How to Target Older Subdivisions for Roofing: A Field Playbook for Finding the Roofs That Are Due

Older subdivisions are the most underworked goldmine in residential roofing, and most crews drive past them every single day. A 1998 tract of 320 homes that all got builder-grade three-tab shingles in the same six-month stretch is not 320 random doors. It is a cohort. Those roofs aged on roughly the same clock, baked under the same sun, and took the same storms. Somewhere around year 18 to 25, that whole pocket starts tipping into replacement range at once. If you can read which subdivisions are sitting in that window right now, you stop knocking cold streets and start knocking streets where a third of the houses genuinely need you.

The problem is that "old neighborhood" is not a strategy. Drive any metro and you will find streets that look tired but have brand-new roofs, and streets that look fine from the curb but are quietly past due. The eye lies. Pin a subdivision on the wrong build year and you waste a $4,000 mail drop or a week of canvasser payroll on houses that re-roofed three years ago. The skill that actually moves your close rate is figuring out, street by street, which roofs are due, then matching your effort to the cohort instead of spraying the whole map.

What follows is the workflow I would hand a new sales manager who has a territory map, a small mail budget, and two green canvassers. It covers how to read a subdivision's age from public records and imagery, how to score it for storm wear without making claims you are not allowed to make, how to sequence your mail and door knocking so the two reinforce each other, and the specific mistakes that quietly bleed money out of an older-neighborhood program. There are worked numbers, checklists, and scripts you can lift directly.

Why older subdivisions beat random territory

Start with the math, because the math is what makes a believer out of an owner who has only ever brute-forced the whole street.

In a randomly chosen residential ZIP, the share of roofs in active replacement range on any given year is small. Asphalt shingles dominate the U.S. market and a typical architectural shingle carries a manufacturer life of 25 to 30 years, while older three-tab runs closer to 15 to 20 in real climates. But the housing stock in a mixed ZIP spans decades, so on any given block you might have a 4-year-old roof next to a 22-year-old roof next to a 9-year-old roof. Knock all three and you are wrong two times out of three before you say a word.

A single-builder subdivision collapses that variance. When a tract goes up, the homes are roofed within a tight window, often the same season, with the same product spec the builder negotiated in bulk. That synchronization is the whole edge. It means age clusters, and clustered age means that when the cohort hits its replacement window, you get a dense pocket of qualified doors instead of scattered ones.

Here is a simplified worked example to make the density real. Two areas, each with 300 homes you could work this month:

Same payroll, same hours, roughly three times the real conversations with people whose roof is actually aging out. The percentages above are illustrative of the structure, not a survey result, but the structure holds in the field: cohort density is the reason an older subdivision out-earns scattered territory at the same cost.

The second reason is conversation quality. In a cohort, your social proof compounds. When you have already done two roofs on Maple Court, the third door knows their neighbors used you, can see the work, and is primed because their roof is the same age as the two that just got replaced. Density turns every completed job into a billboard for the next four. You cannot manufacture that effect in a spread-out ZIP where your nearest job is two miles away.

How a subdivision ages: the cohort window you are hunting

Before you can target by age you need a working model of when a tract becomes worth working. The honest version of this is a range, not a date, and the range depends on roofing material, climate, and storm history.

The replacement-range rule of thumb

For the bread-and-butter asphalt market, here is the field model I use to decide whether a subdivision's original roofs are likely in play:

• Three-tab asphalt (common builder spec pre-2005): real-world replacement pressure builds from roughly 15 to 20 years.
• Architectural / laminate asphalt (common builder spec mid-2000s onward): roughly 20 to 28 years.
• Hot Sun Belt climates (intense UV, high attic temps): shave a few years off the top of the range.
• Northern freeze-thaw and ice-dam climates: flashing and underlayment failures often drive replacement before the field shingle is technically spent.

The takeaway: a subdivision built in 1999 with builder three-tab is dead-center in its window in the mid-2020s. A subdivision built in 2010 with architectural shingle is approaching the early edge of its window now and will be prime in a handful of years. You are hunting more than just old, you are hunting on the leading edge of the window, where the roofs are due but most have not been replaced yet, so the inventory of qualified doors is largest.

Why the second-roof cohort matters too

Here is a piece most crews miss. A subdivision built in 1985 has, in many cases, already been re-roofed once around 2005 to 2010. That second roof is now itself 15-plus years old. So a genuinely old subdivision can have two overlapping cohorts: a minority of survivor originals and a majority of first-replacement roofs that are themselves coming due. The neighborhood reads as "old" by build year, but the relevant clock is the re-roof clock, which is younger than the house. This is exactly why build year alone gets you in trouble, and why you confirm age against the actual roof, never the county's year-built field alone.

Step 1 — Build the build-year map of your service area

You cannot target subdivisions you have not inventoried. The first real work is turning your service area into a ranked list of named subdivisions with an estimated original-roof era for each. This is a few hours of desk work that pays for itself the first week.

Pull the public year-built data

Every U.S. county assessor publishes year-built on the property record. Your sources, cheapest first:

1. County assessor / property appraiser site. Most let you search by parcel or subdivision name and show year built, lot, and often a sketch. Free.
2. County or city open-data / GIS portal. Many publish parcel layers you can filter by year-built ranges and export. This is the fastest way to paint a whole map by era.
3. The U.S. Census Bureau for area-level housing-age context (median year structure built by tract), useful for prioritizing which parts of a metro skew to your target eras.

What you are building is a simple table. For each subdivision or HOA you can name, record: subdivision name, dominant build year (or a 2-3 year range), approximate door count, and the builder if you can find it. A subdivision platted and built in one go will show a tight cluster of identical year-built values on the assessor map. That tight cluster is your signal that it is a true cohort and not a slowly-infilled area.

The trap in county year-built data

Year-built is the year the house was built, not the year the roof was last replaced. The assessor almost never updates year-built when a homeowner re-roofs, because a re-roof under an existing footprint usually is not a reassessable improvement and often does not even pull a separate visible record. So county data tells you the house's age reliably and the roof's age only as an upper bound. A 1999 house has a roof that is at most ~26 years old, but it could be 3. You will resolve that gap in Step 2.

Rank by era, not merely by old

With the table built, tag each subdivision into a priority band:

• Band A — Prime now: original-roof era lands the cohort squarely in replacement range today (e.g., 1998-2006 three-tab, or older tracts whose first re-roof is now 15-plus).
• Band B — Leading edge: cohort entering the window in the next 1-3 years (e.g., 2008-2012 architectural). Worth a mail toehold now, prime soon.
• Band C — Recently worked / too new: large share of confirmed recent roofs, or build year too young. Park it.

Work Band A hard, seed Band B, ignore Band C until it ages in. This single ranking is what separates a targeted program from "let's mail the whole 770xx."

Identify the builder and the original spec

Knowing who built a tract sharpens every later decision, because the builder's original shingle spec sets the replacement clock. A production builder who put down builder-grade three-tab on 300 homes in 2001 created a cohort that ages faster than a semi-custom builder who upgraded everyone to architectural laminate the same year. You can often find the builder from the original plat or subdivision name, older marketing listings, or simply by asking the first one or two homeowners you talk to ("Do you know who originally built out here?"). Veteran homeowners love telling you, and you bank intel for the whole street.

The spec matters in a second way: subdivisions that all got the same discontinued or problem-prone shingle line tend to fail together and visibly, which makes the cohort both more uniform and more receptive. You do not need a database for this, a few conversations and a careful look at a couple of roofs tell you what the tract was built with.

Map the access and route, not only the addresses

A build-year map gets better when you overlay the practical geography of working it. Note which subdivisions have a single entrance versus through-streets, where the cul-de-sacs cluster (dense, walkable, canvasser-friendly), and which tracts have HOA or gated access that will slow a door-to-door push. A 300-home tract of tight cul-de-sacs is a far more efficient canvass than 300 homes strung along a mile of busy arterial. Bake access into your band ranking so the subdivisions you launch first are both prime by age and cheap to physically work.

Step 2 — Confirm roof age against the actual roof

Build year gets you the candidate list. Now you confirm which homes in a candidate subdivision still wear an old roof versus the ones already re-roofed, because every already-done house you mail or knock is wasted spend and, worse, makes your data look unreliable to your own crew.

What aerial and street imagery can and cannot tell you

From good aerial and street-level imagery a trained eye can read signal:

• Granule loss and color fade across the field, especially on south- and west-facing planes that take the most sun.
• Patches and mismatched planes that suggest prior repairs or partial work.
• Obviously new, uniform, high-contrast shingle that flags a recent full replacement, take that house off the list.
• Plane geometry and complexity for later estimating, but that is measurement, a different job from age.

What imagery cannot give you is an exact install date. Nobody can read a date off a roof. The honest output of imagery review is a range and a probability, not a certainty. A roof that reads heavily weathered on a 1999 house is very likely an aging original or an aging first re-roof; a roof that reads crisp and uniform is probably recent. You are sorting the subdivision into likely-due and likely-done, not stamping dates.

A field reference for reading roof age from above

Train your own eye, and your canvassers' eyes, against a consistent set of tells. Here is the read I use, roughly in order of reliability:

The single most useful habit is comparing a roof to its neighbors in the same cohort. If 1640 Maple looks crisp while 1636 and 1644 look chalky and worn, 1640 almost certainly re-roofed recently and comes off your list, even though all three were built the same year. That relative read inside a known cohort is far more reliable than judging any single roof in isolation, which is exactly why subdivision targeting beats scattered addresses: you have a built-in control group on every street.

Watch the imagery capture date

Aerial imagery is not live. A given tile might be eighteen months old or three years old depending on the provider and the area. A roof that reads "new" in a 2022 capture is now several years into its life, and a roof that looked old in an old capture may already be replaced. Always check the capture date before you act, and when in doubt, the ground-level drive-by during canvassing is your tiebreaker. Imagery narrows the list; eyes and the inspection confirm it.

A fast manual confirmation pass

If you are doing this by hand for a target subdivision, a practical pass looks like:

1. Open the subdivision in an aerial imagery tool with recent capture.
2. Spot-check 15-20 homes spread across the tract. Note how many read weathered original versus obviously new.
3. Compute a rough due-rate: weathered ÷ checked. If 12 of 20 read weathered, you are looking at roughly a 60% candidate subdivision, that is a Band A you mail and knock now.
4. Cross-reference any building-permit data your jurisdiction publishes. Some cities expose re-roof permits, which directly flag already-done houses, pull those addresses off your list.

This manual pass is real work, maybe 30-45 minutes per subdivision. It is also exactly the work that gets tedious and error-prone across hundreds of subdivisions, which is where house-by-house age data earns its keep, covered below.

The re-roof permit angle

Many municipalities require a permit for a full roof replacement and publish them in open-data portals. A re-roof permit is the single cleanest "this house is done" signal you can get, because it is a dated record of the exact event. Pulling re-roof permits for a target subdivision lets you subtract the confirmed-done homes before you spend a dollar. The catch: permit data is uneven, some jurisdictions publish it cleanly, others bury it or do not require permits for like-for-like re-roofs at all. Use it where you can get it, do not rely on it everywhere.

Step 3 — Layer in storm wear without crossing the legal line

Age tells you which roofs are tired. Storm history tells you which tired roofs got pushed over the edge faster. The two together are far stronger than either alone, an old roof that also ate a hail event is a more urgent, more legitimate conversation than age alone. But storm and claims territory is where roofers get themselves in trouble, so be precise about what you are and are not allowed to do.

What you may do, and what you may not

Draw this line clearly for your whole sales team, because crossing it can be unlicensed public adjusting in many states:

You MAY:

• Inspect a roof and document its condition thoroughly with photos and notes.
• Identify and document storm-related damage you observe.
• Write an accurate, Xactimate-aligned estimate to repair your own scope of work.
• State facts about your scope to the homeowner and, on your own work, to the carrier.
• Hand the homeowner the documentation and estimate so the homeowner can file.

You MAY NOT (for a fee, on the homeowner's behalf):

• Negotiate, adjust, or "handle" the insurance claim.
• Interpret the homeowner's policy or tell them what is or is not covered.
• Promise a specific payout, approval, or that a claim will go through.
• Promise the deductible will be waived, absorbed, eaten, or "taken care of."
• Advertise a "free roof."
• Represent the homeowner against their insurer.

The safe frame is simple: you document and estimate, the homeowner files, and the insurer decides coverage. Build your storm targeting and your scripts entirely on the document-and-estimate side. The do-not-say list above is not legal advice, confirm specifics with your state's licensing rules, but it is the operating posture that keeps a roofer on the right side of the line.

Where to get real storm history

For the storm side, ground yourself in authoritative public data rather than rumor:

• NOAA's Storm Prediction Center and Storm Events Database for hail and wind event records by date and area.
• The National Weather Service local office archives for storm reports and warnings.
• The Insurance Institute for Business & Home Safety (IBHS) for research on hail and wind damage to roofing assemblies, useful for understanding what damage actually looks like and educating homeowners factually.

The limitation of public storm data is granularity. A Storm Prediction Center report or a hail-swath map tells you a storm passed through an area on a date. It does not tell you which specific roofs took meaningful impact, because hail falls unevenly, roof slope and orientation change impact angle, and an older worn shingle fails at a lower threshold than a new one. So a hail map is a starting filter for "this area had events," not a per-house damage verdict.

Reading age and storm together

The practical synthesis is a two-axis read on every target home:

The bottom-right cell is where your best conversations live, the roofs the storm actually wore out, on houses that were already aging into replacement range. That is the cohort within the cohort.

Step 4 — Sequence mail and door knocking so they reinforce

You have a ranked subdivision, confirmed which homes are likely due, and layered storm context. Now you convert, and the conversion engine for an older subdivision is mail and door knocking working together, not either one alone.

The cadence that works

Mail warms, knocking closes. Run them as one motion per subdivision:

1. Week 0 — First mail drop to the confirmed-due addresses only (not the whole tract). Message is age-and-neighborhood specific, see scripts below.
2. Week 1-2 — Canvass the same blocks while the mail is fresh. Your opener references the mailer and the neighborhood. Density makes the route efficient.
3. Week 3 — Second-touch mail to non-responders, ideally referencing any jobs you have now started in the subdivision ("we are working on Maple Court this month").
4. Ongoing — Yard signs and door hangers radiating out from every job you land. In a cohort, a sign on a completed roof is the most credible mailer you will ever send.

The reason to mail only the confirmed-due homes rather than the whole subdivision is pure economics. At a realistic mailed-piece cost, blasting 300 homes when 140 are due wastes the spend on 160 wrong doors and trains the neighborhood to ignore your envelope. Tight lists keep cost-per-conversation low and your brand from reading as junk.

A cost-per-job worked example

Numbers to make the cadence concrete for a single Band A subdivision of 300 homes, ~140 confirmed-due:

The dollar figures are placeholders for your real costs, plug in your own, but the shape is the point: a tightly targeted older subdivision drives marketing-cost-per-job into a range that scattered territory and bought leads rarely touch, because you stopped paying to reach wrong doors. Compare that to a per-lead resale service where the same homeowner gets sold to four of your competitors.

Door-knock scripts that fit a cohort

The cohort gives you opener material no cold street can. Use it, and keep every word on the document-and-estimate side of the claims line.

Age-only opener (no storm):

"Hi, I'm with [Company]. We're working [Subdivision] this month because a lot of the original roofs out here are hitting the age where they start to go, yours was built around the same time as the neighbors'. No pressure at all, I'd just take a few photos from the ground and let you know honestly where yours stands. Worst case you find out it's got years left and that's good news too."

Storm-context opener (an event is on record):

"Hi, I'm with [Company]. There was a hail event through here back in [month]. We're going door to door documenting roof condition for folks, your roof's also at the age where damage shows up faster. I'll take photos, document what I find, and write you up an honest report. If there's damage worth filing on, that's your call and your insurer's call, I just give you the documentation so you can decide."

Notice what the storm opener does not say: nothing about free roofs, nothing about deductibles, no promise the claim goes through, no offer to handle the claim. It captures the storm intent and answers only on the documentation side, which is both legal and, frankly, more trustworthy to a wary homeowner.

Train green canvassers on the cohort, not the ladder

A new canvasser in a targeted older subdivision can sound like a veteran on day one, because the neighborhood-and-age talking point is true and easy to deliver without roofing expertise. "Your roof is the same age as the two we're replacing around the corner" is a closer's line that a green hire can say honestly. That is also how you keep new reps from quitting: a rep who knocks doors where a real share are due gets conversations, books inspections, makes money, and stays. Sending the same green rep down random cold streets is how you burn through hires in a season.

Handle the four objections you will hear in an older subdivision

A cohort throws the same objections over and over, which is good news, you can drill the responses until they are reflex. Keep every answer on the document-and-estimate side.

"My roof is fine, it doesn't leak." Most roofs do not leak until late in their failure. Granule loss and brittleness happen years before water gets inside. The honest line: "That's the thing, a roof usually looks fine right up until it doesn't. I'd rather tell you it's got five good years left than wait for the ceiling stain. No charge to take a look."

"I just had someone out / I'm getting a bunch of quotes." Lean into the neighborhood: "Makes sense, a lot of folks out here are looking right now because the whole subdivision is the same age. I'm happy to be one of your photos-and-honest-read stops. If yours has years left, I'll tell you that."

"Is this about insurance / are you going to file a claim for me?" This is your compliance moment. "No, I don't file or handle claims, that's not my job and it's not legal for me to. What I do is document the roof's condition with photos and write you an honest report. If there's something worth filing on, you decide and your insurer decides. I just hand you the documentation."

"How much is a new roof?" Do not quote at the door off nothing. "Depends on the size and what's under there, which is why I take the photos and measurements first. I'll get you a real written number, not a guess on your porch." This protects your accuracy and your credibility.

Step 5 — Document and estimate so the homeowner can act

Targeting puts you on the right doorstep; documentation is what turns the conversation into a job, and on storm doors it is the only thing you are legally there to produce. Build a consistent documentation kit your whole crew uses on every inspection, so the homeowner walks away with something real and you walk away with a clean record.

A solid inspection documentation packet includes:

• Dated, geotagged photos of every plane, the ridge, valleys, flashing, penetrations, and any damage, wide shots and close-ups.
• Notes on condition in plain language: granule loss, exposed mat, cracked or missing shingles, flashing condition, soft decking.
• For storm doors: photographs of observed impact marks or wind damage with a reference object for scale, tied to the date of the documented event from NOAA records.
• An accurate, line-item repair estimate for your scope of work, ideally Xactimate-aligned so it reads cleanly to everyone downstream.
• A clear handoff: the homeowner keeps the photos and estimate; if there is damage worth filing on, they file and the insurer decides.

That last point is the legal spine of the whole storm motion. You document and estimate your scope. You do not interpret the homeowner's policy, you do not promise a payout or approval, you do not say a word about their deductible being waived or absorbed, and you never advertise a free roof. Doing the documentation thoroughly and handing it over is both fully compliant and more persuasive than any promise you are not allowed to make, because it gives the homeowner something concrete to act on instead of a sales pitch to distrust.

Mine your own book, the cohort you already own

The most overlooked older subdivision is the one already sitting in your CRM. Every roof you inspected and lost five years ago, every repair customer, every "not yet" from 2019, is a home that is now five years older and likely deeper into its window. These are warm because you already have the relationship and often the roof's history in your notes.

A simple re-engagement pass:

1. Pull every estimate and inspection from 4-8 years ago.
2. Filter to homes whose roof you noted as aging or borderline at the time, they are now squarely due.
3. Cross-reference those addresses against your subdivision priority bands and any storm events since.
4. Reach out with a specific, personal touch: "We looked at your roof in 2019 and gave it a few more years, you're right around the age now where it's worth a fresh look."

This is money already in your book. It costs almost nothing to work, it has the highest trust of any list you own, and it pairs perfectly with subdivision targeting because your old estimates cluster in the same neighborhoods you are already working.

How RoofPredict fits this workflow

Everything above is doable by hand. The reason most crews do not do it consistently is that the desk work, painting build-year maps, spot-checking imagery across hundreds of subdivisions, subtracting the already-done homes, and layering storm context per roof, gets crushing at scale. That is the gap RoofPredict is built to close.

RoofPredict scores roofs house by house from aerial imagery and weather data, and gives you, per address, a roof-age range (not a date, a range, because nobody can read an exact install date off a roof) plus a storm read modeled on that specific roof, rather than only whether a storm passed through the ZIP. The storm piece is the part a hail map cannot do: a hail map shows where it hailed, the per-roof model estimates which roofs the wind and hail actually wore out, accounting for the fact that an older worn shingle fails at a lower threshold than a new one. Paired with age, that is the two-axis read from Step 3 done across your whole service area instead of one subdivision at a time.

Practically, it does three things for an older-subdivision program:

• Ranks your streets and doors so your crew knocks and mails the homes that are actually due and skips the new ones, the Step 1 and Step 2 work, automated and house-by-house.
• Enriches your own list or CRM with roof-age and storm signals, so the re-engagement pass in Step 5 and your direct-mail list both get sharpened with data you would otherwise eyeball.
• Arms a green canvasser with a real, honest per-home talking point so a new hire sounds informed at the door without climbing a ladder.

The honest limits, because a tight trade compares notes: roof age comes back as a range, not an exact date, and the storm read is odds, not proof, it tells you which roofs the storm most likely wore out, it does not certify damage and it does not replace an actual inspection. It also does not handle claims, measure the roof for your estimate, or buy you leads, those are different jobs. What it does is answer the one question this whole playbook turns on, which roofs are due, across every subdivision in your area at once. If you want to see it against ground truth, hand it a roof you already know and judge whether it nailed the call.

Common mistakes that bleed money

The difference between a program that prints money and one that quietly loses it usually comes down to a handful of avoidable errors.

Targeting by build year alone

The biggest one. Build year is the house, not the roof. A 1985 subdivision may be mostly re-roofed; a 2002 subdivision may be mostly original. If you mail by year-built without confirming roofs, you will hit a wall of already-done houses and conclude older-subdivision targeting "doesn't work," when really you skipped Step 2. Always confirm against the actual roof.

Mailing the whole tract instead of the due homes

Blasting all 300 homes when 140 are due nearly triples your wasted spend and trains the neighborhood to trash your mail. Tight, confirmed lists keep cost-per-conversation low and your brand credible. The whole point of targeting is to not spray.

Reading the curb instead of the roof

Tired-looking houses can have new roofs; tidy houses can be past due. Curb appeal is not roof condition. Your canvassers will instinctively prejudge by the yard, retrain them to judge by the plane, the granule loss, the fade, the patches, or by the data, not the lawn.

Crossing the claims line in the storm pitch

The fastest way to turn a great storm opportunity into a legal and reputational problem is a rep who promises a free roof, says the deductible is covered, or offers to "handle the claim." Drill the do-not-say list. Document and estimate, let the homeowner file, let the insurer decide. The compliant pitch is also the more credible one to a skeptical homeowner.

Treating storm and age as separate programs

Roofers often run a storm-chasing motion and an age motion as two disconnected efforts. The strongest doors live where the two overlap, old roofs that took a real event. Layer them. An old roof with storm history is a more urgent and more legitimate conversation than either signal alone.

Working a subdivision once and leaving

A cohort does not convert all at once. Some homeowners need to see two neighbors do it first. Yard signs, door hangers, and second-touch mail compound over a season. The crew that comes back to a subdivision after landing its first three jobs harvests far more than the crew that knocks once and never returns.

Ignoring the re-roof clock on genuinely old tracts

On a 1980s subdivision, the relevant cohort may be the 2005-2010 first replacements, now 15-plus years old, not the survivor originals. Read the roofs, not the house age, and you will find a live cohort hiding inside a neighborhood you might have written off as "already done."

Measure by subdivision, not by month

Most roofers track marketing as one undifferentiated number: spend in, jobs out. That hides the only signal that makes older-subdivision targeting compound. Track every metric by subdivision so you learn which cohorts actually pay and can pour next month's budget into the winners.

The scoreboard that matters per subdivision:

A low inspection-to-signed rate in a subdivision is usually a targeting problem, not a closing problem: you knocked too many already-done or not-yet-due homes, which means tighten Step 2 next time. A high cost-per-job in a tract that looked prime tells you the access was bad or the cohort was already worked by a competitor. Reading these per-subdivision turns a year of canvassing into a map of where your money actually works, which is the entire payoff of targeting in the first place.

Keep it simple to start, a spreadsheet with one row per subdivision beats a fancy system you will not maintain. The discipline of attributing results to the specific tract is worth more than the tooling.

When older-subdivision targeting is the wrong play

Honesty keeps you out of dumb spends. Cohort targeting underperforms in a few real situations, and knowing them saves budget:

• Heavily-worked metros after a big storm. If a hail event already drew a swarm of crews and half the prime tracts are tarped or done, the cohort advantage is mostly spent. Look to the leading-edge Band B tracts the swarm skipped.
• High-churn rental or investor-heavy tracts. Where most homes are rentals, the decision-maker is an out-of-area owner, not the person at the door, and the age play loses its neighbor-social-proof punch. Different motion, different list.
• True custom-home areas. Homes built one at a time over fifteen years have no cohort, the whole edge evaporates. Treat those like a mixed ZIP and lean on per-home roof data instead.
• Premium-material neighborhoods. Tracts roofed in standing-seam metal, tile, or 50-year products age on a different, much longer clock; do not assume a 1995 build means a due roof there.

Knowing where the model breaks is part of using it well. The cohort edge is strongest in stable, owner-occupied, single-builder asphalt-shingle subdivisions in their replacement window, that is the sweet spot, and there are usually more of them in any metro than one crew can work in a year.

A 30-day rollout checklist

If you are starting from a map and a small budget, here is the sequence that gets a targeted older-subdivision program live in a month.

Week 1 — Inventory and rank

• Pull year-built data from the county assessor / GIS for your service area.
• List and name every subdivision you can, with dominant build era and door count.
• Tag each into Band A (prime now), B (leading edge), C (park it).
• Pick your top 2-3 Band A subdivisions to launch.

Week 2 — Confirm and subtract

• Spot-check 15-20 roofs per launch subdivision in aerial imagery; estimate the due-rate.
• Pull re-roof permits where your jurisdiction publishes them; subtract confirmed-done homes.
• Pull storm-event history (NOAA SPC / NWS) for the area and dates.
• Build a clean mail list of confirmed-due addresses only.

Week 3 — Launch the cadence

• Send mail drop 1 to confirmed-due homes.
• Train canvassers on the age and storm openers; drill the do-not-say list.
• Canvass the same blocks while mail is fresh.
• Book inspections; document thoroughly with photos.

Week 4 — Compound and measure

• Send second-touch mail to non-responders, referencing live jobs.
• Plant yard signs and door hangers around every landed job.
• Run a CRM re-engagement pass on 4-8-year-old estimates in the same neighborhoods.
• Track marketing-cost-per-job by subdivision; double down on the bands that convert.

Work that loop, measure cost-per-job by subdivision, and let the winners tell you where to spend next month. Older subdivisions reward the crew that reads the cohort instead of the curb, sequences mail and knocking as one motion, and stays on the right side of the claims line. The roofs are due. The only question is whether you are the crew that knew which ones.

If the house-by-house part, which roofs are due, and which the storm actually wore out, is the piece you would rather not eyeball across hundreds of subdivisions, that is exactly what RoofPredict is for. Book a demo and hand it a street you already know.

FAQ

What is the best build year to target for older-subdivision roofing?

There is no single best year, only the era that puts a subdivision's original roofs in replacement range now. For builder three-tab asphalt (common before the mid-2000s), that is roughly homes built 15 to 20 years ago. For architectural shingle (common since the mid-2000s), it is roughly 20 to 28 years. Target the leading edge of that window, where roofs are due but most have not been replaced yet, so the inventory of qualified doors is largest.

Can I just use the county's year-built data to find old roofs?

Use it to build your candidate list, but never as your final answer. Year-built is the age of the house, not the roof. Assessors rarely update it after a re-roof, so a 1999 house could have a brand-new roof or an aging original. Confirm against the actual roof using aerial or street imagery and, where available, re-roof permit records before you spend on mail or canvassing.

How do I tell which homes in an old subdivision already got new roofs?

Spot-check the roofs in recent aerial imagery: obviously uniform, high-contrast new shingle flags a recent replacement, while granule loss, fade, and patches flag an aging roof. Where your city publishes building permits, pull re-roof permits to subtract confirmed-done addresses directly. Imagery gives you a likely range and probability, not an exact date, so you are sorting likely-due from likely-done, not stamping install dates.

Is it better to mail the whole subdivision or just the older roofs?

Mail only the confirmed-due homes. Blasting an entire tract when roughly half the roofs are due wastes spend on the wrong doors and trains the neighborhood to ignore your envelope. A tight list of confirmed-due addresses keeps your cost-per-conversation low and your brand credible, and it lets you reinvest the saved budget into second-touch mail and canvassing on the doors that count.

How should I sequence direct mail and door knocking in an older neighborhood?

Run them as one motion. Send a first mail drop to confirmed-due homes, then canvass the same blocks within a week or two while the mailer is fresh, referencing it at the door. Follow with second-touch mail to non-responders that mentions any jobs you have started in the subdivision, then plant yard signs and door hangers around every landed job. In a cohort, a completed roof is your most credible advertisement.

How do I add storm history without making illegal insurance claims?

Stay strictly on the document-and-estimate side. You may inspect, document damage with photos, and write an accurate repair estimate for your own scope, then hand it to the homeowner so they can file. You may not negotiate or handle the claim, interpret their policy, promise a payout or approval, say the deductible is waived, or advertise a free roof. Document, the homeowner files, the insurer decides. Confirm specifics with your state's licensing rules.

Where do I get reliable storm history for an area?

Use authoritative public sources: NOAA's Storm Prediction Center and Storm Events Database for hail and wind events by date and location, and your local National Weather Service office archives for storm reports. The limit is granularity, these tell you a storm passed through an area on a date, not which specific roofs took meaningful impact, since hail falls unevenly and older shingles fail at lower thresholds. Treat a hail map as a starting filter, not a per-house verdict.

What is the relevant roof age in a 1980s subdivision that has been re-roofed?

Often the re-roof clock, not the house age. Many 1980s tracts were re-roofed once around 2005 to 2010, so the live cohort is those first replacements, now 15-plus years old, rather than the survivor originals. A genuinely old subdivision can read as already-done by build year while hiding an active cohort of aging first-replacement roofs. Read the roofs themselves, not the house age alone.

How does RoofPredict help target older subdivisions specifically?

It scores roofs house by house from aerial imagery and weather data, returning a roof-age range per address plus a storm read modeled on that specific roof, so you can rank streets and doors by which roofs are actually due and skip the new ones across your whole service area at once. It also enriches your own list or CRM with those signals for re-engagement. The honest limits: age is a range not a date, storm is odds not proof, and it does not handle claims, measure roofs, or buy you leads.

Can a new canvasser work an older subdivision effectively?

Yes, and it is one of the best ways to keep new hires from quitting. The neighborhood-and-age talking point is true and easy to deliver without roofing expertise: telling a homeowner their roof is the same age as the two being replaced around the corner is honest and effective. A green rep who knocks doors where a real share are genuinely due gets conversations, books inspections, makes money, and stays, unlike one sent down random cold streets.