrier" or "info requested" past its clock without a human touching it. A shop that works this board weekly stops losing supplements to silence, which is where the vast majority of them die.

Leak #4: Code items that applied but were never cited

Building codes drive a large share of legitimate restoration scope, and code-driven scope is some of the most defensible scope there is, because it's not a judgment call. The code either requires it or it doesn't. The leak happens when a code requirement genuinely applies but the estimate never cites the specific code section, so the item looks optional and gets stripped.

The big code-driven items:

• Deck re-nailing / re-sheathing where the jurisdiction has adopted requirements for fastening upgrades on reroofs.
• Ice and water shield at eaves and valleys per the International Residential Code in regions subject to ice damming, plus stricter local amendments.
• Drip edge required at eaves and rakes under the IRC.
• Number of layers limits that force a full tear-off rather than an overlay.
• Ventilation minimums tied to attic square footage.
• Underlayment type and coverage per manufacturer and code.

These aren't gray areas. They're written down. But "it's required" is not the same as "it's documented as required." The contractor's job is to cite the adopted code edition and section, attach the jurisdiction's adoption reference where relevant, and let the requirement speak for itself. When you cite IRC R905.2.8.5 for drip edge or the relevant ice-barrier section, you've turned a contestable line item into a documented requirement. That's factual support for your own scope. It's not interpreting anyone's coverage.

A worked example: on a 28-square reroof in a jurisdiction requiring deck re-nailing and ice-and-water at eaves, the un-cited version of the estimate might omit $1,800 of re-nailing labor and $500 of ice-and-water. Cite the adopted sections with the inspection photos and that $2,300 becomes a documented code requirement rather than an argument.

The discipline that makes code recovery reliable is a jurisdiction code sheet: a one-page reference per municipality or county you work in, listing the adopted code edition and the specific local amendments that drive roofing scope. Edition matters because jurisdictions adopt different code years and add their own amendments on top. Re-nailing requirements, ice-barrier requirements, and layer limits all vary by where the roof sits. Build the sheet once per jurisdiction, keep it current, and every estimate in that area cites the right section automatically instead of relying on whoever's writing it to remember the local rule.

A caution worth stating plainly: cite codes, don't interpret coverage. "The adopted code requires X, here's the section" is factual support for your own scope and is firmly in the contractor's lane. "The policy must therefore pay for X" is a coverage conclusion and is not. State the requirement, attach the adoption reference, and route the coverage question to the carrier's decision. You're documenting what the roof needs to meet code, not deciding what the policy covers.

Leak #5: Measurement and pricing errors that quietly round down

This leak is small per item and large in aggregate, because it touches the base of the estimate. Two sub-leaks live here.

Measurement errors. If your squares are low, everything downstream is low: shingles, underlayment, labor, waste, accessories. A roof measured at 27 squares that's actually 29.5 squares under-scopes the entire estimate by roughly 8% across every line. Aerial measurement reports have made this better, but only if the report matches the actual roof planes and someone reconciles facets, hips, valleys, and waste factor against the real structure. Waste factor in particular gets shorted: complex roofs with many hips and valleys legitimately carry higher waste, and a flat default waste percentage leaves material and labor unbilled.

Pricing database lag. Xactimate and similar databases update their regional price lists regularly. A claim written on an old price list and built months later may legitimately re-price on a current list. If nobody checks the price-list date against the build date, the contractor eats the difference. This isn't padding; it's pricing the work at the rate in effect, documented by the database version.

The fix is a pre-submission QA pass: reconcile the measurement report against the roof inventory (facet count, total area, waste factor, ridge/hip/valley lengths), and check the pricing database version on every estimate against the current list. A ten-minute QA step on a $16,000 job routinely recovers $500 to $1,200.

Waste factor deserves its own note because it's the most quietly under-scoped number on a complex roof. A simple gable roof with two planes legitimately runs a low waste percentage. A roof with eight facets, multiple hips, several valleys, and a few dormers cuts far more shingle to fit, and the correct waste percentage is meaningfully higher. Defaulting every roof to the same flat waste number shorts the complex ones on both material and the labor to install it. The reconciliation step should set waste from the actual facet and valley count documented in the inventory, not from a habit.

The pricing-list check is even faster and just as valuable. Note the price-list version on the estimate and the date the work is built. If the database updated between approval and completion and the regional list moved, re-pricing on the current list is pricing the work at the rate in effect, documented by the version number. That's not padding; it's billing at the published rate. The only thing required to capture it is someone comparing two dates.

Leak #6: The handoff between sales and production

The last leak is organizational. The rep who inspected the roof and the production manager who builds it and the coordinator who bills it are three different people, and what one knows the others don't. The rep saw the bent gutter and the cracked skylight flashing. If that knowledge lives only in the rep's head, it never becomes scope.

This leak compounds the others. Missed scope (Leak #1) is largely a handoff failure. Stranded depreciation (Leak #2) is a handoff failure between production ("job's done") and the office ("docs are submitted"). Dead supplements (Leak #3) are a handoff failure between the field ("found rotten decking on tear-off") and the coordinator ("submit it now, with the photo, today").

The structural fix is a single source of truth per job that every role writes to and reads from: the inspection inventory, the photos, the carrier estimate, the reconciliation, the supplements, the completion docs, and the depreciation tracker, all attached to one file that moves with the job. When the rep finds rotten decking mid-tear-off, the photo goes onto the file with a label and a trigger, and the coordinator sees a supplement waiting to be drafted. Nothing depends on someone remembering to mention it.

The tear-off moment is the sharpest example of this leak because it's the one time a hidden condition becomes visible and then disappears again within hours. Rotten decking, double or triple layers nobody knew about, a chimney cricket that was never there, charred framing from an old fire, knob-and-tube near the penetrations: all of it is exposed for a window of a few hours during tear-off and then covered forever. If the crew doesn't photograph it, labeled and dated, in that window, the supplement that would have recovered it has no evidence and dies. A standing rule that the crew lead photographs any surprise condition before it's covered, and that the photo lands on the job file immediately, turns the most perishable evidence in the whole process into a tracked supplement.

Common objections, and why they don't survive the math

A few beliefs keep shops from fixing the leak. Each one sounds reasonable and each one costs money.

"Chasing small items annoys adjusters and slows down approvals." It doesn't, when the items are documented. Adjusters process accessory scope and code items every day; what slows them down is unsupported assertions, not labeled photos with counts and code sections. A complete, factual file is faster to approve, not slower, because there's nothing to go back and forth about.

"The depreciation is the homeowner's problem to collect." On most replacement-cost jobs the depreciation funds your invoice. If it isn't released, you either eat the gap or chase the homeowner for money they don't have. Treating the release as your receivable, and driving the completion packet, is how you actually get paid for work you already did.

"We don't have time for all this documentation." The documentation is faster than the rework. A ten-minute inventory at inspection and a ten-minute reconciliation at the office prevent the hours spent reopening closed files, re-arguing stripped items, and chasing expired depreciation. The shops that feel they have no time for documentation are usually the ones spending the most time on its absence.

"This sounds like public adjusting and we don't want the liability." It's the opposite. Documenting your own scope, your own evidence, and your own completion, and routing everything insurer-facing through human review with factual framing, is precisely the discipline that keeps you out of public-adjusting territory. The liability comes from interpreting coverage and negotiating settlements, which a disciplined documentation system is specifically built to avoid. The contractor documents; the insurer decides; and the do-not-say list is the guardrail that keeps it that way.

Photos: the evidence that makes every leak recoverable

Underneath all six leaks is one shared dependency: evidence. A missed scope item, a stranded depreciation release, a dead supplement, and an un-cited code requirement are all recoverable only if you can put a clear, dated, labeled photo or document in front of a reviewer. The single highest-leverage habit a restoration shop can build is a disciplined photo system, and it's also the cheapest, because every rep already has a camera in their pocket.

The difference between a photo set that recovers money and one that doesn't is organization, not megapixels. A usable set has:

• Coverage. Wide establishing shots of each slope, then mediums of each feature, then tight shots of each defect. A reviewer should be able to walk the whole roof from the photos without guessing.
• Labels. Each photo named to a feature and location: "north slope, pipe boot, cracked," not "IMG_4471." The label is what lets a coordinator attach the right photo to the right line item weeks later.
• Dates. Capture dates intact, so completion photos prove the work was done within the window and inspection photos establish pre-existing condition.
• A consistent sequence. The same order every time (overview, then penetrations, then edges, then flashing, then valleys, then defects), so anyone can find any shot.

The failure mode is universal and expensive: forty unlabeled images on a rep's personal phone, half of them blurry, none tied to a line item. When the supplement coordinator needs to show the cracked boot three weeks later, they can't find it, and the item dies for lack of evidence that was actually captured but never usable. An evidence index, photos and documents labeled and tied to line items, turns a pile of images into a recoverable record. It's the connective tissue that makes the other five fixes work.

The annual math: sizing the prize for your shop

It's worth doing the arithmetic at the company level, because the per-job numbers feel small and the annual numbers do not. Take three illustrative shops:

These are not new sales. They require no new doors knocked, no new ad spend, no new crews. The work was already sold, already approved, already built. The leak is the gap between what was earned and what was collected, and closing even half of it changes the financial picture of the business. A mid-size shop recovering half of a 20% leak adds $400,000 to the top line at close to 100% margin, because the cost of the work is already in the books. There is no cheaper revenue in roofing than the revenue you already earned and never billed.

This is also why the leak deserves an owner. Sales has an owner. Production has an owner. The claim revenue cycle, the path from approved scope to collected dollar, usually has no single owner, which is exactly why it leaks. Assigning one person accountable for reconciliation, supplements, and depreciation releases across every file is often the highest-ROI org change a restoration shop can make.

Putting a real number on your own leak

General percentages are useful for orientation, but you should measure your own leak. Here's a way to do it with files you already have. Pull your last 25 closed restoration jobs and, for each one, fill in this audit:

Total the gaps across 25 jobs, divide by the total approved value of those jobs, and you have your real leakage percentage. Most shops that run this honestly for the first time land between 10% and 25%, and the single biggest line is almost always either stranded depreciation or dead supplements. That number is your annual opportunity: multiply it by your restoration revenue and you've sized the prize.

A documentation system that plugs the leaks

Everything above points to one conclusion: the leaks are documentation leaks, and they close with documentation discipline, not with more aggressive negotiation. The shops that collect their full earned revenue aren't tougher on adjusters. They're more complete in their files. They walk into every interaction with a page-cited, photo-backed, code-cited, reconciled record of what the roof needs, and they track every open item with a clock and an owner.

The operating system that does this has six parts:

1. A standardized inspection inventory. Every roof inspected against the same checklist, photo-backed, so scope is captured at the source.
2. A reconciliation step. Contractor's documented scope compared line-for-line against the carrier's estimate, with every gap surfaced as a specific, evidence-linked item.
3. An evidence index. Every photo and document labeled and page-cited, so any line item can be backed by a named, dated piece of evidence in seconds.
4. A supplement tracker with clocks. Every open supplement has a trigger, evidence, an owner, and a next-action date that the system surfaces.
5. A depreciation release tracker. Every RCV claim tracked from work-complete to docs-submitted to check-received, with the policy window on a clock.
6. A human-review gate. Everything insurer-facing drafted internally and reviewed by a qualified person before it goes out, worded as factual documentation and requests for review, never as demands or coverage interpretations.

Where RoofPredict fits

RoofPredict's claim revenue-cycle side (RoofClaimRCM) is built to be that documentation system on the contractor's side of the line. It turns every document on a claim into verified, page-cited structured data, so the carrier estimate, the contractor estimate, and the inspection inventory can be reconciled automatically and every discrepancy becomes an evidence-linked item rather than something a coordinator has to catch by hand. It tracks each gap, each supplement, and each depreciation release from kitchen table to depreciation check, with the clock and the owner attached, so nothing dies in an inbox. And it keeps a human-approval gate in front of everything insurer-facing: the system drafts and organizes, a qualified person reviews and approves, and the framing stays factual.

Honest limits. RoofPredict does not represent the homeowner, negotiate with the carrier, interpret policy or coverage, or tell anyone what they're entitled to recover. Those things are not the contractor's role and the platform is built to keep you on the right side of that line. Coverage disputes, denials, appraisal, and proof-of-loss questions route to a licensed public adjuster or attorney. What the platform does is make the contractor's own documentation complete, fast, and traceable, so the carrier's decision is based on a full and accurate record of the roof. It documents; it does not decide.

The targeting side of RoofPredict is a separate capability worth a sentence here only because it changes the economics of the whole operation: it estimates which roofs in a territory are due for replacement, as an age range per address from aerial imagery combined with storm exposure modeled per roof. That's odds, not proof, and a range, not a date. It feeds the top of the funnel; RoofClaimRCM plugs the leaks at the bottom.

A 30-day plan to stop the bleeding

You don't need to rebuild your operation to start collecting your earned revenue. Sequence it.

Week 1: Measure. Run the 25-file audit above. Get your real leakage percentage and identify your biggest single leak. Most shops find it's depreciation or supplements.

Week 2: Stop new leaks. Stand up a depreciation release tracker for every open RCV claim today, with date of loss, depreciation amount, policy window, and status. No open claim without a tracked release. Simultaneously, put a clock and an owner on every open supplement.

Week 3: Standardize the inventory. Build one inspection checklist and one accessory list. Require both on every new inspection, photo-backed. Start reconciling carrier estimates against the inventory line-for-line before anything goes out.

Week 4: Add the review gate and the evidence index. Label and page-cite the evidence on every file. Route everything insurer-facing through one qualified reviewer who confirms the framing is factual: documented observations and requests for review, never demands or coverage interpretations.

By the end of the month you'll have a measured leak, a stopped bleed on depreciation and supplements, a standardized capture at the source, and a review gate keeping you on the right side of the line. Then you optimize.

The discipline that earns the money

The money roofers leave on insurance claims is not exotic and it's not contested. It's drip edge that was on the roof. It's depreciation that was already approved. It's a supplement that was valid and just needed a fourth email. It's a code section that applied and was never cited. Every dollar of it is recoverable with documentation discipline that keeps the contractor squarely in the contractor's lane: documenting your own scope, your own evidence, your own completion, and routing everything insurer-facing through human review.

The contractors who collect their full earned revenue aren't gaming anything. They're just complete. They capture scope at the source, reconcile it against the carrier estimate, back every item with labeled evidence and code citations, track every open supplement and depreciation release with a clock, and let a qualified human approve anything that touches the carrier. The homeowner files the claim. The insurer decides what it owes. The contractor's only job is to make that decision rest on a full, accurate, page-cited record of what the roof actually needs. Do that consistently and the 10-30% you've been leaving on the table starts showing up where it belongs: on your invoices and in your margin.

FAQ

How much money do roofers actually leave on insurance claims?

A typical restoration contractor leaves roughly 10-30% of legitimately earned revenue on the table. The largest leaks are stranded recoverable depreciation (often 20-35% of an older roof's RCV when completion paperwork is never submitted) and dead supplements (commonly 5-15% of total restoration revenue when follow-up fails). The exact figure is best measured by auditing your own last 25 closed files against approved scope.

What is recoverable depreciation and why does it get lost?

On a replacement-cost claim, the carrier pays actual cash value first and holds back depreciation, which becomes payable once the work is completed and documented. That recoverable depreciation can be a substantial share of the claim. It gets lost when the completion packet (final invoice matching approved scope, certificate of completion, dated photos) is never properly submitted, when the homeowner-held check is never coordinated, or when the policy's claim window expires.

Why do roofing supplements die so often?

Almost always because nobody follows up. A supplement gets submitted, the carrier requests more information or doesn't respond, and the coordinator is buried under other files, so the fourth follow-up never happens. Supplements also die when they're submitted without supporting evidence (a labeled photo, a measurement, a cited code section) that makes the documented condition clear and reviewable.

What scope items are most commonly missed on roofing estimates?

Drip edge and rake metal, pipe jacks and vent boots, ridge vent versus box vents, step and counter flashing, ice and water shield, starter strip billed separately, detach-and-reset of accessories, and steep/high labor charges. Individually they look small; together they routinely add up to $1,500-$3,500 of present-but-unbilled scope on an average residential reroof.

How do code citations help recover legitimate scope?

Code-driven items like drip edge, ice-and-water barrier, deck re-nailing, and layer limits are some of the most defensible scope because the code either requires them or it doesn't. The leak is failing to cite the specific adopted code edition and section, which makes a requirement look optional. Citing the section with inspection photos documents the requirement as factual support for your own scope, without interpreting anyone's coverage.

Is recovering this money the same as public adjusting?

No, when done correctly. The contractor documents the contractor's own inspection, scope, evidence, and completion, and submits factual documentation with requests for review. Public adjusting means representing the insured, negotiating settlement, and interpreting coverage, which a contractor should not do without a license. Coverage disputes, denials, appraisal, and proof-of-loss questions route to a licensed public adjuster or attorney. The contractor documents; the insurer decides.

How do I measure my own claim revenue leakage?

Pull your last 25 closed restoration jobs and audit each for: recoverable depreciation collected in full, final invoice matching approved scope, all accessories billed, supplements resolved or abandoned, code items cited, and price-list date matching build date. Total the gaps, divide by the total approved value of those jobs, and you have your real leakage percentage and your annual opportunity.

What documentation is needed to release recoverable depreciation?

Typically a final invoice whose line items reconcile to the carrier's approved scope, a certificate of completion, and often dated completion photos, submitted within the policy's claim window. The release stalls when the invoice doesn't match approved scope, when the packet is incomplete, or when the homeowner-held check isn't coordinated. Tracking each release from work-complete to check-received prevents expiration.

Does aerial roof measurement eliminate the measurement leak?

It helps but doesn't eliminate it. Aerial reports still need to be reconciled against the actual roof, facet count, total area, ridge/hip/valley lengths, and especially waste factor, since complex roofs legitimately carry higher waste than a flat default percentage. Low squares under-scope every downstream line item proportionally, so a quick reconciliation step is worth several hundred dollars per job.

How does RoofPredict help without crossing the public-adjusting line?

RoofClaimRCM turns the contractor's documents into page-cited structured data, reconciles contractor scope against the carrier estimate, surfaces every gap as an evidence-linked item, and tracks supplements and depreciation releases with a clock and an owner, all behind a human-approval gate for anything insurer-facing. It does not represent the homeowner, negotiate with carriers, or interpret coverage; those route to a licensed professional. It makes the contractor's own documentation complete and traceable so the insurer's decision rests on a full record.