Best Tools to Verify Whether a Roof Was Already Replaced

Ask any roofing owner what their single most expensive recurring mistake is, and after a beat they'll tell you it's the same one: spending money and labor on roofs that don't need work. The flashiest version is the storm chaser who knocks a whole neighborhood that was already re-roofed after the last hail event. The quieter, more common version is the estimator who drives forty minutes, climbs a ladder, and writes a beautiful quote for a homeowner who replaced their roof eighteen months ago and is just price-shopping for an insurance dispute. Either way, you paid for a confirmation you could have gotten for free or for pennies, before anyone left the office.

Verifying whether a roof has already been replaced sounds like a single yes/no question. It isn't. It's a layered evidence problem, and the reason it trips people up is that every individual data source is wrong some meaningful fraction of the time. Permits get pulled and never closed. Satellite imagery is a year stale. A homeowner swears the roof is "only a couple years old" when the shingles are clearly cupping. The skill is not finding one magic tool. It's knowing which tools to stack, in what order, and where each one quietly lies to you.

I've spent years building target lists, running canvassing teams, and cleaning up estimating pipelines, and I've watched good reps burn whole weeks chasing already-done roofs. What follows is an honest evaluation of the tools and data sources that actually answer this question, what each is genuinely good at, where it falls down, and how to combine them into a workflow that holds up. I'll name real tools and be fair about all of them, including the one I'm closest to.

What "verifying a roof was replaced" actually requires

Before evaluating tools, get clear on what you're really trying to establish, because "was the roof replaced" hides at least four different questions and they need different evidence.

• Was a full replacement done, or just a repair or overlay? A repair patch, a partial slope, or a shingle-over-shingle overlay is not a new roof, and it changes both the sales conversation and any claim-documentation work. Imagery and permits can both miss this distinction.
• When was it done? "Replaced" is meaningless without a date. A roof replaced before the last qualifying storm is a very different opportunity than one replaced after it. Most of your real-world value comes from pinning the date relative to known hail or wind events.
• Who did it and is it documented? A permitted, manufacturer-warranty-registered replacement leaves a paper trail. A weekend cash job by an unlicensed crew leaves almost none, and those are exactly the roofs that look new from the street but fail early.
• Does it match what the homeowner is telling you? Half of verification is reconciling records against the human in front of you, who is sometimes mistaken and occasionally not being straight with you.

Keep those four straight and the tool choices make sense, because no single source answers all four. Permits answer "who and roughly when" but miss unpermitted work. Imagery answers "did the surface change and roughly when" but can't tell you a warranty exists. Field inspection answers "what's actually up there right now" but can't reconstruct history. You triangulate.

The evaluation criteria I judge these tools on

Every tool below gets measured against the same yardstick, earned from using them on real territories:

1. Truth rate. When the tool says a roof was or wasn't replaced, how often is it right? More precisely, what are its characteristic false positives and false negatives? A tool that's confidently wrong is worse than one that admits uncertainty.
2. Date precision. Can it tell you whether a roof changed and also when it changed, tightly enough to place it before or after a specific storm? A two-year imagery gap is useless for a storm that hit last spring.
3. Coverage. Does it work everywhere you sell, or only in certain counties, certain price tiers, certain imagery refresh zones? Rural and small-municipality coverage is where most tools quietly collapse.
4. Cost and effort per address. Some answers are free but take fifteen minutes of clicking. Some are instant but cost a per-report fee. At scale, the difference between a free-but-slow source and a paid-but-instant one is real money.
5. Scalability. Can you check one address, or ten thousand? A tool that's great for vetting a single appointment may be useless for cleaning a 20,000-record mailing list.
6. Defensibility. If this verification ends up in a file an adjuster or a court might see, is the evidence dated, sourced, and reproducible? A screenshot with a visible capture date beats a salesperson's hunch every time.

With that yardstick in hand, here are the tools and sources, each as its own mini-review.

The tools, evaluated one by one

1. County and municipal building permit records

What it actually is. Nearly every jurisdiction in the U.S. requires a permit for a full roof replacement, and most now publish permit records through an online portal — searchable by address, often showing permit type, issue date, contractor, valuation, and inspection status. Some counties use named platforms (Accela Citizen Access, Tyler Technologies' portals, eTRAKiT, OpenGov) so once you learn one interface you can reuse it across many jurisdictions.

What it's genuinely good at. This is your single strongest piece of documentary evidence that a roof was professionally replaced and roughly when. A closed, finaled re-roof permit dated after a storm is close to a smoking gun. It also names the contractor, which tells you whether it was a real company or a permit pulled by a homeowner. For claim-documentation work, a permit record is dated, official, and reproducible — exactly the kind of artifact that belongs in a file.

Where it falls short. Three big holes. First, unpermitted work is invisible here, and in some markets a startling share of residential re-roofs never get permitted — cash jobs, rural areas with light enforcement, "repairs" that were really replacements. Second, an issued permit is not a completed roof; people pull permits and never do the work, or the work fails final inspection. Always check the inspection/close status rather than mere issuance. Third, coverage and data quality are wildly uneven — big metros have clean searchable portals, while plenty of small towns and unincorporated areas still make you call or visit a counter. Permit lag is also real: a roof finished last month may not appear in the portal for weeks.

Rough pricing posture. Usually free to search directly. Aggregators that normalize permits across jurisdictions (see Shovels and BuildZoom below) charge for the convenience and the coverage.

Best for: confirming and dating a documented, professional replacement, and identifying the contractor — when you're verifying a specific address and the jurisdiction has a usable portal.

2. Permit-data aggregators (Shovels, BuildZoom)

What it actually is. Services that ingest permit records from thousands of jurisdictions, clean and standardize them, and let you search or pull them by address or in bulk. Shovels leans toward contractors and data customers with an API and analytics over national permit data; BuildZoom is more consumer- and contractor-directory-facing but surfaces permit history per address and per contractor.

What it's genuinely good at. Coverage and scale. The whole point is that you don't have to learn forty county portals or know which platform each town uses. For checking many addresses, or for enriching a list with "has a re-roof permit in the last N years," an aggregator with an API is the only sane path. Shovels in particular is built for that kind of programmatic, multi-jurisdiction querying.

Where it falls short. It inherits every weakness of the underlying permit data — unpermitted work is still invisible, and a jurisdiction the aggregator hasn't onboarded or that doesn't publish digitally simply won't be there. There's also ingestion lag on top of the jurisdiction's own lag, so very recent work can be missing. And normalization is imperfect: roof work sometimes hides inside a general "alteration" or "reroof/repair" bucket, so you have to read the description rather than trust a category flag alone.

Rough pricing posture. Paid, typically subscription or API/usage-based, scaling with volume and whether you need bulk and historical access. Reasonable for an operation doing this at scale; overkill if you check a handful of addresses a week.

Best for: teams that need permit history across many jurisdictions at once, or want to enrich a mailing/target list programmatically rather than one address at a time.

3. Current high-resolution satellite and aerial imagery (Google Earth Pro, Nearmap, Bing/Apple Maps)

What it actually is. Top-down imagery of the roof as it exists. Google Earth Pro (free desktop) is the workhorse because it includes a historical imagery timeline — a slider that lets you step back through past captures of the same roof. Nearmap sells higher-resolution, more frequently refreshed aerial imagery in covered metros. Bing Maps and Apple Maps give you additional capture dates that sometimes fill gaps Google has.

What it's genuinely good at. Seeing change over time. The killer move is opening Google Earth Pro's historical slider and watching the roof transition from weathered/streaked to clean and uniform between two capture dates — that brackets the replacement to a window, sometimes a tight one. Current imagery also reveals roof complexity, square footage, and obvious issues without leaving your desk. Nearmap's refresh cadence and resolution, where it covers, can pin dates much tighter than free sources and is clear enough to often distinguish a new shingle field from an old one.

Where it falls short. Date precision is entirely hostage to capture frequency. In a rural county, Google's historical captures might be eighteen months apart, so "replaced sometime in this year-and-a-half window" is the best you'll do — useless for a storm that hit in that gap. From straight overhead, a clean overlay or a same-color reroof can be nearly impossible to distinguish from the original, and a steep, complex, or tree-shaded roof hides detail. Imagery tells you the surface changed; it can't tell you whether it was a full tear-off, an overlay, or just an unusually thorough cleaning. And Nearmap's beautiful imagery only helps inside its coverage footprint.

Rough pricing posture. Google Earth Pro and the consumer map apps are free. Nearmap is a paid subscription, priced for businesses, and worth it mainly if you operate inside its covered areas and need the resolution and refresh.

Best for: desk-checking whether and roughly when a roof's surface changed, especially in metros with frequent captures — your fastest free first filter.

4. Historical imagery comparison (the Google Earth Pro time slider as its own technique)

What it actually is. Not a separate product, but the single most underused free technique in this whole category, so it earns its own entry. Inside Google Earth Pro you click the clock icon and get a timeline of every past capture Google holds for that spot. You step through it and watch the roof.

What it's genuinely good at. Bracketing a replacement date for free, with a visible capture date on every frame — which makes it defensible. You find the last capture where the roof looks old and the first where it looks new, and the replacement happened between them. Cross-reference those two dates against known storm dates (more on storm data below) and you can often answer the question that actually matters: was this roof replaced before or after the event?

Where it falls short. Same capture-frequency ceiling as all imagery — if Google only photographed that area once a year, your bracket is a year wide. Image quality varies frame to frame (season, sun angle, cloud, resolution), and an old capture can look deceptively "new" or vice versa. It's also manual and one-address-at-a-time, so it doesn't scale to list cleaning. And like all overhead views, it struggles with overlays and same-color replacements.

Rough pricing posture. Free. Costs only your time.

Best for: dating a suspected replacement on a single high-value address before you commit a truck roll — the highest-value free move on this list.

5. Real estate / property records (county assessor, MLS, Zillow/Redfin, ATTOM)

What it actually is. Public and quasi-public property data. County assessor records sometimes note roof material or a recent improvement. MLS listings (and the consumer sites that mirror them — Zillow, Redfin) frequently include agent-entered notes like "new roof 2023" and, more usefully, photo sets from when the home last sold. Property-data providers like ATTOM and CoreLogic sell structured property attributes and transaction history in bulk.

What it's genuinely good at. Two things. First, listing photos with a date are gold — a sale three years ago with a photo of a visibly old or visibly new roof anchors your timeline precisely, and you also get the prior condition for free. Second, agent remarks sometimes flat-out tell you "roof replaced 2022." For bulk work, ATTOM-style datasets let you filter or enrich large lists by year built, last sale, and sometimes improvement signals.

Where it falls short. Reliability of the roof-specific fields is poor. Assessor "roof" attributes are often the original construction material and rarely updated for a re-roof. Agent remarks are marketing copy — unverified, sometimes aspirational, occasionally wrong. A home that hasn't sold in fifteen years gives you nothing recent. And consumer-site "facts" are aggregated from mixed sources of varying quality, so treat any roof claim there as a lead to verify, not a verified fact.

Rough pricing posture. Assessor and consumer sites are free to browse. MLS access requires a license or an agent relationship. ATTOM/CoreLogic are paid, enterprise-priced for bulk data.

Best for: anchoring a date and prior condition via old listing photos, and as a bulk enrichment layer — never as a standalone proof, always as corroboration.

6. Storm and hail event data (NOAA SPC, NWS, IBHS, and hail-mapping vendors)

What it actually is. Records of when and where damaging hail and wind actually occurred. NOAA's Storm Prediction Center and the National Weather Service publish storm reports and event data; the Insurance Institute for Business & Home Safety (IBHS) does the research on what hail does to roofs. Commercial vendors (HailTrace, Interactive Hail Maps, and the hail layers inside platforms like HailRecon) sell finer-grained, address-level storm-swath history.

What it's genuinely good at. This is the context layer that turns a date into a decision. Verifying that a roof was replaced is only half the job; the question is whether it was replaced relative to the storm that matters. If your imagery brackets a replacement to early last year and the qualifying hail event was last spring, those storm dates tell you whether the new roof predates or postdates the event. Commercial hail maps tighten this from "the county got hail sometime that month" to "this address sat under a swath on this specific day."

Where it falls short. Storm data tells you a storm occurred, never that a specific roof was damaged — that's a forecast-of-odds, not proof. Public NOAA/NWS reports are coarse and depend on human spotter reports, so small or rural events are under-recorded. Vendor hail maps are modeled estimates with real uncertainty, not ground truth. None of this verifies a replacement by itself; it only contextualizes a date you established with another tool.

Rough pricing posture. NOAA, NWS, and IBHS resources are free and authoritative. Commercial address-level hail history is paid, subscription-based.

Best for: converting a replacement date into a before/after-the-storm answer — essential alongside imagery or permits, useless on its own for verification.

7. Field inspection and the human conversation (the irreplaceable ground truth)

What it actually is. A person actually looking at the roof — from a ladder, a drone, or the ground with binoculars — plus a straight conversation with the homeowner. Drone tools (DJI hardware with apps, or measurement platforms like DroneDeploy) make the visual part safer and more documentable; photo-documentation tools like CompanyCam timestamp and geotag what the crew sees.

What it's genuinely good at. Ground truth that no record can give you. Up close you can read the actual condition and distinguish a true tear-off from an overlay (count the layers at a rake edge or eave; an overlay shows two courses), spot brand-new flashing, fresh pipe boots, crisp shingle granules, clean valleys, and warranty-grade installation versus a sloppy cash job. The homeowner conversation resolves the rest — most people will tell you when and by whom, and their answer either confirms or contradicts your records. CompanyCam-style timestamped photos make whatever you find defensible later.

Where it falls short. It's the most accurate source and the most expensive one — a truck roll costs real money and time, which is the entire reason you do the desk verification first. It also can't reconstruct history beyond what's visible now: a five-year-old roof and a one-year-old roof can look similar from the ground, and a skilled installer's overlay can hide its layers. Homeowners are sometimes wrong about their own roof's age, and a small number aren't candid, especially mid-dispute.

Rough pricing posture. Drones and DJI hardware are a one-time capital cost; DroneDeploy and CompanyCam are paid subscriptions. The real cost is labor and windshield time per address.

Best for: the final, definitive confirmation on an address worth the truck roll — never your first move, always your last.

8. Roofing-platform property intelligence (where ranking/targeting tools fit, including RoofPredict)

What it actually is. A newer category of contractor platforms that blend property data, roof-age estimation, and storm exposure to rank which roofs in an area are likely due — and, as a byproduct, to flag which are likely already done. EagleView and GAF QuickMeasure are adjacent here: they're measurement-report tools, but their imagery and roof reports incidentally help you judge condition and recency. RoofPredict sits in the ranking/targeting slice of this category, combining roof-age bands, storm exposure, and opportunity scoring house-by-house.

What it's genuinely good at. Pre-filtering at scale. The honest framing matters: these tools don't prove a roof was replaced — they estimate, from age bands and storm history, which roofs are likely due versus likely recently done, so you spend verification effort where it pays. RoofPredict's value here is honest about its own limits: its scoring is roof-age plus storm-exposure heuristics, not a magic detector, and roof age is delivered as a range, not an exact date — which is genuinely all the underlying data supports. Used right, it shrinks a 20,000-address territory down to the houses worth checking with permits and imagery, and it can carry tracked direct mail, per-home microsite/PDF reports, and a canvassing app on top of the targeting. EagleView/QuickMeasure shine specifically when you need an accurate measurement and a clean recent image of the roof.

Where it falls short. Be clear-eyed: a heuristic age-band score is not address-level proof of a replacement, and anyone selling it as certainty is overselling. RoofPredict won't tell you a permit was finaled or show you a layer count at the eave — it points you at the right doors; permits, imagery, and a ladder still do the actual verifying. EagleView/QuickMeasure cost per report and are measurement-first, so they're not built to date a replacement on their own. And these platforms inherit the same coverage and freshness limits as the property and imagery data underneath them.

Rough pricing posture. EagleView and QuickMeasure are per-report (QuickMeasure is GAF's lower-cost option; EagleView is the premium incumbent). Platform tools like RoofPredict are subscription/SaaS, priced for an operation rather than a one-off check.

Best for: narrowing a large territory to the addresses worth verifying, and — for EagleView/QuickMeasure — getting an accurate measurement and current roof image as part of that look. Use them to prioritize verification, not to replace it.

The comparison table

Scored on the six criteria from earlier. Ratings are my qualitative judgment from field use, not lab measurements — High / Medium / Low.

Tool / source	Truth rate	Date precision	Coverage	Cost-effort per address	Scalability	Defensibility
County/municipal permits	High (when permitted)	High (if finaled)	Medium (uneven)	Free but slow	Low (manual)	High
Permit aggregators (Shovels, BuildZoom)	High (when permitted)	High	Medium-High	Paid, low effort	High	High
Current imagery (Google Earth, Nearmap, Bing)	Medium	Medium	Medium-High	Free–paid, low	Low–Medium	Medium-High
Historical imagery slider (Google Earth Pro)	Medium-High	Medium (capture-bound)	Medium-High	Free, moderate effort	Low	High
Property records (assessor, MLS, ATTOM)	Low-Medium	Medium (via photos)	High	Free–paid	High (bulk)	Medium
Storm/hail data (NOAA, IBHS, hail vendors)	N/A (context only)	High (event dates)	High	Free–paid	High	High
Field inspection + homeowner	Highest	High (current state)	Universal	Expensive	Low	Highest
Platform intelligence (RoofPredict, EagleView)	Medium (prioritization)	Low–Medium	Medium-High	Paid	High	Medium

Read the table as a portfolio, not a leaderboard. The cheap, scalable sources (permits, imagery, property data, platform scoring) are for filtering thousands of addresses down; the expensive, accurate ones (field inspection) are for confirming the few that survive the filter. No row wins alone.

How to actually combine them: a verification workflow

Here's the sequence I'd run, ordered cheapest-and-fastest to most-expensive, so you spend money only on addresses that survive each gate.

Step 1 — Establish the storm baseline (once per territory). Before checking any address, pin the dates of the qualifying hail/wind events for the area using NOAA SPC, NWS storm reports, and IBHS context (plus a commercial hail map if you have one). Every later "when was it replaced" answer gets compared against these dates. Do this once and reuse it.

Step 2 — Desk-check the imagery (free, ~3-5 min/address). Open the address in Google Earth Pro, run the historical timeline, and look for the old-to-new transition. Cross-check Bing/Apple for extra capture dates. If you have Nearmap or EagleView/QuickMeasure in your stack and the address is high-value, pull a current image for resolution. Outcome: either a bracketed replacement window, or "surface looks unchanged and weathered" (a likely-due roof).

Step 3 — Pull the permit record (free–paid, ~2-5 min/address). Search the jurisdiction portal, or query an aggregator (Shovels/BuildZoom) if you're doing volume. Look specifically for a re-roof permit, its issue date, finaled inspection status, and the contractor name. A finaled re-roof permit dated after your storm baseline is near-conclusive that the roof is done. No permit doesn't prove it wasn't replaced — it might be unpermitted — so it lowers but doesn't eliminate suspicion.

Step 4 — Corroborate with property records (free–paid, ~3 min/address). Check the last MLS/Zillow/Redfin listing for dated photos and any "new roof" remark, and the assessor record. You're looking for a second independent date anchor or a contradiction. An old listing photo showing a clearly aged roof, dated after your imagery "new" frame, would be a red flag worth resolving.

Step 5 — Reconcile and decide. Lay the dates side by side: imagery window, permit date, listing photo date, storm dates. If two or more independent sources agree the roof was replaced after the storm, you have a verified "already done" — drop it from the prospect list (or route it to the right workflow if it's a claim-documentation situation). If sources conflict, or everything points to an old roof, it survives to the field.

Step 6 — Field-verify the survivors only (expensive). For addresses worth a truck roll, the inspector confirms on site: layer count at the eave/rake (overlay vs. tear-off), flashing and boot condition, granule and valley condition, and a straight homeowner conversation about when and by whom. Document with timestamped photos (CompanyCam-style). This is your ground truth and your defensible record.

The whole point of the order is leverage: steps 1-5 are cheap and kill most false starts at the desk, so step 6 — the only truly expensive one — runs against a short, pre-qualified list.

A worked example

Let me make it concrete with an illustrative (not real) walkthrough so the reconciliation logic is clear.

Say your storm baseline for a suburban county shows a significant hail event on May 14 of last year. You're vetting 123 Maple Street, flagged by a targeting score as borderline.

• Imagery: Google Earth Pro shows captures on August of two years ago (roof streaked, dark weathering, visible patch) and June of this year (roof uniform, clean, lighter). Replacement happened in that window.
• Permit: The county portal shows a reroof permit issued April 2 of this year, finaled May 20 of this year, contractor "Acme Roofing LLC."
• Listing: No recent sale; assessor lists original 3-tab from the build year, unhelpful.
• Reconciliation: Two independent sources (imagery window + finaled permit) place the replacement in spring of this year — after the May 14 storm of last year. Verdict: roof already replaced, professionally and recently. Drop from the replacement-prospect list. You spent maybe ten minutes and zero windshield time.

Now flip one fact. Suppose the permit instead read issued March of last year, finaled April of last year — before the May 14 storm. Same "already replaced" verification, completely different opportunity: a roof that's only a year old but sat under a qualifying hail event afterward is a candidate for documenting your own inspection findings, not a candidate for a replacement quote. Same tools, opposite business decision — which is exactly why dating relative to the storm is the whole game.

Where pros get this wrong (the failure modes)

After watching a lot of teams do this, the same mistakes recur:

• Trusting an issued permit as a finished roof. Issuance is intent; the finaled inspection is completion. Always read the status. Permits get pulled and abandoned constantly.
• Trusting absence of a permit as proof no work was done. Unpermitted re-roofs are common. "No permit on file" lowers the odds of a documented replacement; it does not prove the roof is original.
• Letting stale imagery date a replacement too tightly. If the captures are eighteen months apart, your window is eighteen months wide. Don't claim a precision the capture cadence doesn't support, especially if a claim file might lean on it.
• Mistaking an overlay for a new roof from overhead. Top-down imagery can't see layers. A clean overlay reads as "new" in imagery and on assessor records but is not a tear-off — only an edge inspection settles it.
• Taking the homeowner's date as fact. Plenty of people genuinely don't know when their roof was done, or repeat what a previous owner told them. Treat it as one input to reconcile, not the answer.
• Treating storm data as damage proof. A storm in the area is odds, not evidence that this roof was hit. Verification of a replacement and assessment of damage are two separate jobs; don't let one masquerade as the other.
• Skipping verification because a targeting score "said so." A heuristic age-band/opportunity score is a prioritizer, not a verifier. It tells you where to look, not what's true. Confusing the two is how you end up quoting a two-year-old roof anyway.

How to pick the right stack for your situation

The correct toolset is not universal — it changes with what you're actually doing. Here's how I'd choose for the common situations I see.

If you're vetting a single appointment before you drive out. You don't need to buy anything. Run the free chain: Google Earth Pro historical slider, the county permit portal, and a quick Zillow/Redfin look for dated listing photos, all checked against your storm baseline. Fifteen minutes at the desk routinely saves an hour of windshield time and a wasted estimate. Reserve the truck roll for genuine ambiguity. This is the highest-ROI habit a small shop can build, and it costs nothing but discipline.

If you're cleaning a large mailing or canvassing list. Per-address manual work doesn't scale past a few dozen, so this is where paid coverage earns its place. A permit aggregator with an API (Shovels, BuildZoom) lets you flag every record with a recent finaled re-roof permit and suppress those addresses before you print a single mailer. A targeting platform that ranks by roof-age band and storm exposure narrows the rest to the likely-due houses. You're not verifying each address to certainty here — you're removing the obvious already-dones and prioritizing the rest, then verifying individually only the ones a rep is about to engage.

If you operate inside a Nearmap or frequent-capture metro. Lean harder on imagery. High-resolution, frequently refreshed aerials can pin a replacement window tightly enough to decide most cases without a permit pull, and they double as a measurement and condition source. The same workflow in a rural county with yearly captures has to lean on permits instead, because the imagery window is too wide to be decisive.

If the verification feeds claim-documentation work. Defensibility outranks speed. Favor sources that produce dated, official, reproducible artifacts: finaled permit records, imagery frames with visible capture dates, dated listing photos, and timestamped field photos. Avoid resting a conclusion on a homeowner's recollection or a heuristic score, because neither holds up in a file. Build the evidence index from records that carry their own provenance.

If you're a high-volume storm operation moving fast after an event. Speed-to-door matters, but so does not knocking the houses that were already done after the prior storm. The efficient pattern is to pre-suppress recently-permitted and recently-imaged-new addresses at the list stage, then let canvassers confirm condition at the door. The cost of a verification miss here is multiplied across a whole crew's day, so a little upfront filtering pays back fast.

If you're a one-person or new operation with no budget. The free stack does almost everything you need at low volume, and it's frequently more defensible than a paid score because every artifact is dated and official. Add paid tools only when manual effort becomes the bottleneck — that's the signal you've outgrown free, not a fixed list size.

Reading the roof up close: the field tell-tales that settle it

When an address survives the desk and earns a look, knowing what to read turns a vague "looks newish" into a real verdict. These are the signals a seasoned inspector checks, and they're worth training every rep on because they're the difference between a confident call and a guess.

• Layer count at the rake and eave. The cleanest single tell. Sight along an exposed edge and count shingle courses stacked at the starter: one layer is a tear-off-and-replace, two (or more) is an overlay laid over the old roof. An overlay reads as "new" from overhead and on records but is not a fresh deck-up installation, and it ages differently.
• Flashing, pipe boots, and penetrations. New step flashing, fresh counterflashing, and crisp rubber pipe boots are installed with a replacement; weathered, cracked, or mismatched ones that don't match supposedly-new shingles suggest a partial job or an older roof dressed up. Boots crack and chalk with age, so they're a quiet odometer.
• Granule condition and color uniformity. A genuinely new asphalt field has dense, evenly distributed granules and consistent color. Bald patches, exposed asphalt, and heavy granule loss in the gutters point to an older roof, regardless of what a listing remark claims.
• Ridge, valley, and hip detailing. Recent work shows clean ridge caps, properly woven or cut valleys, and tidy hip lines. Sloppy or aged detailing at these stress points is where a cash job or an older roof gives itself away.
• Deck and decking signs from the attic. When you can get inside, fresh underside nailing, new felt or synthetic underlayment shadows, and replaced sheathing are strong confirmation of a real tear-off rather than a surface overlay.
• The conversation, asked well. "When did you do the roof, and who did it?" gets a date and a name you can check against the permit. Follow with whether it was a full tear-off — homeowners often know if there was "the old roof torn off" versus shingles added. Reconcile their answer with your records; agreement confirms, disagreement is a flag to resolve before you act on it.

Document whatever you find with timestamped, geotagged photos so the call is reproducible later. A dated edge photo showing a single layer is worth more in a file than any number of confident assertions.

A note on staying on the right side of the line

If any of this verification feeds into insurance-claim or supplement work, keep your role clean. A contractor may document its own inspection, estimate, scope, and evidence; pull and OCR public records; build a dated evidence index; and track its own paperwork. It may not represent or negotiate the claim for the homeowner, interpret their policy or coverage rights, tell them what they're "entitled" to recover, or use deductible-waiver or "free roof" messaging. Verifying when a roof was replaced and documenting your own findings is squarely contractor-side work. Advising the insured on their claim is not. The homeowner files, the insurer decides, and you document your own scope — keep those lanes separate and the verification work stays clean.

Tricky cases the standard workflow misses

A few situations break the clean sequence above, and knowing them ahead of time saves you from a confident wrong call.

• Partial replacements and single-slope jobs. A homeowner who replaced only the storm-facing slope after a directional wind event has a roof that's half new, half old. Overhead imagery may show a color difference on one plane; permits may describe "partial reroof." Don't let a new front slope read as a whole new roof. Note which planes changed and when, because the un-replaced slopes are a different conversation entirely.
• Insurance-driven replacements. After a major hail event, a neighborhood can re-roof in a single season. These usually are permitted and imaged, so they verify cleanly — but they cluster, which means a targeting score may still flag the whole block as "storm-exposed" even though the roofs are now new. This is exactly why a score is a prioritizer and the permit/imagery check is the verifier; the score can't see that the block already cashed in.
• Solar installations. Panels frequently go on within a year of a new roof because nobody wants to mount solar on an old deck. A roof that sprouted panels between two imagery captures was very likely re-roofed just before — the panels themselves become a dating clue. Permits for the solar job sometimes reference the roof work too.
• Recently sold homes. A sale is a natural re-roof trigger (a condition that came up in inspection, or a buyer credit). Always check whether a replacement clusters around the last transaction date, and pull the listing photos from before the sale to see the prior condition. The transaction date is one of your best free anchors.
• Manufactured-color matches. When a homeowner re-roofs in the same shingle color as the original, overhead imagery can show almost no visible change, and you'll miss it entirely from the desk. If records suggest a replacement that imagery doesn't corroborate, don't conclude "no replacement" — conclude "imagery can't see it" and lean on the permit and a field look instead.

The common thread: the workflow gives you a strong default, but these edge cases are where reconciling multiple sources, instead of trusting one, keeps you out of trouble.

The free / manual alternative (and when it's the right call)

You can do nearly all of this with zero software spend, and for a low-volume operation you probably should. The free stack is: NOAA/NWS/IBHS for storm dates, Google Earth Pro's historical slider for the imagery bracket, the county permit portal directly, Zillow/Redfin/assessor for corroboration, and a ladder and a conversation for the survivors. That covers the great majority of single-address verifications for free, and it's often more defensible than a paid score because every artifact carries a visible date and an official source.

Where paid tools earn their keep is volume and coverage: when you're cleaning a 20,000-record list, learning forty county portals by hand isn't viable, so a permit aggregator with an API and a targeting platform to pre-rank become worth the spend. And when you operate inside Nearmap's footprint or routinely need EagleView/QuickMeasure measurements, that imagery resolution pays for itself in fewer wasted truck rolls. The honest rule of thumb: pay for scale and coverage, do the per-address truth-finding with the free stack, and never let any paid score substitute for the cheap, dated, official records that actually hold up.

Bottom line

There is no single best tool to verify whether a roof was already replaced, and anyone who tells you otherwise is selling you something. The best system is a stack: storm dates for context, historical imagery to bracket the change, permit records to document and date it, property records to corroborate, a targeting platform to decide which addresses even deserve the effort, and a field inspection to settle the survivors. Run them cheapest-first so the expensive truck roll only happens on addresses that earned it. Date everything relative to the storm that matters. Treat scores as prioritizers and records as proof. Do that, and you'll stop quoting roofs that were finished two springs ago — which, for most operations, is the cheapest money you'll ever save.

FAQ

What is the single best tool to verify whether a roof was already replaced?

There isn't one. The reliable approach is a stack: a finaled re-roof permit from the county portal is your strongest documentary proof, Google Earth Pro's historical imagery slider brackets when the surface changed, and a field inspection settles anything ambiguous. Each source is wrong some of the time, so you triangulate rather than trusting any single one.

How do I look up a roof permit by address?

Search the building department portal for the city or county where the property sits — many use platforms like Accela, Tyler, eTRAKiT, or OpenGov. Look for a reroof/replacement permit and check its issue date, inspection/finaled status, and the contractor name. If you need to check many addresses or jurisdictions at once, a permit aggregator like Shovels or BuildZoom pulls standardized records across thousands of jurisdictions.

Can satellite or aerial imagery tell me when a roof was replaced?

It can bracket it. Google Earth Pro's historical timeline lets you step through past captures and find the last frame where the roof looks old and the first where it looks new — the replacement happened between those dates. Precision is limited by how often the area was photographed, so a rural location with yearly captures only gives a year-wide window. Nearmap offers higher resolution and more frequent captures inside its coverage.

Does no permit on file mean the roof was never replaced?

No. Unpermitted re-roofs are common, especially cash jobs and work in areas with light enforcement. Absence of a permit lowers the probability of a documented professional replacement but does not prove the roof is original. Corroborate with imagery, listing photos, or a field look before concluding.

How can I tell a full tear-off from an overlay?

Overhead imagery and assessor records usually can't — both can read an overlay as a new roof. The reliable check is at the roof edge: count the shingle courses at a rake or eave. Two layers indicate an overlay; one indicates a tear-off and new deck-up installation. That distinction often requires a ladder or a close drone shot.

Are property sites like Zillow or county assessor records reliable for roof age?

Treat them as leads, not facts. Assessor roof fields are often the original construction material and rarely updated after a re-roof, and listing remarks are unverified marketing copy. The genuinely useful piece is dated listing photos from a past sale, which anchor both the timeline and the prior roof condition. Always verify a roof claim from these sources against permits or imagery.

Why does the storm date matter when verifying a replacement?

Because the business decision hinges on whether the roof was replaced before or after the qualifying storm. A roof replaced after the event is simply done; a roof replaced before it but exposed to the event afterward is a different situation entirely. Pin storm dates from NOAA's Storm Prediction Center, the NWS, and IBHS research, then compare every replacement date against them.

How does a targeting platform like RoofPredict fit into verification?

It prioritizes, it doesn't verify. Platforms like RoofPredict use roof-age bands and storm-exposure heuristics to rank which roofs in a territory are likely due versus likely recently done, which shrinks a large list to the addresses worth checking. The age it gives is a range, not an exact date, and the score isn't proof — you still confirm with permits, imagery, and a ladder. Use it to decide where to look, not to conclude what's true.

Can I do all of this without paying for software?

For low volume, yes. NOAA/NWS/IBHS for storm dates, Google Earth Pro's historical slider, the county permit portal, Zillow/Redfin/assessor for corroboration, and a ladder plus a homeowner conversation cover most single-address verifications for free — and every artifact carries a visible date and official source, which is often more defensible than a paid score. Paid tools earn their place mainly for volume, coverage, and high-resolution imagery.

Is verifying a roof's replacement date allowed if it relates to an insurance claim?

Documenting your own inspection findings and pulling public records is contractor-side work and is fine. What crosses the line is representing or negotiating the claim for the homeowner, interpreting their policy or coverage rights, telling them what they are entitled to recover, or using deductible-waiver or 'free roof' messaging. Verify and document your own scope; the homeowner files and the insurer decides.