How to Get to Aging Roofs Before Competitors Knock Them

Every roofing market has the same quiet problem. There are far more worn-out roofs than there are crews willing to do the legwork to find them. A neighborhood built in 2003 with original architectural shingles is, right now, a street full of roofs reaching the end of their service life. Some of them took a beating in the last wind event. A few homeowners have already noticed granules in the gutter or a stain on a bedroom ceiling. And yet most of those doors will get knocked by whoever happens to drive down that street on a Tuesday afternoon, in no particular order, with no idea which houses are actually due.

That randomness is the opening. The contractors who win the replacement and restoration work in a market are almost never the smoothest talkers. They are the ones who show up at the right houses first, while the roof is still a problem the homeowner is thinking about and before three other yard signs go up on the block. Getting there first is a logistics problem dressed up as a sales problem. Solve the logistics and the selling gets dramatically easier.

What follows is the actual playbook: how to figure out which roofs in your market are aging out or storm-stressed, how to rank streets so your reps spend their hours on the doors most likely to pay, how to knock and document the right way, and how to keep a territory warm so a competitor's canvasser can't poach it out from under you. It is written for owners and sales managers who run crews and want fewer wasted miles, not for anyone looking for a shortcut around doing the work.

Why "first" beats "best" on an aging roof

Roofs do not fail on a schedule you can see from the street at a glance, but they do fail on a curve you can predict in aggregate. Asphalt shingles, which cover the overwhelming majority of U.S. single-family homes, have a usable service life that depends on the product grade, the install quality, the slope and ventilation, and the local climate load. A 3-tab roof in a hot, high-UV market might be tired at 15 years. A laminated architectural shingle in a milder climate can run past 25. The National Roofing Contractors Association and shingle manufacturers publish service-life ranges precisely because there is no single number; what matters for your sales operation is that a roof installed in a known era is statistically inside or past its replacement window.

Here is the part most reps underestimate. The homeowner's willingness to act is highest in a narrow window, and that window opens and closes whether or not you are standing on the porch. It opens when something prompts attention: a storm, a neighbor's new roof, a real estate appraisal, a leak, an insurance non-renewal letter. It closes when the homeowner either acts with someone else or files the problem under "deal with it later" and stops thinking about it.

When you knock a genuinely aging or storm-hit roof inside that window, you are not creating demand. You are catching it. That is why "first" beats "best." The second-best crew that knocks on Tuesday morning closes the roof that the best crew would have closed on Tuesday afternoon. Speed compounds: the first reputable contractor on a block often becomes the de facto referral, because the next four neighbors who get curious ask the household that already went through it.

How a roof actually wears out, and why the curve matters

To target by age you have to understand what age is doing to the roof. An asphalt shingle is a mat saturated in asphalt and surfaced with mineral granules. Those granules are the roof's sunscreen; they block ultraviolet light from breaking down the asphalt underneath. As a roof ages, granules loosen and wash off, first in trace amounts, then in the steady trickle a homeowner notices in the gutter or at the bottom of a downspout. Once enough granules are gone, the exposed asphalt oxidizes, hardens, and grows brittle. A brittle shingle loses its flexibility, curls at the edges, cracks, and gives up the seal strip that bonds each course to the one below it. That lost seal is what lets wind get under a shingle and peel it.

The practical upshot for targeting is that the failure curve is not linear. A roof spends a long, quiet middle stretch looking fine and performing fine, then degrades quickly as it crosses into the last few years of service life. That accelerating tail is exactly the window you want to catch. A roof at year 12 may genuinely have nothing to talk about. The same roof at year 22 can be one wind event away from active leaks. The whole reason age bands are worth scoring is that a relatively small difference in years, late in the curve, maps to a large difference in how ready the homeowner is to act.

Ventilation and slope move the curve. A poorly vented attic bakes the underside of the deck and cooks shingles from below, pulling years off the service life. A steep, south-facing slope takes more direct UV than a shaded north slope on the same house, which is why one side of a roof can look a decade older than the other. None of this changes the targeting method; it just explains why imagery sometimes shows obvious wear on a roof your age data calls middle-aged, and why a confidence range beats a single number.

The cost of knocking the wrong doors

Run the math on a rep's day and the waste becomes obvious. A canvasser working a residential street can realistically attempt 40 to 70 doors in a productive afternoon, depending on density and how many conversations actually start. If only one in five of those houses has a roof anywhere near its replacement window, the rep burns most of the shift on roofs that are 6 years old and homeowners who have no reason to talk. Worse, every one of those premature knocks slightly sours the street: people remember being pitched on a roof they know is fine.

Now flip it. If you can sort that same street so the rep starts with the eight houses most likely to be aging out or storm-stressed, the conversation rate climbs, the rep's morale holds, and the homeowners who weren't ready don't get bothered. Same labor, radically different return. The entire point of targeting is to convert miles into conversations and conversations into inspections.

Put numbers on it. Say a rep works 25 productive afternoons a month and attempts 55 doors each. That is 1,375 attempts. At a random 20 percent qualified rate, only about 275 of those knocks hit a roof anywhere near its window, and your rep spent four out of every five conversations on roofs with no reason to act. Lift the qualified rate to 60 percent through targeting and the same 1,375 attempts now land on roughly 825 due roofs. You did not add a single hour or hire anyone. You roughly tripled the number of real opportunities the same paycheck produced. That delta, repeated across a small sales team over a season, is the difference between a company that grows and one that grinds.

There is a softer cost too. Reps burn out fast on cold streets. A canvasser who hears "my roof's fine, it's six years old" forty times in an afternoon stops believing in the work, and turnover in a door-knocking sales force is brutally expensive to absorb. Feeding reps ranked routes where conversations actually start is as much a retention strategy as a revenue one. Good targeting keeps good people, and good people knock more doors.

Build a map of where the old roofs actually are

You cannot rank what you cannot see. Before any rep knocks anything, you want a working picture of roof age across your service area, organized so you can act on it street by street. There are four practical sources of that picture, and the strongest operations blend them.

Source 1: Housing age as a first approximation

The single most available signal is the age of the house itself. A roof is usually original until proven otherwise, and a sizable share of homes are still wearing the roof they were built with or one early replacement. Year-built data is public and broad. County assessor and parcel records carry a year-built field for nearly every property, and the U.S. Census Bureau's American Community Survey reports the median year structures were built down to the tract and block-group level, which is enough to flag whole pockets of a city built in the same window.

Housing age is a blunt instrument. It tells you a 2001 subdivision is statistically loaded with roofs in or near replacement range; it does not tell you which specific houses already got a new roof three years ago. But as a first pass for deciding which neighborhoods even deserve a canvasser, it is fast, free, and surprisingly directional. Pull the year-built distribution for your county, find the bands that map to roofs now 18 to 28 years old, and you have your starting universe.

Source 2: Permit history to subtract the already-done

The most valuable correction to housing-age data is reroof permit history. Many jurisdictions require a permit to tear off and replace a roof, and those permits are often public record. If a 2002-built house pulled a reroof permit in 2016, that roof is 9 years old and you can stop thinking about it. If the same house has no reroof permit on file, the odds that it still wears its original roof go way up.

Permit data is uneven. Coverage and digitization vary wildly by jurisdiction, some homeowners and even some contractors skip permits, and a permit tells you a roof was replaced but not always the exact scope. Treat it as a powerful subtraction filter rather than gospel: use it to remove houses you can confirm were recently done, and accept that a no-permit house is a maybe, not a certainty. Even an imperfect permit overlay sharpens a housing-age map considerably.

Source 3: Aerial and street-level imagery

The eye in the sky is what turns a statistical guess into a visual read. High-resolution aerial and oblique imagery lets you, or a system trained to do it, actually look at the roof: the shingle pattern, obvious patches, discoloration, streaking, missing tabs, tarps, and the general condition of the field. Street-level imagery adds the eave-line view a drive-by canvasser would get without the drive.

Imagery is where roof age stops being a single number and becomes a range. You cannot read a manufacture date off a photo. What you can read is whether a roof looks consistent with a 5-year-old install or a 22-year-old one, and that visual read, combined with house age and permit history, narrows the estimate to a usable band. Be honest about the limits: shade, tree cover, image date, and a recent cleaning can all fool a visual read. The output you want is a confidence-tagged range, not a false precision.

Source 4: Weather and storm exposure per address

The last layer is what the sky has done to each roof. Wind and hail do not fall evenly. A single supercell can hammer one subdivision with 1.75-inch hail and leave the neighborhood two miles away untouched. The National Weather Service, the Storm Prediction Center, and NOAA archive storm reports, and radar-derived hail and wind products can estimate the maximum size and wind speed that passed over a specific location on a specific date.

For an aging-roof targeting strategy, storm exposure does two things. It identifies recently stressed roofs that may have functional damage worth a documented inspection, and it sharpens the urgency on roofs that were already old. A 20-year-old roof that just took 1.5-inch hail is a far stronger door than a 20-year-old roof in a calm pocket. The key discipline, covered in detail later, is that storm modeling gives you odds and priority, never a guarantee that a given roof is damaged or that anything will be covered.

Stack the layers into a single picture

None of these four sources is sufficient alone. House age over-counts. Permits under-cover. Imagery has blind spots. Storm data is probabilistic. Stacked, they correct each other:

The combined output you are after is a list of addresses, each carrying a roof-age range, a storm-exposure flag, and a confidence level. That list is the raw material for ranking.

A note on data hygiene

Garbage in, wasted miles out. Before you trust any stacked map, sanity-check the inputs. Assessor year-built fields carry errors, especially on homes that were substantially rebuilt or had additions that reset the record. Permit portals sometimes file reroofs under generic "repair" categories that a keyword scrape will miss, and sometimes a permit was pulled but the job never happened. Imagery can be two or three years stale depending on the provider, which matters a great deal for a roof near its window. The fix is not to abandon any source; it is to carry a confidence level on every address and to let your reps' field observations flow back into the data. When a rep knocks a house your map called 22 years old and finds a two-year-old roof, that correction should update the record so you never waste a second knock on it. The map is a living thing, not a one-time export.

Privacy and the right way to handle homeowner data

Everything described here uses property-level signals, public records, and imagery of the structure, not surveillance of people. Keep it that way. Work from address-level roof and storm data, respect do-not-knock ordinances and any local solicitation permit requirements, and honor it immediately when a homeowner asks not to be contacted again. Maintain your own internal suppression list and actually use it. Targeting that respects the homeowner is both the legal posture and the one that protects your brand on a block you intend to work for years. A reputation for knocking politely, only where it makes sense, and leaving people alone when asked is itself a competitive moat.

Rank the streets so reps spend hours where they pay

A list of likely-aging addresses is necessary but not enough. Reps work geographically, not by spreadsheet row, so you have to convert the address-level signal into a street-and-route ranking that a human can actually walk. This is the step most contractors skip, and it is where the biggest efficiency gains hide.

Score the address, then roll up to the street

Give every address a simple priority score built from the signals you already gathered. You do not need a data science team; a weighted sum on a spreadsheet works and is easy to tune. A workable starting model:

1. Roof-age band (0 to 40 points). Older estimated range scores higher. A 22 to 26 year band tops out; a 4 to 8 year band scores zero.
2. Storm exposure (0 to 30 points). Recent significant hail or wind over the address scores high; no notable event scores zero.
3. No-recent-permit bonus (0 to 15 points). Confirmed no reroof permit since the build adds points; a confirmed recent reroof drops the address out entirely.
4. Visible condition flags (0 to 15 points). Imagery showing streaking, patches, missing tabs, or a tarp adds points.

Sum those and you have a 0 to 100 address score. Then roll addresses up to the street segment: average the top half of scores on each block, or count how many addresses on the block clear a threshold (say, 60+). A block with eleven high-scoring houses out of fourteen is a route you send a rep to immediately. A block with two is a maybe-later.

Density is half the battle

The ranking has to respect a brutal field reality: a rep's effective hourly output is dominated by walking distance between live doors. Ten qualified houses spread across a half-mile of large lots is a worse afternoon than ten qualified houses on two adjacent cul-de-sacs. When you rank routes, weight density alongside quality. The ideal target is a tight cluster of high-scoring addresses where a canvasser can move from porch to porch in under a minute.

This is why subdivisions built in a single year are gold. They aged together, the roofs are the same vintage, lots are uniform and close, and a storm that hit one hit most. A single 2002 subdivision of 180 homes with no reroof permits and a recent hail flag can be a week of high-conversion knocking for a small crew.

A worked ranking example

Say you are working a mid-size metro and you have three candidate areas this week:

• Maple Ridge: built 2001 to 2003, 210 homes, dense, took 1.5-inch hail nine days ago, roughly 30 percent show reroof permits. Address scores cluster high; storm flag is fresh; density is excellent. Permits knock out 60-some homes, leaving about 145 strong doors in a tight footprint.
• Old Town: built 1955 to 1975, scattered lots, no recent storm, permit coverage spotty so age is uncertain. Some roofs are surely original and ancient, but density is poor and there is no urgency event to anchor the conversation.
• Westbrook: built 2014, 160 homes, dense, no storm. Roofs are 11 years old, mostly inside warranty, nothing to talk about yet.

Maple Ridge wins decisively. It combines an aging-out vintage, a fresh storm trigger, strong density, and a permit filter that has already removed the houses not worth knocking. Westbrook is a calendar reminder for four years from now. Old Town is a slow-drip route for a rep with patience, worked by walking and reading roofs in person rather than by storm urgency. The ranking didn't just tell you where to go; it told you in what order and why, and it kept your best canvasser out of Westbrook.

Tuning the weights to your own market

The point values above are a sensible default, not a law. Different markets reward different signals, and you should expect to move the weights once you have your own close data. A few examples of how local conditions shift the model:

• Hail-alley markets where most replacements run through storm activity should weight storm exposure heavily, sometimes above age, because a freshly hit roof of almost any age becomes an inspection-worthy door. In these markets the storm flag is the trigger and age is the tiebreaker.
• Calm, high-UV markets with little hail should weight age and visible condition far more, because there is no storm event to anchor urgency. Here you are selling age-out replacements on their own merits, and the imagery read carries more of the load.
• Markets with strong permit transparency can lean hard on the no-permit bonus and confidently drop recently re-roofed homes, sharpening the list considerably. Markets where permits are spotty have to lean more on imagery to compensate.
• Dense urban grids versus large-lot rural change how much density weighting matters. In a sprawling exurb, a slightly lower-scoring but tightly clustered street can out-earn a higher-scoring street where the houses sit an acre apart.

Re-tune quarterly at most; weights that change every week are just noise chasing. Make a change, run it for several weeks of routes, and check whether score-to-close actually improved before changing it again.

Don't let the model override an obvious read

A score is a prior, not a verdict. If a rep stands in front of a house your model scored 40 and sees a tarped slope, missing tabs across the field, and a homeowner already complaining about a leak, that roof is a 90 regardless of what the spreadsheet said. Train reps to trust their eyes over the score when the two disagree, and to feed that observation back so the data improves. The ranking exists to point reps at the best streets, not to talk them out of an obvious opportunity standing right in front of them.

Where roof-age and storm data does the heavy lifting

Everything described so far you can assemble by hand: pull assessor data, scrape permit portals, eyeball imagery, cross-reference storm reports, and build the score in a spreadsheet. Plenty of sharp operators do exactly that, and it works. The catch is that doing it by hand for a whole metro, and keeping it current as storms roll through and permits get filed, is a large standing job. That is the specific gap tools like RoofPredict are built to fill.

RoofPredict's role is narrow and worth stating plainly: it estimates a roof-age range per address from aerial imagery, models storm exposure per roof rather than per zip code, and enriches a contractor's own address list, CRM, or mailing list with those signals so you can rank doors and routes without building the data pipeline yourself. In the language of the playbook above, it automates Source 3 and Source 4 and hands you the stacked, scored list to act on.

What it does not do is equally important, and any honest version of this needs the limits stated. It gives you a roof-age range, not a manufacture date, because no system can read an install date off a photo. It gives you storm odds and priority, not proof that a specific roof is damaged or that any claim will be approved. It does not buy or sell you homeowner leads; it ranks the doors on a list you already control. The judgment, the inspection, the documentation, and the conversation are still yours. Used that way, the data removes the part that wastes reps' hours, which is figuring out where to point them, and leaves the part that actually earns the job, which is showing up and doing honest work.

A reasonable way to fold it in: run your service area through the enrichment to get age ranges, storm flags, and confidence per address; layer your own permit knowledge and local route sense on top; and use the combined score to build the week's canvassing routes. The tool is the engine for the targeting layer, not a replacement for the field craft.

Knock the right way once you're there

Getting to the right door first is wasted if the conversation falls apart on the porch. Targeting buys you a warmer audience; it does not write the script. The reps who convert aging-roof doors share a few habits.

Lead with the roof, not the pitch

A homeowner whose roof is genuinely old or recently storm-hit does not need to be sold on the existence of a problem. They need a credible reason to believe you know something specific about their house. Opening with the specific, observable fact, the age of the neighborhood's roofs, the recent storm date, the streaking you can see from the curb, lands far better than a generic "we're doing roofs in the area." Specific beats salesy. You earned the right to be specific by doing the targeting work.

Offer a documented inspection, and mean it

The ask on an aging or storm-hit roof is simple and honest: a no-pressure inspection where you get on the roof, take photographs, and document the actual condition. That is a real service with real value to the homeowner whether or not it leads to a job. It is also where the legal lines matter, and getting them right protects your business.

What you are offering is documentation and, where appropriate, an accurate repair estimate. You inspect, you photograph, you measure, and you write up what you find in an estimate aligned to standard estimating practice. You hand that documentation to the homeowner. If insurance is in the picture, the homeowner files the claim and the insurer decides coverage. You can state facts about what you observed and what your scope to repair would be. That is your lane, and it is a strong, defensible one.

The do-not-say list

The fastest way to turn a legitimate operation into a regulatory problem is to drift across the line into unlicensed public adjusting or deceptive claims. Train every rep on what they may not say or do, regardless of what a competitor down the street is doing:

• Do not, for a fee, negotiate, adjust, or "handle" the homeowner's claim with their carrier. That is public adjusting and it is licensed in most states.
• Do not interpret the homeowner's policy or tell them what is or isn't covered. You are not their coverage advisor.
• Do not promise a specific payout, approval, or that "insurance will pay for it." You don't decide that; the carrier does.
• Do not promise the deductible will be waived, absorbed, or made to disappear. Offering to eat a deductible is insurance fraud in many jurisdictions and can violate state law outright.
• Do not advertise or imply a "free roof." The Federal Trade Commission and state attorneys general treat that as deceptive, and several state insurance departments specifically warn homeowners about it.
• Do not represent the homeowner against their insurer in any way.

The safe frame is consistent and it is also the more credible one on the porch: "I'll get up there, document exactly what I find with photos, and write you an accurate estimate to repair my scope. You decide what to do with it, and if you file, your insurer decides coverage." Homeowners trust that more than a free-roof promise, and it keeps your license and reputation intact.

Handling the four objections you'll hear on an aging roof

Even on a well-targeted door, the homeowner has reflexes. The conversations that convert are the ones where the rep has an honest answer ready instead of a counter-pitch.

• "My roof is fine." It might be. The honest reply is that roofs this age often look fine from the ground while wearing thin where it counts, and a free inspection settles it either way. You are offering information, not insisting on a problem. If the roof checks out, you say so and leave, and you have just earned a household's trust for when it does need work.
• "I already had someone look at it." Good. A second documented set of photos costs the homeowner nothing and gives them something to compare. Never run down the other contractor; just offer thorough documentation and let it speak.
• "I can't afford a new roof." Decouple the inspection from the purchase. The inspection is free and tells them what they are actually dealing with and how urgent it is. Whether and how they pay for any work is a separate conversation that only matters if the roof needs it. Pushing financing before you have even been on the roof is backwards.
• "Is this about the storm? Will insurance pay?" Here you stay rigorously in your lane. You can say you will document the roof's condition and the storm's date thoroughly and write an accurate estimate for any repair to your scope. You cannot say insurance will pay, what is covered, or that the deductible disappears. The homeowner files and the insurer decides. Reps who promise outcomes here create the complaints that sink companies.

Document like the photos will be scrutinized

Because they might be. Whether the roof goes to replacement out of pocket or through a homeowner-filed claim, your documentation is the spine of the job. Build a standard inspection capture so every rep produces the same evidence package:

1. Wide context shots of each slope showing overall condition.
2. Address and date verification, a shot that ties the photos to the property.
3. Close-ups of every flagged area, with a reference for scale where size matters (hail bruising, missing tabs, lifted shingles, flashing failures).
4. Test square documentation where appropriate, photographed clearly.
5. Collateral damage to soft metals, vents, gutters, and screens, which often corroborate a wind or hail event.
6. Interior evidence if the homeowner reports leaks: ceiling stains, attic decking, daylight.

Keep the photos organized by property and slope, time-stamped, and stored so you can produce the full set on demand. Good documentation is what separates a contractor whose estimates get taken seriously from one whose word is just one more opinion.

Be the first crew on the street, not the fourth

Speed to the door is an operational capability, not a personality trait. Three levers control it.

Lever 1: Trigger-based dispatch after storms

When a significant hail or wind event crosses your service area, the clock starts. The contractors who clean up are the ones who can convert a storm footprint into a knocking route within a day or two, while the homeowners are still attentive and before out-of-town storm chasers flood in. That means having your address-level storm flags update quickly, knowing which aging neighborhoods sit inside the footprint, and being able to dispatch reps to ranked routes the same week.

Build the runbook before the storm: who pulls the impacted footprint, who cross-references it against your aging-roof map, who builds the routes, and who knocks. A crew that has to figure this out after every event will always be the fourth crew on the street.

A workable post-storm runbook looks like this:

1. Within hours: confirm the event footprint from storm reports and radar-derived products, and note the maximum hail size and wind speed bands across your service area.
2. Same day: intersect that footprint with your existing aging-roof map so you immediately see which due neighborhoods sit inside the hardest-hit zones. A 24-year-old subdivision under 1.75-inch hail is your first route, not the newer neighborhood next to it.
3. Next morning: build ranked routes for those overlap zones and brief reps on the specific event date and conditions so they can speak to it accurately and honestly.
4. Days two through ten: knock those routes densely, capturing standardized documentation, while attention is still high.
5. Throughout: keep your suppression and do-not-knock lists live so the surge of activity does not stomp on homeowners who asked to be left alone.

The contractors who own a storm are simply the ones who compressed steps one through three from a week into a day. Speed there is the whole advantage.

Beware the storm-chaser trap, and don't become one

Major events draw out-of-town operators who blanket a footprint, overpromise, and leave. They are your competition for the door, but they are also a cautionary tale. The free-roof pitch, the promise that insurance will surely pay, the offer to absorb the deductible: those are exactly the lines that draw state insurance department warnings and attorney-general action, and homeowners are increasingly wary of them. Your edge as the local, accountable contractor is to do the opposite. Show up first because you targeted well, document honestly, write a real estimate, and tell the homeowner the truth about what the carrier does and does not decide. Being the trustworthy crew on a street full of chasers is a position the chasers literally cannot occupy.

Lever 2: Work your owned list relentlessly

The single most underused asset most roofing companies have is their own database: past customers, past estimates that didn't close, neighbors of completed jobs, and any list they've ever purchased or built. Enrich that list with roof-age ranges and storm flags and you can re-rank it the moment conditions change. The homeowner who passed on a roof two years ago because it "wasn't that bad" is a different conversation after their now-22-year-old roof takes hail. You already have their number. Get to them before a stranger knocks.

Lever 3: Saturate, don't sprinkle

When you commit to a high-ranked neighborhood, work it densely and finish it. Sprinkling one rep across five neighborhoods leaves every block half-knocked and ripe for a competitor to clean up behind you. Concentrating a crew on one ranked cluster until it is fully worked does three things: it raises the conversation rate through neighbor-to-neighbor familiarity, it produces the yard-sign density that makes the next knocks easier, and it closes the street before the competition gets organized. Depth beats breadth when the goal is to own a block.

Equip the field so first-to-the-door is repeatable

Reaching aging roofs first is not a one-time campaign; it is a capability your operation either has or doesn't. A few investments make it repeatable rather than heroic.

Get the ranked route into the rep's hand, not a back-office spreadsheet

The ranking is worthless if the rep can't see it where they work. The route, with each address's age range, storm flag, and score, needs to live on the phone in the rep's pocket, ideally on a map they can navigate door to door, with a way to log the result of each knock on the spot. Whether that is a canvassing app fed by your enriched list or a simpler shared map, the requirement is the same: the rep sees the next best door, not a printout from Monday that is already stale by Thursday.

Standardize the capture so every rep produces comparable work

If one rep takes twelve clean photos per inspection and another takes three blurry ones, your documentation quality is a coin flip and your data feedback is noise. Build a fixed inspection checklist and photo sequence, train to it, and review the output. Consistency is what lets you compare neighborhoods, trust your close data, and produce a credible documentation package on any property on demand.

Train safety before speed

Reaching roofs first means nothing if a rep gets hurt getting onto one. Roofing consistently ranks among the most dangerous trades, and fall hazards dominate the injury data. Whatever your canvassing pace, do not let speed pressure push reps onto roofs without proper fall protection, ladder safety, and clear rules about when not to get on a roof at all. A storm push that produces an injury costs more than every roof it could have closed. Bake safety into the runbook so it is never the thing that gets skipped when everyone is moving fast.

Keep score so you can sharpen the targeting

Targeting is not a one-time setup; it is a loop you tighten with feedback. Track a small set of numbers per route and per neighborhood so you learn which signals actually predict closes in your market, which can differ from the averages.

• Doors attempted per rep-hour, by neighborhood.
• Conversation rate, the share of attempts that become a real exchange.
• Inspection rate, the share of conversations that earn a roof inspection.
• Inspection-to-signed rate, the share of inspections that become jobs.
• Score-to-close correlation, whether your high-scored addresses actually closed at higher rates than low-scored ones.

That last one is the whole game. If your 80-plus scored addresses close at three times the rate of your 50s, your model is earning its keep and you should push more reps toward high scores. If there is no spread, your weights are wrong, your data is stale, or your reps aren't following the routes, and you fix that before buying any more data. Treat the score as a hypothesis you are constantly testing against closed jobs.

A simple weekly cadence

1. Monday: refresh the map, pull any new storm footprints and permits, re-rank the week's candidate routes.
2. Tuesday through Friday: knock the top-ranked clusters densely, capturing standardized documentation on every inspection.
3. Friday: log doors, conversations, inspections, and signs by neighborhood; flag any neighborhood that under- or over-performed its score.
4. Following Monday: adjust weights based on what actually closed, and roll the routes forward.

Run that loop for a season and the targeting gets sharp enough that a rep's average afternoon looks like another company's best one.

Common mistakes that hand roofs to competitors

A few patterns reliably surrender aging roofs to the next crew. Watch for them.

• Knocking by convenience, not by score. Reps default to the nearest or easiest streets. Without a ranked route, they spend the afternoon on roofs that aren't due. Give them the list and hold them to it.
• Trusting house age alone. A 2001 build with a 2017 reroof is not your customer. Skipping the permit-and-imagery correction wastes knocks and annoys homeowners who just got a new roof.
• Treating storm data as proof. A hail flag is a reason to inspect, not evidence of damage or coverage. Reps who overpromise on the porch create complaints and chargebacks.
• Drifting into claims handling. Crossing into negotiating claims, interpreting policy, promising payouts, or erasing deductibles is how good companies attract regulators. Stay on the document-and-estimate side.
• Half-working neighborhoods. Spreading reps thin leaves blocks open for competitors to finish. Saturate and complete.
• Never closing the feedback loop. Without tracking score-to-close, you can't tell whether your targeting works. Measure it or you're just guessing with extra steps.
• Ignoring the owned list. Buying new data while a goldmine of past estimates and customers sits un-enriched is backwards. Mine what you already own first.

Avoid those seven and you are already ahead of most of the crews working your market.

Putting it together

The contractors who consistently reach aging roofs first are not luckier or louder. They have turned a guessing game into a process: a stacked map of where the old and storm-stressed roofs are, a ranking that points reps at dense clusters of due roofs, a knocking and documentation discipline that stays firmly on the right side of the law, a dispatch capability that converts storms into routes within days, and a feedback loop that sharpens the whole thing every week. Each piece is doable by hand. Together, run consistently, they compound into a structural advantage that a competitor knocking doors at random cannot match.

The data layer, knowing which roofs are due by age and which got stressed by the last storm, is the part that scales worst by hand and best with the right tooling. That is where RoofPredict fits: a roof-age range per address, storm exposure modeled per roof, and your own list enriched so you can rank doors and routes honestly, without the false precision of a fake install date or a promised payout. The tool points you at the right doors. Showing up first, documenting thoroughly, and telling the homeowner the truth is still the work, and it is still what wins the roof.

If your reps are knocking streets in no particular order today, start with the cheapest win available: pull your county's year-built data, subtract the recent reroof permits you can find, and rank this week's routes by age and density before anyone laces up. Then layer in imagery and storm signals as you go. The first ranked route you send a rep on will tell you everything about how much money the old random way was leaving on the table.

FAQ

How do I find aging roofs in a specific neighborhood?

Start with public year-built data from your county assessor and the Census Bureau to flag neighborhoods built in a window that puts roofs at 18 to 28 years old. Subtract homes with recent reroof permits where that record is available. Then confirm and narrow with aerial or street-level imagery, which gives you a condition-based age range per address. Layer storm exposure last to prioritize. No single source is reliable alone; the combination is what makes a neighborhood map actionable.

Can I get the exact age of a roof from aerial imagery?

No. You cannot read a manufacture or install date off a photo, and any tool claiming an exact date is overselling. What imagery gives you is a confidence-tagged age range based on visible condition, shingle pattern, patches, and discoloration. Combined with house year-built and reroof permit history, that range is precise enough to decide which doors to knock, which is all the targeting actually requires.

Is storm data proof that a roof is damaged?

No. Radar-derived hail and wind estimates and storm reports tell you the odds that a roof was stressed and how hard a given location was hit. They establish priority and urgency, not proof of damage and certainly not proof of insurance coverage. Use storm data to decide which aging roofs to inspect first, then let an actual on-roof inspection determine condition. Train reps never to promise damage or coverage based on a storm flag alone.

How do I rank streets so my reps stop wasting time?

Score each address on roof-age range, storm exposure, absence of a recent reroof permit, and any visible condition flags, producing a 0 to 100 number. Then roll those scores up to the block by counting how many addresses clear a threshold, and weight for density so reps work tight clusters rather than spread-out lots. Send crews to the highest-scoring, densest routes first, and hold them to the route instead of letting them knock by convenience.

What is the difference between this and buying roofing leads?

Buying leads means paying for a homeowner's contact information that a vendor often sells to several contractors at once. Targeting aging roofs means ranking the doors on a list you already control, your service area, your CRM, or your past customers, using roof-age and storm signals so your own canvassers spend their hours on the houses most likely to be due. You own the relationship from the first knock, and you are not competing with three other buyers for the same lead.

What can I legally say to a homeowner about insurance?

You can inspect the roof, document what you observe with photographs, and write an accurate repair estimate for your scope using standard estimating practice, then hand that documentation to the homeowner. You may state facts about what you found. You may not, for a fee, negotiate or handle their claim, interpret their policy or what is covered, promise a payout or approval, promise to waive or absorb the deductible, or advertise a free roof. The homeowner files any claim and the insurer decides coverage. Crossing those lines can amount to unlicensed public adjusting or deceptive practices under state law.

How do reroof permits help my targeting?

A reroof permit is the cleanest signal that a roof was recently replaced. If a 2002-built house pulled a reroof permit in 2017, that roof is young and you can remove it from your list. Permit coverage is uneven and some replacements happen without permits, so treat it as a subtraction filter rather than proof: use it to drop confirmed recent reroofs, and treat a no-permit house as a stronger candidate for still wearing an aging roof.

How fast do I need to move after a storm?

Days, not weeks. The homeowner's attention is highest right after a significant event and falls off quickly, and out-of-town crews flood active footprints fast. Build a runbook in advance: who pulls the impacted area, who cross-references it against your aging-roof map, who builds the routes, and who knocks. Companies that can dispatch reps to ranked routes within a day or two of an event consistently reach roofs before the crews that start organizing only after the storm hits.

Does RoofPredict guarantee a roof needs replacement?

No. It estimates a roof-age range from aerial imagery and models storm exposure per roof, then enriches your address list so you can rank doors and routes. It deals in ranges and odds, not certainties, and it does not promise that any roof is damaged or that a claim will be approved. The inspection, the documentation, and the homeowner conversation are still yours. It removes the guesswork about where to point your reps; it does not replace the field work that earns the job.

What numbers should I track to know if my targeting works?

Track doors attempted per rep-hour, conversation rate, inspection rate, and inspection-to-signed rate by neighborhood, and most importantly the correlation between your address scores and actual closes. If high-scored addresses close at a meaningfully higher rate than low-scored ones, your model is working and you should push more reps toward high scores. If there is no spread, your weights or data need fixing before you invest in anything else. The score is a hypothesis you test against closed jobs every week.