Commercial Flat Roof Age Data for Targeting: A Field Guide to Finding Buildings That Are Actually Due

Most commercial roofing salespeople are working a list that lies to them. They pull a list of buildings in a county, sort by square footage or property value, and start calling. Six weeks later they have a stack of "called, no answer" notes and two appointments, one of which is on a building that got a new membrane eighteen months ago. The problem was never the calling. The problem was the list. Nobody on it was sorted by the one thing that actually predicts a reroof: how old the roof is.

Flat roof age is the closest thing the commercial trade has to a buying signal you can see from the air. A 4-ply built-up roof installed in 1998 is on borrowed time. A single-ply TPO that went down three years ago is not a prospect, no matter how big the building is or how motivated the owner sounds in your head. The whole game is separating the first building from the second before you spend a tank of gas, a sales rep's afternoon, or a printed mailer on it.

This is a working field guide to commercial flat roof age data: where it actually comes from, how accurate it really is (and where it quietly fails), how to read an age signal without fooling yourself, and how to turn a column of numbers into a route your crew can run on Tuesday. It is written for the person who has to hit a number this quarter, not for a data scientist. There is real math here, and there are honest limits, because a list that pretends to be more certain than it is will burn you worse than no list at all.

Why roof age beats every other commercial targeting signal

Walk into any commercial roofing office and look at how they pick targets. The common methods, roughly in order of popularity:

• Square footage. Bigger roof, bigger ticket. True, but a 40,000 sq ft roof that's five years old is worth nothing to you this year.
• Property value or assessed value. A proxy for "can they pay." Useful for qualifying, useless for timing.
• Year built. The single most common mistake in the trade. Year built tells you when the structure went up, not when the roof was last replaced. A 1985 distribution center may be on its third roof.
• Owner type / SIC code. Helpful for the pitch (a national QSR chain buys differently than a mom-and-pop strip mall), but it doesn't tell you which location is due.
• "We drove past it and it looked rough." The most accurate method on this list, and the least scalable. You can't drive every building in three counties.

Roof age is the only signal that directly maps to the buying event. A commercial low-slope membrane has a service life you can put a range on:

These are general service-life ranges drawn from manufacturer guidance and trade references; treat them as planning numbers, not warranties. The point isn't the exact figure. The point is that age plus system type gives you a probability the roof is in or near its replacement window. When you sort a list by estimated roof age and filter to buildings whose roof is within a few years of end-of-life, you've done in an afternoon what a driving crew couldn't do in a month.

Age also enriches every other signal instead of replacing it. Big roof and old roof is a six-figure target. Old roof and in a recent hail swath is a documentation opportunity (more on the legally careful version of that below). Old roof and an owner already in your CRM from a repair three years ago is a phone call you should make today.

How each membrane ages, and why that changes your targeting

Service-life ranges are a starting point, but a sharp commercial team targets differently depending on what's up there. The way a roof fails tells you when the owner feels the pain, and pain is what makes a phone get answered. Here's the practitioner's version of how the common systems age and what it means for timing your approach.

Built-up roof (BUR) and modified bitumen

Asphalt-based roofs fail slowly and then all at once. For years they leak a little, get patched, leak somewhere else, get patched again. The owner has a relationship with a leak — a bucket in the warehouse, a stained ceiling tile in the front office — and a maintenance habit that's quietly bleeding money. When you flag a 22-year-old gravel BUR with a dozen visible patches, you're not selling a roof to someone who doesn't know they need one. You're offering to end an annoyance they've lived with for three winters. The talking point writes itself: chronic leaks, rising patch spend, and a roof past its service life.

Mod-bit (the torch-down or self-adhered asphalt sheets that replaced a lot of BUR) fails through granule loss, surface cracking, and seam splitting. From the air it grays out and you can sometimes see the splits. It hits end-of-life a bit sooner than gravel BUR, so a 16-to-18-year-old mod-bit roof is squarely a prospect.

EPDM (single-ply rubber)

EPDM is the big black rubber roof on a million strip malls and schools. Its two classic failure modes are seam failure and shrinkage. Older EPDM was seamed with adhesives that give out, and the membrane itself shrinks over decades, pulling at the perimeter and at every penetration until flashings tear. A shrinking EPDM roof telegraphs stress at the corners and curbs. Because EPDM has a long service window (often into the high 20s), an EPDM building flagged at 20-plus years with visible seam or flashing distress is a strong, patient target — these owners often know the roof is old and are waiting for a reason or a budget cycle.

TPO and PVC (heat-welded single-ply)

The thermoplastics are the workhorses of new commercial roofing now, and their age story has a wrinkle worth knowing. First-generation TPO from the mid-2000s underperformed — thin membranes and formulation issues led to early failures, so some pre-2010 TPO roofs are due years ahead of the nominal service life. That's a targeting edge: a TPO roof that's only 14 or 15 years old but from that early window can be a genuine prospect, where you might otherwise have skipped it as too young. PVC ages by losing plasticizer and getting brittle, eventually cracking, especially at flashings and around rooftop traffic paths. Both show their age by graying, holding dirt and biological streaking, and developing weld-line problems you can sometimes read from above.

SPF (spray foam) and the recoat motion

Spray polyurethane foam roofs aren't a tear-off prospect on the normal clock — they're a recoat prospect. SPF needs a fresh elastomeric coating roughly every 10 to 15 years to protect the foam from UV. A foam roof that's 11 or 12 years out from its last coat is a maintenance-and-recoat sale, not a reroof, and it's a smaller, faster, less competitive ticket that builds the relationship for the eventual bigger job. Don't lump SPF into your reroof filter and then ignore it; give it its own "due for recoat" tier.

The practical upshot: when your age data carries a system type, use it to shade your messaging. A patched 22-year BUR owner is in pain now. A 21-year EPDM owner is waiting for a budget. An early-generation TPO building is due ahead of schedule. A foam roof wants a recoat call. Same age column, four different conversations.

Where commercial flat roof age data actually comes from

There is no national registry of "the day this roof was installed." Anyone who tells you they have an exact install date for every building is selling you certainty that doesn't exist. What does exist is a set of sources that each get you part of the way, and the trick is knowing what each one can and can't do.

Public records (county assessor, permits)

What you get: Year built, square footage, owner of record, sometimes a permit history.

The roof-age catch: Year built is not roof age. The useful nugget in public records is the permit history — a re-roof permit is the closest thing to a real install date you'll find. If a county exposes permit data and you can find a roofing permit pulled in 2009, you now know that roof is about 15 years old, full stop. The problem is coverage: many commercial reroofs are permitted under vague descriptions ("commercial alteration"), some jurisdictions don't require permits for like-for-like recover, and a lot of older permit data was never digitized. Permit data is gold when it's there and silent when it isn't.

How to use it: Pull assessor data for owner, parcel, and square footage. Pull permit data where available and treat any roofing permit as a hard anchor on roof age. Where there's no permit, you fall back to imagery (next).

Aerial and satellite imagery

What you get: A top-down view of the roof itself, often a current image and a stack of historical images going back 10–20 years depending on the provider and the metro.

The roof-age catch: Imagery doesn't hand you a number; it hands you evidence you (or a model) have to interpret. Two ways imagery produces an age signal:

1. Change detection across a historical stack. If you can see the roof in imagery from 2011 as gravel BUR and in 2019 as bright white TPO, the reroof happened between those dates. That's a genuine, defensible age range — "this roof is 5 to 13 years old" — anchored to the imagery dates. The tighter the imagery cadence, the tighter the range.
2. Condition and material reading from a single current image. Membrane color, ponding stains, patch density, gravel scour, seam shadows, and rust streaking from a single recent image correlate with age and condition. This is softer — it estimates "old / mid / newer," not a year.

Imagery is the workhorse for commercial flat roofs specifically because flat roofs are visible from above in a way sloped residential roofs aren't. You can see the whole membrane, the ponding, the patches, the rooftop units, the prior repairs. That's a real advantage of commercial targeting that residential prospecting doesn't get.

Manufacturer and warranty data

Manufacturers (Carlisle, GAF, Firestone/Holcim, Versico, Johns Manville, and others) hold warranty registration data — effectively an install date — for systems under warranty. You will not get access to this as an outside contractor; it's not a prospecting source. It matters only in that it's the reason you'll occasionally find a building whose roof is younger than imagery suggested (a warranty recover you couldn't see clearly from the air). Mention it here only so you understand why your data will sometimes be wrong in the conservative direction.

Your own history and the obituary file

The most underused commercial roof-age source is the one already in your office. Every roof your company has ever touched has a known install or repair date. Every competitor's job sign you photographed has a date. Every "we'll call you in a few years" note from a maintenance visit is a dated age anchor. Commercial roofing is a small world in any given metro; the buildings you bid on five years ago and lost are now five years closer to due. A disciplined CRM is a roof-age database you already own.

Modeled / synthesized estimates

This is the category that has matured fastest: combining imagery, public records, and learned patterns into a single estimated age range per building. A model trained on millions of roofs can look at the imagery stack, the visible material, the ponding and patch signals, the parcel data, and produce "this roof is most likely 16–20 years old" with a confidence level. Done honestly, this is a range with odds attached, not a date. Done dishonestly, it's a fake-precise number that gets your reps yelled at on site. The next section is about telling the difference.

The unglamorous part: matching, hygiene, and getting the owner right

Targeting fails as often on plumbing as on strategy. You can have a perfect roof-age estimate and still waste it because the parcel doesn't match the building, the owner record points at a shell LLC, or you're mailing the property manager's old address. Spend an hour on the boring parts and the rest of the program works.

Parcel-to-roof matching. Big commercial sites break the one-parcel-one-building assumption. A single parcel can hold three separate buildings with three different roofs and three different ages. A multi-tenant retail strip is one roof but several owners-of-interest. Before you trust a building-level age number, confirm the age estimate is tied to the actual roof footprint you intend to sell, not averaged across a campus. When a parcel has multiple roof planes, treat each major plane as its own line item.

Owner of record vs. decision-maker. Assessor data gives you the legal owner, which is frequently a holding company, a REIT, or an out-of-state LLC. The person who decides on a roof is often a property manager, a facility director, an asset manager, or a regional ops lead. Part of qualifying a Tier A building is finding the human who controls the roof budget. National and regional chains usually centralize this; a local owner-operator is the easy case. Build the habit of recording the real decision-maker the moment you find them — that contact is worth more than the age estimate.

Deduplicate and reconcile. The same building can appear under two parcels, or under slightly different addresses. Your own CRM history has to be reconciled against the parcel pull so you don't cold-call a building you reroofed four years ago (embarrassing) or pitch a maintenance customer as a stranger (worse). Address standardization before you do anything else saves you from both.

Suppress what you shouldn't touch. Pull your existing customers, your active warranties, and roofs you know you recently did out of the prospecting list and into a separate "protect and upsell" track. A recent-warranty building isn't a reroof target; it's a maintenance and reference-customer relationship. Mixing the two makes your reps look like they don't know their own book.

Keep a confidence column, not only an age column. Every age estimate should carry how it was derived (permit / imagery transition / condition read / your own record) and a confidence level. A rep working a Tier A building should be able to see at a glance whether the "19 years" is anchored to a 2005 permit or inferred from a graying membrane. The derivation changes how hard they lead with it.

This is the layer that separates a list that performs from a list that looks good in a spreadsheet. None of it is exciting. All of it is the difference between a rep walking into the right meeting with the right person and a rep apologizing to a property manager for a mailer addressed to a company that sold the building in 2019.

How accurate is flat roof age data, really?

This is the section most vendors skip, which is exactly why you should read it. If you understand the error modes, you can sell into them instead of getting blindsided by them.

Age is a range, not a date

The honest output of any imagery-based method is a range, because the underlying evidence is a range. If the last visible reroof in the imagery stack happened between the 2014 and 2018 captures, the roof is 6 to 10 years old today. That's not vagueness; that's accuracy. The dishonest move is to collapse "6 to 10 years" into "installed June 2016" to look impressive. When a vendor or a tool gives you a single date with no range, distrust it.

A good rule for reading age ranges:

The error modes that bite commercial roofers specifically

• Recovers and overlays. A new membrane installed over the old one can look like a reroof in imagery (good) or, if it's a same-color recover, can be nearly invisible (bad). You'll sometimes flag a roof as 20 years old that quietly got recovered eight years ago. On-site this looks like two layers at the edge metal — your estimator should always check for it anyway, because a recover changes the whole scope and tear-off price.
• Phased reroofs. Big roofs get done in sections across multiple years. The east half is 4 years old and the west half is 14. A single building-level age number hides this; a good read flags "mixed age" when the membrane color or seam pattern changes across the roof plane.
• Imagery date drift. "Current" imagery in a given metro might be 18 months stale. Your age estimate inherits that lag. It's small, but in a fast-growing market it means some "recent" reroofs won't show yet.
• Solar arrays and equipment. A roof covered in a rooftop solar array is hard to read, and also a building where reroofing just got much more expensive and politically complicated. Treat heavy-solar roofs as a separate, harder bucket.
• Material misread. First-generation TPO and PVC can look similar from above; BUR with a coating can read as single-ply. Material type affects service life, so a misread shifts your due-date estimate. Condition signals (ponding, patches) are more reliable than precise material ID from the air.

The two ways a list lies, and which one is worse

Every targeting list has two failure modes. A false positive flags a building as due when it isn't — you knock, the roof is four years old, you wasted a visit. A false negative misses a building that is due — you never knock, a competitor gets the job. For a commercial reroof prospecting list, false positives cost you a little time and false negatives cost you a whole job. So you should deliberately tune your list to tolerate some false positives in exchange for catching more of the truly-due buildings. A list that's been scrubbed to be "100% certain" has thrown away your best opportunities to look clean. Verify on site; that's what the site visit is for.

Reading a flat roof from the air: the condition signals that sharpen an age estimate

Even when you have a modeled age range, your estimator and your better salespeople should be able to read a flat roof image themselves. It makes the age number make sense, and it turns a cold call into "I noticed the ponding on the northeast corner of your roof." Here's what to look for, top-down:

• Ponding water and ponding stains. Dark rings and bathtub-ring stains where water sits after rain. Chronic ponding is both an age accelerant and a leak predictor. Drainage problems are a real, specific thing you can mention.
• Patch density. Count the visible patches and repairs. A roof with a dozen patches is a roof at the end of its maintenance rope — the owner is spending money to keep a dying roof alive, which is the exact moment a reroof pitch lands.
• Seam shadows and splits. On single-ply, you can sometimes see seam lines telegraphing or lifting. On EPDM, shrinkage pulls the membrane and stresses seams and flashings.
• Gravel scour and bald spots (BUR). Wind and foot traffic scour gravel off built-up roofs, exposing the bitumen. Bald, shiny areas on a gravel roof signal age and UV exposure.
• Membrane discoloration. Bright white TPO/PVC dulls and grays with age and biological growth. A gray, streaked single-ply reads older than a crisp white one.
• Rust streaking. Rust off equipment curbs, drains, and fasteners stains the membrane and signals a roof that's been up a while.
• Rooftop equipment age. Old, rusted HVAC units and the staining around their curbs are a secondary age tell for the whole roof.
• Edge metal and parapet condition. Where you can see it, lifting or rusted edge metal indicates an aging detail package.

None of these alone dates a roof. Together they let you sort a "old/mid/new" bucket and, more importantly, walk into a meeting sounding like you already know the building.

Building a commercial reroof prospecting list, step by step

Here's the actual workflow, the way a disciplined commercial team runs it. Assume you sell TPO/EPDM/BUR reroofs and maintenance and you want a quarter's worth of targets in two counties.

Step 1 — Define the geography and the building profile

Pick the territory you can actually service profitably (drive time, crew capacity, your sweet-spot building size). Set a minimum and maximum roof size — most commercial shops have a band they're efficient in (say 10,000 to 80,000 sq ft). Going too big chases bids you'll lose to nationals; too small isn't worth the mobilization.

Step 2 — Assemble the building universe

Pull every commercial/industrial parcel in the geography from assessor data: parcel ID, owner, owner mailing address, square footage, year built, property type. This is your raw universe — maybe 3,000 buildings.

Step 3 — Attach a roof-age estimate to each building

For each parcel, attach the best age signal available, in priority order:

1. A roofing permit date (hard anchor) where it exists.
2. An imagery change-detection range where a reroof transition is visible.
3. A condition/material read (old/mid/new) where there's no clear transition.
4. Your own CRM history (you bid it, repaired it, or lost it — you know the date).

This is the step where a modeled age-range dataset earns its keep, because doing it by hand for 3,000 buildings is a part-time job for a month. Whether you build it or buy it, the output you want per building is: estimated age range, confidence, system type if known, and visible condition flags.

Step 4 — Score and tier the list

Don't just sort by age. Build a simple score so your reps know what to do with each building. A workable model:

Then tier:

• Tier A — Due now, high confidence. Past or at end-of-life, distress visible, in your size band. These get a rep's time, in person.
• Tier B — Due soon / due but lower confidence. Within a few years of end-of-life, or due but the estimate is soft. These get a call and a drive-by to confirm.
• Tier C — Watch list. Mid-life roofs. These get a mailer or a once-a-year touch so you're top of mind when they fail.
• Tier D — Skip. Recent roofs. Do not spend a dollar here this year. The discipline to not work these is half the value of the list.

Step 5 — Route it

Geographically cluster your Tier A and B buildings so a rep isn't crossing the metro for two appointments. A good day is six to ten Tier A/B buildings in one quadrant, with the rep's tablet showing the aerial and condition flags for each before they walk in.

Step 6 — Verify on site and close the loop

The site visit confirms or corrects the data. Every visit should feed back: actual system, actual condition, recover or not, real or estimated age. This is the part teams skip, and it's the part that compounds. After a year of feeding site truth back into your list, your age estimates in that metro get sharply better, and your CRM becomes the best roof-age database in your market.

A worked example: turning 3,000 buildings into a Tuesday route

Numbers make this concrete. Say you pull two counties and get 3,200 commercial/industrial parcels in your size band.

• Attach age signals. Roughly speaking, a chunk will have permit anchors, more will have imagery-based ranges, the rest get condition reads. Say you end with a usable age signal on ~2,900 of them.
• Filter to roofs at or past end-of-life for their system type. Commercial roof stock skews old, but in a given pull you might land ~18% in the "due now or overdue" band. That's roughly 520 buildings.
• Layer condition: of those 520, the ones with visible ponding or heavy patching — the buildings actively spending to keep a dead roof alive — might be ~140. That's your Tier A.
• 140 in-person targets is a full quarter for one or two reps if they work them properly (visit, document, bid, follow up). The remaining ~380 due-but-cleaner roofs are Tier B — calls and drive-bys.
• The ~2,400 mid-life and recent roofs you don't touch this quarter. You mail the mid-life Tier C ones once and move on.

The exact percentages will vary by market and building stock — these are illustrative, not a promise. The structure is the point: you converted an unworkable 3,200-building pull into 140 in-person targets and 380 phone targets, and you know exactly why each one made the cut. Compare that to sorting by square footage and dialing from the top.

Turning the age signal into the actual conversation

A roof-age list is only worth what your reps do with it on the phone and at the door. The data hands them an opening that doesn't sound like a cold pitch — it sounds like someone who already looked at the building. Here's how to use each tier without overreaching on what the data actually proves.

What you can honestly say

The age signal is a range and the storm signal is odds, so your reps' language has to match. Train them to say "our records show this roof is likely in the 18-to-22-year range" rather than "your roof is 20 years old." The first is true and disarming; the second is a claim you can't back and an easy thing for an owner to swat down ("actually we recoated it in 2019"). When the owner corrects you, that's a win — now you have the real date and a conversation. The honest frame also protects you: you're offering to inspect and confirm, not declaring a verdict from a satellite.

Tier A — due now, visible distress (in person)

These owners are usually already living with a problem. Lead with the specific, observable thing: ponding on a named corner, a cluster of patches, an obviously aged membrane. "I was looking at low-slope buildings on the south side and yours stood out — looks like there's standing water collecting near the northeast drains and a fair number of repairs. We think the roof's likely past 20 years. Worth me coming out to confirm what's actually going on up there?" You're not selling a reroof on the call; you're earning an inspection. The inspection sells the reroof.

Tier B — due but cleaner, or lower confidence (call + drive-by)

Here you're planting a flag for when the roof gives out. "Your building's roof is getting up there in age — likely in the range where we start seeing the first real problems. Nothing urgent today, but I'd rather be the contractor you already know when the first leak shows up than a stranger you find in a panic. Can I do a no-pressure look and leave you a condition report for your files?" The condition report is the gift; it's also the documentation that makes you the obvious call later.

Tier C — watch list (mail, light touch)

You're not having a conversation yet; you're staying visible. A simple, honest mailer or once-a-year touch that says, in effect, "we track building roofs in this area, here's what we do, keep us in your file." The goal is that when a Tier C roof slides into Tier A, your name is already in the facility manager's drawer. Don't waste rep time here; waste a stamp.

The condition report as a door-opener

The single best commercial sales artifact built on age data is a clean, factual condition report you leave behind: the roof's estimated age range, the visible condition issues, dated photos from your inspection, and an honest assessment of remaining life. No claim promises, no scare tactics, no payout talk — just a professional document that makes you look like the adult in the room. Owners keep these. Property managers forward them up the chain. It's the artifact that converts a targeting list into a relationship, and it costs you an inspection and an hour of write-up.

What not to say

Don't promise an exact install date you can't prove. Don't tell an owner their insurance will cover it or that you'll "handle the claim." Don't say anything about their deductible. Don't promise a payout or an approval. Keep every sentence on the roof, the condition, your scope, and your price. The data gives you a reason to be there and something true to talk about; it does not give you license to promise outcomes you don't control.

Where RoofPredict fits

Most of the workflow above is a data-assembly problem, and that's the part that eats a commercial team alive if they do it by hand. RoofPredict exists to do the heavy part: it reads aerial imagery and models the storms each roof has actually taken, and returns, per building, an estimated roof age range with confidence, plus a per-roof storm history. You hand it your territory or your existing prospect list, and it comes back enriched — every building scored by how old the roof is and what weather it's been through — so your Tier A/B/C/D sorting is done for you instead of being a month of parcel-by-parcel grunt work.

Two things it does that matter for commercial flat-roof targeting specifically. First, the age output is an honest range with odds, not a fake install date — because that's the only thing the imagery can truthfully support, and a fake-precise number gets your reps embarrassed on site. Second, on the storm side, it models the storm on each roof rather than just telling you a hailstorm passed through the ZIP. A hail map shows you where it hailed; modeling per roof gets you closer to which buildings the storm actually wore out, paired with the age signal so you're not chasing a fresh roof that happened to be under the storm.

The honest limits, because you should hear them before you decide: it estimates a range, not a certainty, and it will sometimes miss a same-color recover or split the difference on a phased roof — which is exactly why the on-site verification step stays in your workflow. It enriches your own list and your own CRM; it is not a lead service handing you the same building it sold to five competitors. And it tells you which roofs are likely due — the relationship, the bid, and the close are still yours. Used the way it's meant to be used, it turns the targeting step from your bottleneck into the easy part. You can hand it a building you already know and check whether the age range matches what's actually up there.

The storm angle, done the legal way

A lot of commercial roof age searches are really storm searches in disguise: a roofer wants the old roofs that also sit inside a recent hail or wind event, because those are the ones where there's both age-driven failure and a fresh weather event. This is legitimate and valuable — and it's also where commercial roofers get themselves in trouble by drifting into territory they're not licensed for. So let's be precise.

What you can do, all day long: Inspect the roof. Document its condition thoroughly with dated photos. Identify and record storm-related damage you observe. Write an accurate, Xactimate-aligned estimate to repair the scope of work you would perform. Hand that documentation and estimate to the building owner. State plain facts about your own scope. Pair roof age with storm history to decide which buildings are worth inspecting first.

What you must not do, for a fee, unless you are a licensed public adjuster: Negotiate, adjust, or "handle" the owner's insurance claim. Interpret their policy or tell them what's covered. Promise a specific payout, approval, or that a claim will be accepted. Tell a commercial owner their deductible will be waived, absorbed, or made to disappear. Advertise a "free roof." Represent the owner against their insurer. Those acts are unlicensed public adjusting in most states, and doing them — even with good intentions — is how a roofing company ends up in front of a state regulator.

The safe and frankly more professional frame for a commercial owner: you document the roof thoroughly, you write an accurate repair estimate for your scope, you hand it to them, and they decide whether to file. The owner files; the insurer decides coverage. You stay on the documentation and estimating side of the line. On a commercial building, the owner usually has a risk manager or broker who handles claims anyway — your job is to be the contractor whose documentation is so clean and whose estimate is so defensible that they want you doing the work. Age + storm data tells you which buildings to walk first; it does not change the rules about what you can say.

If you want a one-line compliance check for your sales team: if a sentence promises an insurance outcome, predicts coverage, or makes a deductible vanish, cut it. If it describes the roof, the damage, your scope, or your price, keep it.

Common mistakes that wreck a commercial targeting program

A decade of watching teams do this badly produces a short list of repeatable failures. Check your program against it.

1. Treating year built as roof age. Covered above, still the number one mistake. A 1979 warehouse may be on its third roof; a 2015 building has its original. Sorting by year built puts your worst targets at the top.
2. Chasing the biggest roofs regardless of age. A 120,000 sq ft roof that's six years old is a $0 target this year and a tempting trap. Size qualifies the ticket; age qualifies the timing. You need both.
3. Believing a single date. Any tool that gives you an exact install date with no range is hiding its uncertainty. Demand the range and the confidence, and tune your effort to it.
4. Scrubbing the list until it's "clean." Over-filtering to avoid false positives throws away due buildings. Tolerate some wasted visits to catch more real jobs; that's the correct trade for commercial tickets.
5. Ignoring recovers and phased roofs. Train estimators to look for double-layer edge metal and mixed membrane age. A building flagged "due" can have been quietly recovered; a building flagged "recent" can be half old.
6. No feedback loop. Reps visit buildings and the truth they learn dies in their heads. Every site visit should correct the data so next quarter's list is sharper. Skipping this means you re-make the same mistakes forever.
7. Letting the data do the closing. The list gets you to the right door with the right talking point. It does not write the bid, build the relationship, or handle the objection. Teams that expect data to sell for them are disappointed; teams that use it to point their human effort win.
8. Drifting into claims handling. The storm angle is real, the temptation to over-promise on insurance is real, and the line is bright. Stay on documentation and estimate; never on negotiation and payout.
9. Working the whole list instead of the tiers. A tiered list only works if reps respect the tiers. If everyone works Tier A and nobody mails Tier C, you've got no pipeline behind this quarter. The watch list is the seed of next year.

A 30-day plan to stand this up

If you're starting from a spreadsheet and a hunch, here's a month to get to a real targeting program.

Week 1 — Define and pull. Lock your geography, size band, and ideal building profile. Pull assessor data for the whole commercial/industrial universe in that geography. Export your own CRM history of every building you've bid, repaired, or lost — that's your free age-anchor layer.

Week 2 — Attach age. Get a roof-age estimate onto every building, by permit, imagery, condition read, or your own records. Decide whether you're building this or buying enriched data; for most shops, buying the enrichment and spending your time selling is the better math. Flag confidence on each estimate.

Week 3 — Score, tier, and route. Build the simple score, sort into A/B/C/D, and geographically cluster the A/B buildings into routable days. Load the aerials and condition flags onto your reps' tablets so they walk in informed.

Week 4 — Work it and close the loop. Run the routes. Every visit, capture: real system, real condition, recover or not, corrected age, outcome. Feed it back. By the end of the month you have a live, self-improving targeting list and a quarter of qualified Tier A work in front of your reps.

The difference between this and the old way isn't effort — both are work. The difference is that every hour your reps spend is pointed at a building that's actually due, with a talking point already in hand, instead of dialing down a list sorted by the wrong number. Roof age is the right number. Get it attached to your buildings, respect its honest range, verify it on site, and let it point your people. That's the whole game.

The short version

If you remember five things from all of the above:

1. Roof age — as a range — is the only commercial targeting signal that maps to the buying event. Square footage and year built don't.
2. The data comes from permits (hard anchors), imagery change-detection (real ranges), condition reads (soft ranges), and your own CRM (free and underused). Stack them in that priority.
3. Age is a range with odds, never a date. Distrust fake precision, and tune your effort to the confidence.
4. Tier the list (due now / due soon / watch / skip) and respect the tiers. The discipline to skip recent roofs and to mail the watch list is where the program pays off.
5. On storms, stay on the documentation-and-estimate side of the line. Document, photograph, estimate your scope, hand it over. Never negotiate the claim, interpret coverage, promise a payout, or erase a deductible.

Do that and your reps stop working a list that lies to them and start working the buildings that are actually due.

FAQ

Is roof age the same as the year a building was built?

No, and confusing the two is the most common targeting mistake in commercial roofing. Year built tells you when the structure went up; roof age tells you when the membrane was last installed. A building from 1985 may be on its third roof, while a 2016 building still has its original. Always target by roof age, not year built.

How accurate is flat roof age data from aerial imagery?

Honest imagery-based age data is a range, not a date. If a reroof transition is visible between two historical captures, you get a tight range (often within 3 to 5 years). Where there's no clear transition, you get a softer old/mid/new read. The right answer is always a range with a confidence level. Any source giving you an exact install date with no range is hiding its uncertainty.

Why is a range better than an exact roof install date?

Because the underlying evidence is a range. The imagery shows a reroof happened between two capture dates, so the truthful output is 'this roof is 6 to 10 years old,' not 'installed June 2016.' A range is the accurate answer; a fake-precise date just sets up your reps to be wrong on site, which costs you credibility on the building.

What roof conditions can you read from an aerial image?

Ponding water and ponding stains, patch density, seam splitting on single-ply, gravel scour and bald spots on built-up roofs, membrane discoloration, rust streaking, and aging rooftop equipment. None of these dates a roof alone, but together they let you sort an old/mid/new bucket and walk into a meeting already knowing the building's problems.

Why does flat roof age data work better for commercial than residential?

Flat low-slope roofs are fully visible from above, so you can see the entire membrane, ponding, patches, prior repairs, and equipment. Sloped residential roofs hide much of their condition at an angle. Commercial flat roofs also have longer, more predictable service-life windows by system type, which makes an age-to-due-date mapping cleaner.

How do I build a commercial reroof prospecting list from roof age data?

Define your geography and building size band, pull the full commercial parcel universe from assessor data, attach a roof-age estimate to each building (permit, imagery, condition read, or your own CRM), score and tier them into due-now / due-soon / watch / skip, cluster the top tiers into routable days, and verify on site. Feed every site visit back into the data so it sharpens over time.

Should I worry more about false positives or false negatives in a targeting list?

False negatives. A false positive (flagging a recent roof as due) costs you one wasted visit. A false negative (missing a truly due roof) costs you a whole commercial job to a competitor. So tune your list to tolerate some false positives in exchange for catching more genuinely due buildings, and use the site visit to confirm.

What's a recover, and why does it throw off roof age data?

A recover is a new membrane installed over the existing one rather than after a full tear-off. If it's a different color it shows up in imagery; if it's a same-color recover it can be nearly invisible, so a roof that looks 20 years old may have been recovered eight years ago. Estimators should always check for double-layer edge metal on site, since a recover changes tear-off scope and price.

Can I use storm data with roof age data to target commercial buildings?

Yes. Pairing roof age with recent hail or wind exposure helps you prioritize buildings that have both age-driven wear and a fresh weather event. The legal line matters, though: you may inspect, document damage with dated photos, and write an accurate repair estimate for your own scope. You may not, for a fee, negotiate or handle the owner's claim, interpret their policy, promise a payout or approval, erase a deductible, or advertise a free roof. The owner files and the insurer decides coverage; you stay on the documentation and estimate side.

Does roof age data replace sending reps to inspect buildings?

No. The data points your reps at the buildings that are actually due and arms them with talking points, but the site visit confirms the system, condition, whether there's a recover, and the real scope. The visit also feeds corrected truth back into your list, which improves every future pull. The data is the targeting engine; your people still do the inspecting, bidding, and closing.