How to Pre-Screen Which Roofs Will Likely Qualify for a Storm Claim Before You Inspect

A storm rolls through, and within 48 hours you've got a map full of pins, a couple of crews itching to canvass, and a hard question: out of these 4,000 homes, which ones are actually worth a ladder? Most of them aren't. Some have roofs new enough to shrug off marble-sized hail. Some sat under the edge of the cell where the wind never got above 40 mph. Some are tile or metal that didn't take functional damage. And some — the ones you actually want — are 16-year-old three-tab asphalt that caught the core of the storm and have bruising the homeowner can't see from the driveway.

Pre-screening is the work of separating that last group from the other three before you spend payroll, fuel, and your reps' energy walking the whole subdivision. Done right, it's the difference between a crew that inspects 12 roofs to find 8 worth filing on, and a crew that inspects 30 to find the same 8. Same result, less than half the labor, and far fewer doors slammed on a rep who showed up at a house with a two-year-old roof.

This is a long, operational walkthrough of how to do that screening. It's written for the owner or sales manager building the route, and for the rep who has to make the call house by house. We'll cover what actually drives claim eligibility, the data you can gather without a ladder, a repeatable scoring workflow, the edge cases that fool people, and the compliance lines you do not cross. One thing up front, said plainly: nobody can know a roof qualifies for a claim before inspecting it. Coverage is the carrier's decision, and damage has to be physically verified. What you can do is rank addresses by the odds they have functional, claimable storm damage — and stop wasting ladders on the ones that don't. That's pre-screening, and it's a skill.

What "qualifies for a storm claim" actually means (so you screen for the right thing)

Before you can pre-screen, you have to be honest about what you're screening for. A roof "qualifying" for a storm claim is not one thing. It's a chain, and every link has to hold:

1. A covered peril hit the property. For most residential policies, that's wind and hail — both are typically covered perils on an HO-3 form. The storm had to actually reach this address with enough energy to do work.
2. There is functional damage caused by that peril. Not cosmetic. Not wear. Functional damage to asphalt shingles from hail generally means the impact fractured the mat or dislodged enough granules that the asphalt is exposed and the shingle's service life is shortened. Wind damage means creased, torn, or missing shingles, or compromised seals.
3. The damage is recent and tied to the date of loss. Adjusters are trained to distinguish fresh impacts (sharp, granule loss with bright underlying asphalt, no oxidation) from old hits and from manufacturing or installation defects.
4. The cost to repair exceeds the deductible, and the policy doesn't exclude it. A home with a 2% deductible on a $400,000 dwelling has an $8,000 hurdle. A few creased shingles won't clear it; a roof that needs full replacement will.

When you pre-screen, you are estimating the probability that links 1, 2, and 3 are present, because those are the ones you can read signals for from the ground and the air. Link 4 is largely the homeowner's policy math, which you'll handle at the kitchen table, not on the route. And the final call on all of it belongs to the carrier's adjuster, full stop.

This framing matters because it kills the two most common screening errors. Error one: chasing the storm's path instead of its impact. A hail swath on a TV weather map tells you where it hailed, not which roofs it wore out. Error two: chasing any roof in the swath regardless of age and material, which fills your inspection list with new roofs and metal roofs that ate the same storm and walked away fine.

The three variables that drive 80% of your screening accuracy

If you strip pre-screening down to its load-bearing parts, three variables do most of the work. Get these three right per address and your hit rate climbs hard.

1. Roof age (as a range, not a date)

Age is the single biggest filter, because it controls how vulnerable the roof was when the storm arrived. A standard architectural asphalt shingle has a service life in the ballpark of 15 to 30 years depending on quality, climate, and ventilation, and three-tab shingles run shorter. The relationship that matters for screening: older asphalt is more brittle, has already shed granules, and fractures from impacts that a new shingle would absorb. A 2-year-old roof can take pea-to-marble hail and show nothing functional. A 20-year-old roof can take the same hail and have a dozen fractures per square.

You will almost never get an exact install date, and you shouldn't pretend to. What you can establish is a range — "this roof is roughly 16 to 22 years old" — and a range is plenty to screen with. The threshold most storm-restoration outfits use as a rough cut is: roofs in the back half of their service life are worth a closer look; roofs in their first 5 to 7 years usually aren't, absent very large hail.

Where age comes from without a ladder:

• Aerial and street imagery over time. If you can see the roof in imagery from, say, eight years ago and it already looked weathered, that's a data point. If it was visibly re-roofed (different color, crisp lines) three years ago, that's a stronger one.
• Permit records. Many jurisdictions log re-roof permits. A pulled permit from 2009 with no permit since suggests a roof pushing 17 years.
• Real-estate and listing history. Old listing photos sometimes show a roof in a known year. "New roof" in a sale listing dates it.
• The shingle itself, read from the street. Curling, cupping, heavy granule loss in the gutters, fading, and exposed fiberglass all read as age even from the sidewalk with binoculars.

Be careful with the easy-but-wrong source: year built is not roof age. Zillow, the county assessor, and Google will happily tell you a house was built in 1998. That tells you nothing about whether it was re-roofed in 2014. Re-roofs are invisible to "year built," and screening on year built alone will route you to homes that already replaced.

2. Storm intensity at that specific roof

The second variable is how hard the storm actually hit this structure — not the neighborhood, the structure. Hail size, wind speed, impact angle, and duration vary across surprisingly short distances. Two houses on the same street can see different outcomes if one sat under the densest part of the core and the other caught the trailing edge.

The inputs that matter:

• Hail size. Functional damage to asphalt typically starts becoming likely around 1 inch (quarter size) and up, and the probability rises steeply with diameter. Sub-1-inch hail can still damage old, brittle roofs but is far less reliable. The National Weather Service and the Storm Prediction Center publish hail reports and the SPC keeps storm reports databases you can pull by date and location.
• Wind speed and gusts. Shingle wind ratings are tested to standards (e.g., the ASTM D7158/D3161 classes), but real-world seal failure, aging, and installation quality mean field damage can show up below the rated speed. Peak gusts in the 58+ mph range (the severe-thunderstorm threshold) are where wind claims start getting common, and EF-scale or measured gusts well above that make wind damage likely.
• Direction and exposure. Wind damage clusters on the windward and leeward field and along rake and ridge edges. Hail damage clusters on the slopes facing the storm's approach and on soft metals (vents, flashing, gutters, AC fins) that bruise visibly.

This is the variable most contractors screen worst, because the convenient data — a hail swath polygon or a wind map — is painted at neighborhood or county resolution. "It hailed here" is true for thousands of homes that don't have a claim. What separates a good screen is getting closer to per-address storm intensity: modeling how the storm's energy actually loaded this roof given its slope, orientation, and the storm's track. We'll come back to this, because it's where the modern tooling earns its keep.

3. Material and roof type

The third variable filters out roofs that took the storm and don't behave like claimable asphalt. From imagery and the street you can usually classify:

• Asphalt shingle (three-tab or architectural) — your bread and butter; behaves predictably with age and hail.
• Tile (concrete or clay) — can crack from large hail but the failure mode and claim pattern is different, and small hail rarely does functional damage.
• Metal (standing seam, screw-down, stone-coated steel) — often only cosmetically dented by hail; many carriers treat cosmetic metal dings as excluded, and policies increasingly carry cosmetic-damage exclusions specifically for metal.
• Wood shake, slate, synthetic — specialty behavior, specialty screening; usually a separate workflow.

For a typical residential storm-restoration crew, the highest-probability target is aging asphalt shingle. That's not a rule, it's a base rate. Screening should surface it.

A worked example: turning a 4,000-home swath into a 280-address route

Let's make this concrete. A supercell drops hail across the north side of a metro on a Tuesday evening. By Wednesday you want a route. Here's the funnel, with rough numbers so you can see the leverage at each stage.

Stage 0 — The raw swath: ~4,000 homes. That's everything inside the reported hail polygon. If you canvass this blind, your reps knock 4,000 doors to find maybe 200 to 300 worth filing on. Brutal economics.

Stage 1 — Cut to the real impact zone, not the polygon. The reported swath is generous. The core — where hail actually reached 1.25 inches or the gusts topped 60 mph — might be 55% of the polygon. Pull SPC storm reports, radar-estimated hail size (MESH) for the date, and any spotter reports. Cut the edges. ~4,000 → ~2,200.

Stage 2 — Filter by material. Drop the obvious tile and metal homes and the new-construction pockets where everything is sub-5-years. Imagery classification handles most of this. ~2,200 → ~1,500 asphalt-and-likely-claimable.

Stage 3 — Filter by roof age range. Rank by estimated age. Keep roofs in roughly the 12+ year band where hail of the observed size is likely to have done functional work; deprioritize the obviously young roofs. ~1,500 → ~600.

Stage 4 — Score per-roof storm load and rank. Now combine age and per-address storm intensity into a single priority score. The 600 aren't equal: a 19-year-old roof on a south-facing slope that took the core is a far better bet than a 13-year-old roof on the swath's edge. Rank them. Your reps work top-down. ~600 ranked, top ~280 form the first-pass route.

That's the leverage. You went from 4,000 cold doors to 280 prioritized ones, and the 280 are ordered so the best bets get knocked first while the storm is fresh and homeowners are receptive. You did it before a single ladder went up. Your inspection-to-file ratio on that route will crush a blind canvass, and — just as important — your reps spend their day at houses where the conversation makes sense, which is how you keep reps from burning out and quitting.

The rest of this guide is how to actually execute each stage.

The math: why screening pays for itself before lunch

Contractors who haven't run the numbers underrate how much a screen is worth, so let's put real arithmetic on it. None of these figures are claims about your specific market — plug in your own — but the structure is what matters.

Say a canvassing rep can effectively work 40 to 60 doors in a productive day. On a blind swath canvass, suppose 1 in 15 of those doors is a roof actually worth inspecting (old enough, hit hard enough, right material). That rep generates roughly 3 to 4 real inspection opportunities a day, and spends the rest of the day getting turned down at houses that were never going to be jobs. Now run the same rep on a screened, ranked route where the hit rate is 1 in 4. The same 50 doors produce 12-plus inspection opportunities. You didn't add a rep, add hours, or buy a single lead. You changed which doors got knocked.

Stack that across a crew and a storm season and the gap compounds:

The morale row is not a soft metric. Roofing has notoriously high front-line turnover, and a green canvasser who spends a week getting rejected at houses with two-year-old roofs is a canvasser who quits before they ever learn to close. Replacing and re-training that person costs you weeks. A screen that routes them to winnable doors is, quietly, one of the cheapest retention investments you have.

There's a second-order cost to skipping the screen, too: reputation burn. Every time a rep knocks a house with a clearly new roof and launches a storm-damage pitch, that homeowner files you mentally under "roofers who blanket every storm." Do that across a subdivision and you've trained a whole neighborhood to distrust the knock — including the houses that did have damage. Screening protects the credibility of the knock itself.

The pre-screening workflow, step by step

Here's the repeatable process. Build it once as a checklist and your whole team runs it the same way.

Step 1: Lock the date of loss and the storm's footprint

Everything keys off the date of loss (DOL). Before you screen a single roof, pin down:

• The exact date(s) the storm hit the area. Multiple events in a season complicate attribution; know which one you're working.
• The storm's reported footprint and intensity. Sources: NOAA/NWS local office summaries, the SPC storm reports database (searchable by date), radar-derived hail products, and IBHS research on hail and wind damage thresholds. Save these. You'll want the documentation of the event later, and so will the homeowner.
• The peril type. Hail-dominant, wind-dominant, or both? It changes what your reps look for and what damage signals you weight.

Note on honesty: the storm data establishes that a peril occurred in the area. It does not establish that any specific roof was damaged. Keep those two ideas separate in your own head and in every conversation with a homeowner. The forecast or the hail report is context, never proof of damage on a given roof.

A note on reading storm data correctly

The storm sources you'll lean on each have a personality, and knowing it keeps you from over-trusting any one of them:

• SPC storm reports are ground-truth spotter and public reports — someone actually measured or eyeballed a hailstone. They're trustworthy where they exist, but they're sparse: a 1.75" report at one intersection doesn't mean the whole grid got 1.75", and the absence of a report doesn't mean it didn't hail. Treat them as confirmed anchor points, not a coverage map.
• Radar-estimated hail (MESH and similar) gives you continuous spatial coverage, which is exactly what spotter reports lack, but it's an estimate derived from reflectivity aloft. It can over- or under-call surface hail size, and it doesn't know about melting, wind drift, or how the hail actually fell on a given slope. Use it to interpolate between the confirmed SPC points.
• NWS local storm reports and event summaries give you the narrative — storm motion, peak gusts, tornado tracks — which is what tells you direction, and direction tells your reps which slopes to inspect first.
• Wind data is the weakest link to read from afar because gusts are so local and shingle seal failure depends heavily on age and install quality. Measured gusts at the nearest ASOS/AWOS station are a reference point, not a per-roof reading.

The honest synthesis: no single public source tells you what happened on one specific roof. They triangulate a probability surface over the area. That gap — from "area-wide probability surface" to "per-roof load" — is the whole reason per-address storm modeling exists, and it's the difference between a sharp screen and a swath dump.

Step 2: Pull the address universe inside the footprint

Get the list of structures inside the impact zone. Parcel data, your CRM's geographic tools, or a mapping platform will give you addresses. At this stage you want quantity; you'll filter hard next.

Step 3: Classify material and drop the non-targets

Run the universe against imagery to tag roof material and obvious new construction. Drop or deprioritize:

• Metal and tile (separate, lower-base-rate workflow).
• Visibly new roofs (crisp, uniform, recently re-roofed).
• Commercial and multifamily if you're a residential crew (or vice versa).

Step 4: Estimate roof age range per address

For each remaining address, build an age range from the best sources you have: imagery-over-time, permit records, listing history, and visible weathering. You're not after a date. You're after a band — "roughly 8–12," "roughly 15–20," "roughly 22+" — good enough to bucket.

A simple bucketing scheme that works:

• 0–7 years: deprioritize unless hail was very large (1.75"+).
• 8–14 years: moderate priority; screen in if storm intensity was solid.
• 15+ years: high priority; these are the roofs storms wear out, and even moderate hail can do functional damage.

Step 5: Estimate per-address storm intensity

This is the step that separates a pro screen from a swath dump. For each address, estimate how hard the storm actually loaded that roof: hail size at that point, peak gusts, and the orientation of the slopes relative to the storm's approach. The closer you get to per-roof resolution — modeling the storm's energy on the actual structure rather than reading a neighborhood-wide polygon — the better your ranking.

Step 6: Combine into a single priority score and rank

Fold age and storm intensity (and a material weight) into one score per address. It doesn't have to be fancy. Even a simple weighted model — say, 45% age, 45% storm load, 10% material/edge factors — sorts the route far better than gut feel. Rank descending. That ordered list is your route.

Step 7: Hand reps a per-address brief, not a bare pin

The rep at the door closes better when they know why this house is on the list: the roof's age range, what the storm did here, and which slopes to look at first. A pin on a map is a guess; a one-line brief is a reason. This is also where a branded, homeowner-facing summary helps a green canvasser sound like a veteran without overclaiming.

Step 8: Inspect, document, and let the homeowner own the claim

Screening ends at the ladder. The actual inspection — photos of fresh impacts, a chalk test, soft-metal evidence, the bruise pattern — is what determines real damage. You document conditions and provide an honest estimate. The homeowner decides whether to file. The carrier's adjuster decides coverage. You never cross into filing, handling, or negotiating the claim, and you never make promises about what the carrier will do.

Reading roofs from the ground: a rep's field checklist

Much of the final pre-screen happens in the 30 seconds your rep stands in the driveway. Train them to read these signals before they ever ask to get on the roof. None of these prove a claim; they raise or lower the odds that an inspection is worth doing.

Age and condition signals (from the street, binoculars help):

• Granules in the gutters and at downspout splash blocks (shedding = age + possible impact loss).
• Curling, cupping, or clawing shingle edges (advanced age, brittleness).
• Bald spots or visible fiberglass mat (heavy wear, high vulnerability).
• Fading and color loss, especially uneven across slopes.
• Patches or mismatched shingles (prior repairs, partial age).

Storm-impact signals (collateral, visible without a ladder):

• Soft-metal bruising. Look at gutters, downspouts, fascia, garage doors, mailboxes, and AC condenser fins. Hail that dents an aluminum gutter or pocks an AC fin almost certainly hit the roof too. This is the single best ground-level proxy for hail impact.
• Window screens, deck boards, and fence tops for impact marks.
• Wind tells: lifted or missing shingles on ridge and rake edges, debris in the yard, displaced ridge caps, fence sections down (directional wind evidence).
• Spatter marks on painted surfaces, decks, and the dirty film of gutters where hail cleaned off oxidation — a fresh, directional spatter pattern dates and directionally locates the hail.

Disqualifiers (lower the odds, save the ladder):

• A roof that's obviously 2–4 years old with no collateral damage.
• Metal or tile with only cosmetic dings (likely excluded).
• No collateral damage anywhere on soft metals despite a claimed big-hail event — possible the core missed this address.

A practical rep rule: if the soft metals are clean and the roof is young, don't burn a ladder. If the gutters are dimpled, the AC fins are flattened, and the roof reads 15+, that's a high-confidence inspect.

What confirms the screen on the roof (so reps know what they're verifying)

Pre-screening hands the rep a high-probability address. The ladder is where probability becomes fact. Reps screen better from the driveway when they understand what they're trying to confirm up top, so it's worth spelling out what a real, claimable hail or wind finding looks like — and what fakes people out.

What functional hail damage actually is on asphalt: a fresh impact that fractured the shingle mat, usually showing as a soft, bruised spot you can feel as a give underfoot, with granules knocked away and the underlying asphalt freshly exposed (dark, no oxidation, no dirt). The classic confirmation steps:

• Chalk a test square (commonly a 10' x 10' square) on a representative slope and count the functional hits inside it. Carriers and adjusters think in hits-per-square; so should your rep. A handful of hits in a test square scales very differently from one or two.
• Check every slope and the soft accessories — ridge caps, hip shingles, and the soft metals up top (vents, pipe boots, flashing, valley metal). Hail damage is directional, so the storm-facing slopes carry the most evidence.
• Confirm freshness. A bruise with bright asphalt and crisp granule loss reads recent; a hit that's oxidized, dirty, or ringed with algae is old and won't tie to your date of loss.

What functional wind damage looks like: creased shingles (a hard fold line where the shingle lifted and slapped back down, breaking the seal), torn or missing shingles, lifted tabs with failed adhesive, and damage concentrated at rakes, ridges, and the field edges where uplift is strongest.

The false positives a good rep won't write up as storm damage:

• Blistering (small raised bumps from trapped gas/moisture in manufacturing) gets mistaken for hail bruising; it's typically round, uniform, and not associated with directional collateral damage.
• Mechanical/foot traffic damage — scuffs and granule loss from someone walking the roof, including prior inspectors.
• Manufacturing granule loss and normal weathering — diffuse, non-directional, with no fresh asphalt.
• Algae streaking and lichen that looks like spatter from the ground but isn't.

Why put this in a screening guide? Because the rep who knows the difference screens the driveway better — they're not fooled into a ladder by blistering visible through binoculars, and they're not talking themselves out of one when the real tells are present. And it keeps everyone honest: you document what's actually there, in a form an adjuster recognizes, and you never inflate weathering into a storm loss. That honesty is both the ethical line and the durable business strategy — manufactured or exaggerated damage is fraud, and it's the fastest way to lose a market.

Safety belongs in this handoff too. Roof inspections are fall-exposure work; OSHA fall-protection requirements apply, and "we screened it so reps move fast" should never translate into skipping tie-offs or sending someone up a wet, steep, or storm-loosened roof. A good screen actually helps here: fewer, higher-value ladders means more time to set up each one safely instead of racing across a whole subdivision.

Where RoofPredict fits: ranking the route before the ladder

Everything above is doable by hand. The problem is it doesn't scale. Pulling age ranges, classifying material, and — the hard part — estimating per-address storm intensity across thousands of homes is days of analyst work you don't have in the 72-hour window when a fresh storm matters most. This is the specific gap RoofPredict is built to close.

RoofPredict takes aerial imagery plus weather data and produces, per address, a roof-age range and a per-roof storm model — and that second part is the differentiator worth understanding. Most storm tools hand you a hail swath: a polygon that says "it hailed here." A swath can't tell two houses on the same street apart. RoofPredict models the storm on each roof, not only where it passed — it scores how the hail and wind actually loaded that structure, then pairs that with the roof's estimated age so you can rank which roofs the storm most likely wore out. The plain version of the pitch: a hail map shows you where it hailed; this shows you which roofs it actually worked over.

In the funnel from earlier, that collapses Stages 2 through 6 into one ranked list. Instead of an analyst spending two days cutting a polygon down by hand, you get the storm-hit area scored house by house — material classified, age estimated as a range, per-roof storm load modeled — and your reps work the ranked list top-down. The roofs most likely to have claimable, storm-caused damage rise to the top; the new roofs and the edge-of-swath roofs sink. You knock and inspect the right doors first, while the storm is fresh.

What it does not do, and where I want to be straight with you:

• It does not tell you a roof qualifies for a claim. It ranks probability, from age and storm physics. The damage still has to be verified on a ladder, and coverage is always the carrier's call.
• Roof age is a range, not an install date. Treat it as a band for prioritizing, not a fact to quote a homeowner.
• The storm model is odds, not proof. A high score means "this roof probably took a beating, go look" — never "this roof is damaged." Don't let a rep wave a score at a homeowner as evidence of damage; that's both inaccurate and a compliance problem.
• It doesn't replace the inspection, the documentation, or your reps' judgment at the door. It decides which doors, not whether there's a claim.

Used that way — as a route-ranking engine that gets your crew to the right addresses before anyone climbs — it turns the pre-screening workflow in this guide from a manual grind into something you can actually run at storm speed. It's the targeting layer, not the inspection and not the claim.

The compliance lines you do not cross

Storm restoration sits close to a legal trip-wire, and pre-screening is exactly where contractors get sloppy because they're excited about a fresh storm. Read this section as carefully as the workflow.

You document conditions; you do not adjust claims. In most states, negotiating, handling, or advising on the settlement of an insurance claim for a fee is the practice of public adjusting, and doing it without a license is illegal under unauthorized-public-adjusting (UPPA) statutes. Roofing contractors have been penalized for it. A 2024 Texas case (Stonewater Roofing) underscored that even labeling yourself a claims or insurance "specialist" can cross the line. The safe lane is narrow and clear: you inspect, you document the roof's condition with photos and measurements, you provide an honest repair/replacement estimate, and you hand that documentation to the homeowner. The homeowner files. The homeowner's carrier decides coverage. The homeowner owns the claim.

Things to never say — on a door, in a text, in your marketing, or in a screening brief:

• "You qualify for a claim" / "this roof qualifies" — you can't know that; the carrier decides.
• "We'll get your claim approved" / "we handle the claim" / "we'll negotiate with your adjuster."
• Anything about waiving, covering, eating, or rebating the homeowner's deductible — that's insurance fraud in many states, period.
• "Free roof" — it's not free; it's an insured loss with a deductible the homeowner pays.
• Using a forecast, a hail map, or a probability score as proof the roof is damaged. It's context and prioritization, not evidence of a specific loss.

Document everything legitimately. Date-stamped photos, the storm event documentation (NWS/SPC reports for the DOL), the inspection findings, and a clear estimate. Good documentation supports the homeowner's claim honestly; it never fabricates or exaggerates damage. Adjusters and carriers can spot manufactured damage, and creating it is fraud that ends careers.

If you're building claims-adjacent language into your sales process, get it reviewed by counsel for your state. The lines move state to state, and "everybody in my market says it" is not a defense.

Edge cases and the mistakes pros still make

The basic workflow gets you most of the way. The edge cases are where experience shows.

The new roof that still has a claim

A 3-year-old roof normally screens out — but if the hail was large enough (think 1.75"+ or bigger), even a new roof can take functional damage, and a manufacturer-defect interaction can muddy it further. Don't hard-exclude young roofs in a big-hail event; just weight them lower and let the collateral-damage signals override. A young roof with flattened AC fins and dimpled gutters after golf-ball hail is still an inspect.

The old roof that won't make the deductible

A 22-year-old roof with light, scattered hail might have a few functional hits but not enough to clear a high-percentage deductible on an expensive home. Age screens it in for inspection, but the kitchen-table math may screen it out for filing. Know the difference; don't oversell the homeowner on a claim that won't pencil.

Prior claims and roof history

A roof that already had a storm claim paid in the last few years complicates a new one — the carrier may argue the prior payment should have covered replacement, or that current damage is unrepaired prior damage. It doesn't kill the opportunity, but it changes the conversation and the documentation burden. Worth flagging in your screen if you can see prior permits.

Cosmetic vs. functional, and the policy fine print

More policies now carry cosmetic damage exclusions, especially for metal roofs and sometimes for shingles in hail-prone states. A dented-but-functional metal roof can be a dead end under such an exclusion. Your reps should know to ask whether the homeowner's policy has a cosmetic exclusion or a separate, higher wind/hail deductible — it changes whether an inspection leads anywhere.

Attribution across multiple storms

In an active season, a market can take three hail events in four months. Tying the damage to a specific date of loss matters because the policy in force, the deductible, and the filing window all key off the DOL. If your screen lumps all events together, you'll misdate losses and create headaches. Keep events separate.

The "whole swath" temptation

The biggest mistake, still, is treating the hail polygon as the target list. It's the starting universe, not the route. Contractors who canvass the whole swath burn their reps out on new roofs and edge-of-storm homes, tank their inspect-to-file ratio, and train homeowners to think roofers just blanket every storm. Screening down to ranked, high-probability addresses is the entire point.

Confusing measurement tools with screening tools

EagleView, HOVER, Roofr and similar are excellent at measuring a roof you've already decided to bid — squares, pitch, facets. They are not age-and-storm screening tools; they answer "how big is this roof," not "which roof should I be on." Different category. Use measurement tools after screening has put you on the right roof, not as the screen itself. Likewise, "year built" from Zillow or the assessor is not roof age — re-roofs are invisible to it.

Building the screen into your operation

A few notes on making this stick beyond a single storm.

Standardize the score. If every manager weights age and storm intensity differently by feel, your routes are inconsistent and you can't improve them. Write down the weighting, run it the same way every storm, and adjust based on results.

Track the feedback loop. Log, per route, your inspect-to-file and file-to-approval ratios (approval being the carrier's, not yours). If a screening tweak raises inspect-to-file, keep it. This is the only way to know your screen is actually working rather than just feeling productive.

Protect your reps with good routes. A rep who knocks 30 doors to find one worth inspecting quits. A rep who knocks the right doors closes, makes money, and stays. Pre-screening is a retention tool as much as a sales tool — the route quality directly drives whether your green canvassers survive their first season.

Keep the documentation tight from day one. The same storm data you used to screen (NWS/SPC event records for the DOL) becomes part of the honest documentation package for homeowners who do file. Capture it once, reuse it.

Stay inside the lines, every time. The fastest way to lose a market is a UPPA complaint or a fraud allegation. Screen aggressively, inspect honestly, document carefully, and let the homeowner and carrier own the claim. The contractors who last are the ones who never made it about "getting claims approved" — they made it about being on the right roofs with honest work.

Pre-screening between storms (and why age alone keeps a crew busy)

Storm screening is reactive — a cell hits, you have 72 hours, you run the funnel. But the same machinery works in the slow weeks when nothing has hit, and the smart shops run it then too, because a crew that only eats when it hails starves between storms and overstaffs after them.

The off-season version drops the storm-intensity variable and leans on age and condition alone. The question changes from "which roofs did this storm wear out" to "which roofs in my service area are simply aging out and due for replacement regardless of weather." A 22-year-old three-tab roof that's curling and shedding granules is a real opportunity whether or not a storm comes — it's a retail replacement or a maintenance conversation, and it keeps your crews working when the radar is quiet.

Running that screen on your own territory does two things. First, it builds a standing list of aging roofs you can work as retail replacements, so your pipeline doesn't live and die by the next supercell. Second, it pre-loads your storm response: when a cell does hit, you already know which homes in that footprint were old and vulnerable going in, so your post-storm screen starts from a sharper baseline. The roofs that were 18 years old yesterday and took the core today are the very top of your route.

This is also where mining your own customer book pays off. The estimates you wrote three years ago and lost, the past customers whose roofs were borderline then — those are aging on a known clock. A homeowner you quoted at year 17 is at year 20 now. Screening your existing CRM by age surfaces money already sitting in your records, with no ad spend and a warm prior relationship. Storm or no storm, owning your own streets and your own book is more durable than waiting on the next event or renting the same homeowner from a lead site that already sold them to four competitors.

The tooling that ranks storm-hit roofs by age + per-roof storm load is the same tooling that ranks a quiet neighborhood by age alone — which is why a screen you stand up for storm season ends up being a year-round targeting engine, not a seasonal one.

Putting it together

Pre-screening which roofs are likely to qualify for a storm claim isn't fortune-telling and it isn't a hail map. It's a disciplined funnel: start with the real impact zone, not the generous swath; filter out the materials and the new roofs that don't behave like claimable asphalt; estimate each roof's age as a range; model how hard the storm actually loaded that structure; combine it into a single ranked list; and send your crew up the ladders that the data says are worth climbing — in priority order, while the storm is fresh. The damage gets verified on the roof, the homeowner owns the claim, and the carrier decides coverage. Your job is to make sure the ladder goes up on the right house.

That ranking — roof age as a range, storm physics modeled per roof, the right doors surfaced to the top — is exactly the targeting layer RoofPredict is built to give a storm-restoration crew, so you can run this whole workflow at storm speed instead of by hand. Hand it a storm-hit area and it ranks the roofs the storm most likely wore out, so your reps inspect the right addresses first. It won't tell you a roof qualifies — nothing honest can — but it will tell you which roofs are worth your ladder, and on a 4,000-home swath in a 72-hour window, that's the difference between a profitable storm and a burned-out crew. If you want to see it run on a storm you already know, book a demo and hand it a roof you've already inspected — then judge whether it nailed it.

FAQ

Can you really tell which roofs will qualify for a storm claim before inspecting them?

No — and any tool or rep that says they can is overselling. Coverage is the insurance carrier's decision, and damage has to be physically verified on the roof. What you can do is estimate the probability that a roof has functional, storm-caused damage by combining its age, the storm's actual intensity at that address, and its material. That probability lets you rank which roofs are worth a ladder, but it never replaces the inspection or the carrier's call.

What roof age makes a storm claim most likely?

There's no hard cutoff, but as a screening rule of thumb, asphalt roofs in the back half of their service life — roughly 15 years and older — are the most vulnerable, because aged shingles are brittle and fracture from impacts that a new roof would absorb. Roofs in their first 5 to 7 years usually only take functional hail damage when the hail is large (around 1.75 inches or bigger). Always treat roof age as a range, not an exact install date — re-roofs are invisible to 'year built' records.

How big does hail have to be to damage an asphalt roof?

Functional damage to asphalt shingles generally starts becoming likely around 1 inch (quarter size), and the probability rises steeply as diameter increases. Smaller hail can still damage old, brittle roofs but is far less reliable. Hail size varies across short distances, so a neighborhood hail report isn't the same as the hail size at a specific address — which is why per-roof storm intensity matters more than a swath polygon.

Why isn't the hail swath map on the weather report enough to build a route?

A swath polygon tells you where it hailed, not which roofs it wore out. It's painted at neighborhood or county resolution, so it can't distinguish two houses on the same street where one took the core and one caught the edge, or a new roof from a 20-year-old one. The swath is your starting universe, not your route. Good screening cuts it down by real impact intensity, material, and age, then ranks per address.

What's the best signal a rep can read from the driveway without getting on the roof?

Soft-metal bruising. Check the gutters, downspouts, fascia, garage door, mailbox, and the AC condenser fins. Hail that dents an aluminum gutter or pocks an AC fin almost certainly hit the roof too, so it's the strongest ground-level proxy for hail impact. Pair it with age signals — granules in the gutters, curling or balding shingles — and you have a high-confidence read on whether a ladder is worth it.

Is 'year built' the same as roof age for screening?

No, and screening on year built is a common, costly mistake. Year built from Zillow or the county assessor tells you when the house went up, not when it was last re-roofed. A 1998 home re-roofed in 2014 has an 11-year-old roof, not a 27-year-old one. Use imagery over time, permit records, listing history, and visible weathering to estimate an age range instead.

How does RoofPredict help pre-screen roofs for storm claims?

RoofPredict takes aerial imagery plus weather data and produces, per address, a roof-age range and a per-roof storm model — it scores how the hail and wind actually loaded each individual structure, not only where the storm passed, then pairs that with estimated roof age. That ranks a storm-hit area by which roofs the storm most likely wore out, so your crew inspects the right doors first. It ranks probability and which addresses to work; it does not determine that a roof qualifies for a claim — that's still verified on the ladder and decided by the carrier.

Can I tell a homeowner they qualify for a claim based on a storm report or risk score?

No. A storm report or a risk score is context and prioritization, never proof of damage on a specific roof, and telling a homeowner they 'qualify' crosses into territory you can't back up and that can create compliance problems. The honest framing: a peril occurred in the area, the roof's age and the storm suggest it's worth inspecting, and the inspection plus the carrier's adjuster determine everything else. You document conditions; the homeowner owns the claim.

Can a roofing contractor handle or negotiate the insurance claim for the homeowner?

Generally no. Negotiating, handling, or advising on the settlement of an insurance claim for compensation is the practice of public adjusting, which requires a license in most states; doing it without one violates unauthorized-public-adjusting (UPPA) statutes, and contractors have been penalized for it. Your safe lane is to inspect, document the roof's condition honestly, and provide an estimate. The homeowner files and owns the claim; the carrier decides coverage. Never touch the deductible, and get claims-adjacent language reviewed by counsel for your state.

What about metal and tile roofs in a hail storm?

They behave differently from asphalt and belong in a separate, lower-base-rate workflow. Metal roofs are often only cosmetically dented by hail, and many policies now carry cosmetic-damage exclusions that make those dents a dead end. Tile can crack from large hail but the failure mode and claim pattern differ. For a typical residential storm-restoration crew, aging asphalt shingle is the highest-probability target, so screening should surface it and deprioritize metal and tile.