How to Find the Right Doors to Knock for Roofing

Knocking doors is the cheapest pipeline a roofing company can build. It is also the easiest one to waste. Two crews can knock the same number of doors on the same Saturday and walk away with wildly different results, and nine times out of ten the difference is not the pitch, the polo, or the weather. It is which doors they chose to walk up to in the first place.

The roofers who treat canvassing like a numbers game eventually burn out their reps and their reputation. The ones who treat it like a targeting problem — where every door is a small bet, and you want the odds stacked before you ring the bell — build a machine that prints inspections year after year, storm or no storm. This is a working field manual for the second kind. It covers how to read a roof from the curb, how to think about storm damage house by house instead of by ZIP code, how to score and sequence neighborhoods, how to build routes a green rep can run, and how to keep score so the system gets smarter every week.

No fluff, no inspirational nonsense about hustle. Just the decisions that separate a productive door day from a long walk.

Why most door-knocking dies on the wrong street

Let's start with the math, because the math is brutal and most reps never run it.

Say a rep knocks 80 doors in a four-hour block. On a typical residential street, roughly a third of those are not-homes — no answer, renter who can't decide, language barrier, dog, gate. So you're really talking to about 50 people. Of those, the overwhelming majority have a roof that does not need replacing right now. A composition shingle roof on a tract home usually carries a manufacturer rating of 25 to 30 years and realistically lives somewhere in the high teens to mid-twenties depending on slope, ventilation, sun exposure, and whether it has taken a real storm.

So if you knock a subdivision that was built six years ago, the honest replacement rate on that street is close to zero. You can have the best opener in the county and you will still come home with a sunburn and one "maybe call my husband." The pitch was never the problem. The street was the problem.

Flip it. Knock a neighborhood where the homes are 18 to 22 years old, on their original roofs, and one good hailstorm rolled through two springs ago. Now a meaningful share of those 50 conversations are with homeowners who genuinely have an aging or damaged roof — they just haven't thought about it because it still looks fine from the driveway. Same rep, same script, same hours, completely different day.

The lesson every veteran canvasser learns the hard way: door knocking rewards selection far more than it rewards volume. Doubling your door count gives you linear returns and burns your reps out. Doubling the quality of your door selection compounds, because better doors mean more inspections, more inspections mean more contracts, more contracts mean reps who make money and stay, and reps who stay get good enough to close the harder doors. Selection is the lever. Everything in the rest of this manual is about pulling it.

What "the right door" actually means

A door is worth knocking when at least one of these is true, and ideally two or three:

• The roof is old enough that replacement is plausible on its own (original roof, 15-plus years of age depending on material and climate).
• The roof has taken a storm — hail or high wind — recent enough that damage is real and a claim window may still be open.
• The homeowner profile suggests they can act: owner-occupied, has owned long enough to have equity, not a rental, not a brand-new build still under builder warranty.
• There is a visible curb signal: granule loss, missing or lifted shingles, a tarp, a competitor's yard sign two doors down, an obvious patch job.

A door is a waste when the roof is young, the home is a rental with an absentee owner, the house was recently re-roofed (you'll learn to spot this), or the neighborhood took no storm and is too new to age out. The whole game is sorting the first kind from the second kind before you spend your legs on it.

Reading a roof from the street: the curbside inspection

You will not climb every roof, and you should not. The fastest canvassers qualify a house in the four seconds it takes to walk from the sidewalk to the porch. Train your eyes and your reps' eyes to read these signals.

Age signals you can see without a ladder

Granule loss and color fade. Asphalt shingles are protected by mineral granules. As a roof ages and bakes, those granules shed. From the street, an aging roof looks duller, lighter, and blotchy compared to the rich uniform color of a newer roof. Look at the area below valleys and around downspout outlets — washed-out, bald-looking patches mean granules have been running off into the gutters for years.

Curling, cupping, and clawing. Old shingles lose their flexibility and start to deform. Edges that lift up are cupping; edges that turn down while the center rises is clawing. From the ground this reads as a rough, shadowed, "shaggy" texture across the field of the roof instead of a clean flat plane. A roof that looks like it has a five-o'clock shadow is usually well past 15 years.

Sagging or wavy lines. A ridgeline that dips, or a roof plane that ripples, often signals decking or structural issues underneath — and a homeowner who already half-knows something is wrong.

The neighbors' roofs. Tract subdivisions were built and roofed at the same time, so they age in cohorts. If three houses on a street have already been re-roofed (you'll see the bright new shingle against the faded ones), the rest of that street is on borrowed time. A cluster of new roofs is one of the strongest "this street is due" signals there is.

Multiple layers. From the rake edge (the angled side edge of the roof) you can sometimes see a thick shingle stack — a sign of a roof-over, which is at end of life and often won't pass a fresh layover, meaning a full tear-off is the only option.

Storm signals you can see without a ladder

Soft metal as a tell. Hail leaves dents in soft metals long before it leaves obvious marks on shingles. From the ground or the driveway, glance at metal roof vents, valley flashing, gutters, downspouts, the garage door, gutter aprons, and any AC condenser fins. Random dents in these surfaces are some of the clearest evidence a hailstorm actually hit this specific house at meaningful size.

Wind damage. Look for missing shingles, lifted or flapping tabs, a crease line across shingles where they bent back and didn't fully reseal, and debris in the yard. Wind damage clusters on the windward-facing slopes and along edges, ridges, and rakes.

Fresh disturbance. Granules in concentrated piles at downspout splash blocks after a recent storm, a tarp on part of the roof, a damaged fence or dented mailbox — these say the storm here was real and recent.

What gets pros in trouble: the false positives

Newer reps over-flag. A few traps to teach:

• Mechanical and blistering marks look like hail but aren't. Blistering (popped bubbles from trapped moisture/asphalt), foot traffic scuffs, and manufacturing flaws all leave round marks. Real hail hits are random in pattern, show a bruise you can feel (soft spot) on the mat, and knock granules loose exposing fresh asphalt. Directional, lined-up, or evenly-spaced marks are usually not hail.
• A new-looking roof might be a recent re-roof, not a young house. Check the house age against the roof. If a 25-year-old home has a crisp uniform roof, someone already replaced it — skip it.
• Algae streaking (the black stains) is cosmetic, not age or damage. Plenty of homeowners think those streaks mean a bad roof; they don't. Don't build a pitch on them.

The honest curbside read is a filter, not a diagnosis. Its job is to decide whether this door is worth a knock and whether a ladder inspection is worth scheduling — not to declare damage from the sidewalk.

Match the signal to the material

Different roof materials age and fail differently, so the same curb signal means different things depending on what you're looking at. A quick reference for the materials a residential crew actually meets:

The practical takeaway: most of your door-knocking volume in a typical American subdivision is asphalt, and asphalt is where the 15-to-25-year age window does the most work for targeting. When you hit a tile or metal street, slow down — the roof can look rough and still have years left, and a re-roof conversation there is usually about underlayment or specific failures, not whole-roof age.

Picking neighborhoods before you pick doors

Door selection starts a level up, at the neighborhood. Walking a great street is ten times more productive than great pitching on a bad street, so spend real planning time here. You're looking for the overlap of three things: roofs old enough to be due, storms that actually hit, and homeowners who can act.

Age the housing stock

The single most useful planning fact about a neighborhood is when it was built, because original roofs age with the house. Public sources get you most of the way:

• County assessor / property appraiser records list year built for nearly every parcel, usually free and searchable online. This is your bedrock data.
• Census data (the American Community Survey publishes "year structure built" by tract) lets you spot whole tracts that hit the 18-to-25-year window where original roofs start failing.
• Permit records. Many jurisdictions publish building permits, including reroof permits. A neighborhood with few recent reroof permits and 20-year-old housing stock is a field of original roofs. A neighborhood already saturated with reroof permits has been picked over.

The sweet spot for an original asphalt roof is roughly 15 to 25 years old, shifted by climate: harsh sun and big temperature swings age roofs faster, milder coastal climates slower. A subdivision built in the late 1990s through mid-2000s, never restormed, is a classic age-out target right now.

One caution worth stating plainly: year built is not roof age. It is the best free proxy, but a re-roof resets the clock and is invisible in "year built." That gap — between when the house went up and when the roof was actually last replaced — is exactly where a lot of wasted knocking comes from, and it's the gap good data tools exist to close.

Map where storms actually hit

For storm work, the planning question is not "did it storm this spring" but "where, exactly, did damaging hail and wind touch down, and how big." Free public data gets you the regional picture:

• NOAA's Storm Prediction Center publishes daily storm reports and archives of hail and wind reports, including estimated hail size.
• The National Weather Service issues and archives warnings and local storm reports by office.
• NOAA's Storm Events Database is a searchable historical record of severe weather by county and date, including hail size and wind estimates.

The limit of all of these is resolution. A storm report is a point or a polygon over a county or a swath. Hail does not fall evenly across a swath — it falls in streaks and cores, and two streets a half-mile apart can have completely different damage. A regional hail map tells you the storm passed through the area. It does not tell you which roofs it actually wore out. That distinction is the entire difference between knocking a whole ZIP code and knocking the specific blocks that took the core.

Read the homeowner profile

Age and storm get you to a roof that needs work. The homeowner profile gets you to a roof that can be sold.

• Owner-occupied beats rentals. Renters can't authorize work and absentee landlords are slow. Assessor records and mailing-address-vs-situs-address mismatches flag rentals.
• Tenure and equity. A homeowner who has owned for several years generally has the equity and the standing to move on a roof. Brand-new buyers and underwater owners are harder.
• Median home value and income at the tract level (Census again) tell you whether a neighborhood can carry a re-roof or a deductible without flinching. You're not looking for the richest blocks — you're looking for solidly middle, owner-occupied, well-kept homes where a roof is a normal expense, not a crisis.
• HOA presence can cut both ways: HOAs sometimes mandate roof upkeep (good — forced demand) but can also slow approvals and restrict materials.

Stack the three layers — old enough, hit hard enough, owners who can act — and you've gone from a city of a hundred thousand doors to a short list of blocks worth your crew's legs.

A 7-step system to choose which doors to knock

Here is the workflow, start to finish, the way a sharp sales manager runs it for a crew.

Step 1 — Define the catchment

Draw the area you're willing to service profitably. Drive time matters: a crew that travels 40 minutes each way loses an inspection a day to windshield time. Most production roofers draw a 20-to-30-minute radius around the yard or the current job and work concentrically out from there.

Step 2 — Pull the age layer

For that catchment, pull year-built data from the assessor and/or Census tract data, and any reroof permit history you can get. Shade your map: neighborhoods 15-plus years old in one color, the prime 18-to-25 window in a hotter color, new builds grayed out (skip). This alone eliminates most of the doors that would have wasted you.

Step 3 — Overlay the storm layer

Pull recent hail and wind reports for the catchment from SPC / NWS / Storm Events. Mark the swaths and, where you can, the estimated hail size. Hail under about one inch rarely does functional shingle damage; one inch and up is where claims get real, and 1.5-inch-plus does serious damage. Where storm swaths overlap your hot age neighborhoods, you've found your A-list.

Step 4 — Filter by homeowner reality

Knock out (or deprioritize) the rentals, the brand-new buyers, the blocks that obviously can't carry the work. You want owner-occupied, settled, maintainable homes.

Step 5 — Rank the streets

Now score and stack-rank the streets in your remaining set. A simple, honest scoring rubric a manager can keep on one screen:

Rank streets by score. Your reps work top-down, A-streets first.

Step 6 — Pre-qualify with a windshield drive

Before you deploy a crew, drive the top streets yourself (or send a lead). Confirm the curb signals from the section above: granule loss, neighbor reroofs, soft-metal dents, tarps. A 20-minute drive-by upgrades or downgrades streets that looked good on paper. This is the highest-leverage 20 minutes in the whole process.

Step 7 — Cut the route and assign it

Turn the ranked, drive-verified streets into a sequenced walking route (more on routes below) and hand each rep a list they can run without thinking. The thinking — the selection — already happened. The rep just executes.

A worked example

Make it concrete. A two-rep crew, suburban metro, spring after a hail event.

• Catchment: 25-minute radius around the yard, about 40,000 homes.
• Age layer: cut to neighborhoods built 1998-2007, never mass-reroofed. Down to ~9,000 homes.
• Storm layer: SPC and NWS reports show a hail swath with 1.75-inch reports cutting across the north third of that area. Overlap with the age-hot neighborhoods: ~2,600 homes.
• Homeowner filter: drop two large rental-heavy tracts and a brand-new infill. ~2,100 homes.
• Rank + drive: top eight streets confirmed by drive-by (neighbor reroofs, dented gutters, one tarp). ~700 homes across those streets.
• Route: split into two-day walking routes per rep.

That crew now spends its week on 700 genuinely promising doors instead of randomly sampling 40,000. The inspection rate on a list like that can be several times what the same reps would get cold-walking an unscreened subdivision — not because they got better, but because the doors got better.

Where roof-due data and per-roof storm modeling fit

Everything above can be done by hand with public data, a spreadsheet, and a lot of evenings. Plenty of good roofers built their book exactly that way. The catch is the two blind spots no free source closes well, and they're the two that cost you the most doors.

Blind spot one: year built is not roof age. Re-roofs are invisible in assessor data unless a permit was pulled and published — and a huge share never are. So you walk up to a 22-year-old house, deliver your aging-roof pitch, and the homeowner tells you they replaced it four years ago. You burned the door and a little credibility. Across a season, those silent re-roofs are a real tax on your hours.

Blind spot two: a hail map is not per-roof damage. Storm reports are swaths and points. Hail falls in streaks; within a single "affected" street, one side may be hammered and the other barely touched. Knocking a whole storm polygon means knocking a lot of roofs the storm didn't actually wear out.

This is the gap RoofPredict is built to close. It reads aerial imagery to estimate roof age as a range per address — narrowing "this house is old" down to which specific houses are likely on an original, aging roof, including catching re-roofs that reset the clock that year-built data can't see. And rather than handing you a regional hail map, it models the storm physics — hail trajectory and wind — on each individual roof, so you get a per-house read on which roofs a storm likely wore out, paired with that age range. A hail map shows you where it hailed; modeling each roof shows which ones it actually battered.

In practice that means your Step 2 and Step 3 above — the age layer and the storm layer — come pre-built and per-address, and your street ranking in Step 5 starts from a ranked list of which roofs are most likely due rather than a manual scoring exercise. For a crew, a per-home read also becomes a back-pocket talking point: a green rep can knock with a specific reason this house, sound like a veteran, and skip the new roofs without climbing anything.

Now the honest limits, because a tool that oversells gets you in trouble at the door:

• Roof age is a range, not a birth certificate. The output is an estimate window (for example, an 18-to-22-year range), not an exact install date. Treat it as "this roof is very likely due" — strong for prioritizing doors, not a fact to assert to a homeowner as gospel.
• A storm model is odds, not proof. Modeling says this roof was likely worn by the storm. It is not a substitute for getting on the roof and documenting actual conditions. The only thing that proves damage is an inspection.
• It is not a lead service and it does not knock for you. It ranks doors and routes; your crew still has to walk up, build rapport, inspect, and earn the job. It sharpens outbound you already do — it does not replace it.

Used that way — as a sharper version of the manual targeting in this manual, not a magic button — it does the one thing that moves door-knocking results most: it gets your reps in front of more roofs that are actually due, and keeps them off the ones that aren't.

Turning a target list into a knockable route

Great targeting falls apart if the route is a mess. Reps waste an astonishing amount of a day backtracking, crossing busy streets, and skipping houses. Tight routing is its own multiplier.

Principles of a good walking route

• Walk one side of the street, then the other. Crossing back and forth doubles your steps and halves your doors. Run all the odd numbers down, all the evens back.
• Keep the rep on foot, the car staged. Driving between individual houses kills density. Park, work a tight cluster of blocks fully, then move the car.
• Sequence by density, not by score alone. An A-street two miles from the cluster can be worth less than a B-street in the middle of it, because windshield time is dead time. Cluster first, then prioritize within the cluster.
• Size the route to the day. A productive rep covers roughly 60 to 100 doors in a four-hour block depending on density and conversation length. Hand them a route a little longer than that so they never run out of doors, but not so long it sprawls.
• Build in callback geography. Group your "not home" and "come back" houses so the second pass is one efficient loop, not a scavenger hunt across town.

Logging the door so the route gets smarter

Every door is data. Reps should log a disposition on the spot — not at the end of the day from memory. A clean disposition set:

That "new roof" disposition is gold over time: it tells you which streets your age data was wrong about, so you stop sending reps there. The dispositions feed the scorecard in the next section.

Timing: the hours and seasons that actually convert

The right door at the wrong time is a not-home. Timing is part of targeting.

Time of day. The reliable residential windows are late afternoon into early evening on weekdays — roughly 4:30 to dusk — when people are home from work, and mid-morning to early afternoon on Saturdays. Avoid dinner hour resentment by reading the house: lights on, cars in the drive, kids' toys out means someone's home and settled. Knock too early on a weekend and you're the jerk who woke the house.

Day of week. Weekday early evenings and Saturday daytime are the workhorses. Sundays are regionally touchy — in many markets Sunday knocking reads as disrespectful, so know your area. Friday evenings convert poorly; people are out or winding down.

Season and storm timing. For age-out work, spring and summer are prime — long daylight, homeowners thinking about the house. For storm work, the window matters enormously: you want to be canvassing the affected blocks within days to a few weeks of the event, while the damage is fresh, undeniable, and the homeowner hasn't already signed with someone else. Wait too long and the out-of-town crews have already worked it; move too slow on your own backyard and you lose your home-field advantage. Most carriers also require claims to be filed within a defined window after the loss, so prompt, accurate documentation matters to the homeowner regardless of who they hire.

Light and safety. Stop knocking when you can't be clearly seen at the door. After dark, you read as a threat, not a contractor, and you'll get doors slammed and police called. Daylight is a conversion tool and a safety tool.

Staying legal and welcome: permits, lists, and the laws that bite

Nothing kills a campaign faster than a city ordinance violation or a contractor-fraud complaint. Build compliance into your targeting.

• Solicitation permits and "no-knock" rules. Many municipalities require a solicitation or peddler's permit to canvass, and many neighborhoods (especially HOAs) post no-soliciting rules. Some cities maintain do-not-knock registries. Check the local ordinance for every city you work, carry the permit, and honor posted and registered no-soliciting requests. Reps should always carry ID and visible branding.
• Honor do-not-knock and do-not-contact on the spot. If someone asks you not to come back, log it as do-not-contact and never return. It's the law in places and just basic decency everywhere.
• Storm-chasing and AOB laws. Numerous states have passed contractor laws specifically aimed at post-storm door-knocking abuses — rules on contract cancellation windows, prohibitions on a contractor advancing or rebating a homeowner's insurance deductible (treated as fraud in many states), and limits on assignment-of-benefits. Know your state's roofing and storm-restoration statutes before you knock a disaster area.
• Stay in your lane on insurance. This is the one that quietly sinks roofers. Your job at the door and on the roof is to document conditions and provide an estimate. You can hand the homeowner clear evidence — roof age, photos of damage, measurements — that supports their conversation with their carrier. What you cannot do is act as the homeowner's insurance representative: negotiating, adjusting, or handling the claim on their behalf can cross into unlicensed public adjusting, which is illegal in most states and has been enforced against roofers who merely marketed themselves as claims specialists. The clean line: the homeowner owns the claim, the insurer decides coverage, and you document and build. Say nothing about guaranteeing approval and nothing about absorbing or covering a deductible.

None of this is legal advice — verify the specific rules for your city and state, and get counsel sign-off on any claims-related script. But baking these constraints into how you choose and approach doors keeps the whole operation out of the ditch.

At the door: a qualification that respects everyone's time

This is a targeting manual, not a pitch manual, but the first thirty seconds at the door are really a final qualification step — you're confirming whether this is a door worth your next twenty minutes. A quick, honest framework:

1. Open with the specific reason you're here. Reference the real thing — the storm that came through, the age of roofs on this street, the neighbor's new roof. Specificity earns the next sentence. "We've been on this street because the hail that hit on the 14th was big enough to bruise shingles, and I noticed some dings on a few gutters" beats any generic script.
2. Confirm ownership and roof status fast. Are they the owner? Do they know how old the roof is? Has it been replaced since they owned it? One "we replaced it three years ago" and you politely thank them and move on — that's a win, because you just saved twenty minutes for a real door.
3. Offer the inspection, not the sale. The ask at the door is a free roof inspection, not a contract. Low commitment, high value, and it's where you actually find out what's going on.
4. Document, don't diagnose claims. If you get up there and find damage, you document it — photos, measurements, the roof's condition — and walk the homeowner through what you see. You let them decide whether to call their carrier. You don't promise an approval and you don't talk deductibles.

The reps who qualify out fast at the door — who are genuinely happy to find a new roof and move on — are the ones who hit more real opportunities per day, because they don't waste twenty minutes being polite to a door that was never going to buy.

Keeping score: the metrics that make targeting smarter every week

Targeting is a loop, not a one-time setup. Every week of dispositions tells you which selection bets paid off, so you sharpen the next week. Track these.

The core canvassing funnel

The one to obsess over for targeting is inspection set rate and doors per contract. If your selection is good, fewer doors produce each contract. When you compare two neighborhoods, the one with the lower doors-per-contract is the one your data and instincts read correctly — go find more streets like it.

Closing the loop on your data

The "new roof" and "renter" dispositions are your data-quality report card. If a street you ranked A throws back a pile of "we just re-roofed" dispositions, your age signal was wrong there — likely a wave of silent re-roofs — and you adjust. This is exactly where a per-address roof-age estimate earns its keep over raw year-built data: it catches many of those re-roofs before you send a rep, instead of after.

Review the funnel weekly with the crew. Celebrate low doors-per-contract, not high door counts. The rep who knocks 60 great doors and sets four inspections beat the rep who knocked 120 random doors and set one — reward the first one, or you'll train your team to optimize the wrong number.

The doors you already own: working your own list

The best doors a roofing company has are often not on a map at all — they're sitting in the company's own records, and almost nobody works them systematically. Before you spend a dime walking cold streets, mine the demand you've already paid to create.

Old estimates that never closed. Every roofer has a graveyard of inspections and proposals that didn't convert — homeowner wasn't ready, timing was off, they got three bids and stalled. A roof you inspected three years ago and flagged as marginal is now three years older, and a storm may have hit it since. These addresses are pre-qualified: you already know the roof, you already have a relationship, and you have a legitimate reason to knock again. Pull every proposal that didn't close from the last three to five years, cross-reference it against recent storm swaths, and you have a warm route before you touch a cold neighborhood.

Past customers' neighbors. When you complete a job, the surrounding ten houses are now an A-list: same building era, same storm exposure, same age cohort, and you have a fresh sign, a recent truck on the street, and a real reference two doors down. "We just replaced the Hendersons' roof around the corner" is one of the strongest openers there is, and it's free. Build a standing rule that crews knock the immediate neighbors of every completed job before the dumpster leaves.

Aging warranties and old installs. If you've been in business long enough, roofs you installed 15 or more years ago — or roofs you've serviced or inspected — are aging into replacement now. Those homeowners already trust you. A maintenance-check or end-of-warranty touch is a clean, non-pushy reason to get back on the roof.

The reason this matters for door selection: these are doors where the targeting work is already done and the relationship already exists, so the conversion rate dwarfs cold streets. Run your own list first, then fill the rest of the route with the cold-targeting system above. A per-address roof-age and storm read makes this even sharper — you can sort your own old list by which of those roofs is now most likely due, instead of re-knocking all of them blind.

Common mistakes that quietly bleed a door program

The failure modes are predictable. Watch for these.

• Knocking new construction. It feels productive — friendly homeowners, easy conversations — and it produces nothing. Gray those neighborhoods out and don't let reps drift into them because they're "nicer to walk."
• Treating a storm polygon as a target list. The storm "hit the area" is not the same as "hit this roof." Working an entire swath uniformly wastes the half of it that took glancing damage. Tighten to the cores and the streets that show curb signals.
• Chasing the same picked-over blocks everyone else is on. After a notable storm, the obvious neighborhoods get swarmed by every crew in three states. The edge is often in the slightly-less-obvious adjacent blocks the swarm skipped, and in age-out work that doesn't depend on a storm at all.
• Volume worship. Pushing reps to maximize door counts trains them to knock fast and shallow, skip qualification, and burn good streets. Reward conversion, not count.
• No dispositions, no memory. A crew that doesn't log doors re-knocks the same not-homes, forgets the come-backs, and never learns which streets were good. The data is the asset.
• Letting reps freelance the route. When reps pick their own doors house-to-house, they cherry-pick the easy-looking ones, skip the awkward ones, and lose density. Hand them a sequenced route and a disposition app.
• Over-promising on insurance. A rep who tells a homeowner "we'll get this approved" or "we'll cover your deductible" exposes the company to fraud and unlicensed-adjusting liability. Train the legal line hard.
• Ignoring the come-back pile. A huge share of contracts come from second and third touches. Reps who only count first knocks leave money on streets they already paid to walk.

A simple weekly operating rhythm

Put it all together into a cadence a sales manager can actually run.

Sunday or Monday (plan): Pull or refresh the age and storm layers for the week's catchment. Rank the streets. Do the windshield drive on the top candidates. Cut routes and assign them.

Tuesday-Saturday (execute): Reps run assigned routes in the right time windows, log every door, set inspections, and feed the come-back and not-home piles into a callback loop.

End of each day (capture): Reps sync dispositions. Manager eyeballs inspection-set rate by street.

Friday or Monday (review): Walk the funnel with the crew. Find the streets with the lowest doors-per-contract and queue more like them. Flag streets that threw back re-roofs or rentals and fix the targeting. Adjust the week's plan.

Do this for a season and the compounding shows up: your street list gets sharper, your reps spend their legs on roofs that are actually due, your doors-per-contract drops, and the reps who are suddenly making money stick around to get even better. That's the whole game — not knocking more doors, but knocking the right ones, and getting a little smarter about which ones those are every single week.

The short version

If you remember nothing else: door knocking is a selection problem wearing a stamina costume. Pick neighborhoods where roofs are old enough to be due, where storms actually hit hard enough to matter, and where owners can act. Read roofs from the curb to filter further. Rank and route so reps execute instead of wander. Qualify fast and honestly at the door, document instead of diagnosing claims, and stay on the right side of local and insurance law. Then keep score, close the loop, and let the data sharpen your next list. Do that, and you stop renting your pipeline from the weather and the lead sites and start owning the streets you work — storm or no storm.

FAQ

How do I know which neighborhoods to knock for roofing?

Layer three things over a map of your service area. First, age: pull year-built data from the county assessor or Census tract data and target neighborhoods roughly 15 to 25 years old on original roofs. Second, storms: overlay recent hail and wind reports from NOAA's Storm Prediction Center and the National Weather Service to find where damage actually fell. Third, homeowner reality: favor owner-occupied, settled blocks over rentals and brand-new builds. Where old roofs, real storm impact, and able owners overlap is where you knock. Then drive the top streets to confirm curb signals before deploying a crew.

How can I tell a roof's age from the street without getting on a ladder?

Look for granule loss and color fade (a dull, blotchy, washed-out surface, especially below valleys and downspouts), curling or clawing shingles that give the roof a rough shaggy texture, and sagging ridgelines. The strongest single tell is the neighbors: tract homes were roofed in cohorts, so a cluster of bright new roofs against faded ones means the rest of that street is aging out. Remember this is an estimate to decide whether a door is worth knocking, not an exact age, and a new-looking roof on an old house is usually a recent re-roof you should skip.

What's the difference between a hail map and per-roof storm data for targeting doors?

A hail map or storm report shows where a storm passed as a point or swath over a county. Hail falls in streaks and cores, so within one 'affected' street one side can be hammered and the other barely touched. A regional map tells you the storm came through the area; it does not tell you which specific roofs it wore out. Tools that model storm physics on each individual roof give you a per-address read on which houses were likely battered, so you knock the streets that took the core instead of the whole polygon. Either way, only a roof inspection proves actual damage.

Why does 'year built' fail as a way to find old roofs?

Year built is the best free proxy for roof age, but a re-roof resets the clock and almost never shows up in assessor data unless a permit was pulled and published. So you can walk up to a 22-year-old house on its third roof and deliver an aging-roof pitch to someone who replaced it four years ago. Those silent re-roofs are a real tax on your hours. Per-address roof-age estimates from aerial imagery exist specifically to catch re-roofs that year-built data can't see, narrowing 'this house is old' down to which houses are actually on an aging original roof.

What time of day and which days are best for knocking roofing doors?

The reliable residential windows are weekday late afternoon into early evening, roughly 4:30 to dusk when people are home, and mid-morning to early afternoon on Saturdays. Read the house for lights, cars, and signs of life before you knock. Avoid dinner hour, early weekend mornings, and Friday evenings. Sundays are regionally sensitive, so know your market. Always stop when it gets dark enough that you can't be clearly seen at the door, both for conversion and for safety. For storm work, canvass affected blocks within days to a few weeks of the event while damage is fresh.

How many doors should a rep knock, and is more always better?

A productive rep covers roughly 60 to 100 doors in a four-hour block depending on density, but raw door count is the wrong target. Selection matters far more than volume: a rep who knocks 60 well-chosen doors and sets four inspections beats a rep who knocks 120 random doors and sets one. The metric to obsess over is doors-per-contract, which good targeting drives down. Rewarding high door counts trains reps to knock fast and shallow and burn good streets, so reward conversion instead.

What are the biggest mistakes roofers make when choosing doors to knock?

Knocking new construction because it's pleasant but produces nothing; treating an entire storm swath as a target list instead of tightening to the cores; piling onto the same picked-over blocks every other crew is swarming after a storm; worshipping door volume over conversion; failing to log dispositions so the crew re-knocks the same not-homes and never learns which streets worked; letting reps freelance the route and cherry-pick easy doors; ignoring the come-back pile where many contracts actually close; and over-promising on insurance, which creates real legal exposure.

What can I legally say about insurance when I knock a storm-damaged door?

You can document the roof's condition and provide an estimate, and you can hand the homeowner clear evidence such as roof age, damage photos, and measurements that supports their conversation with their insurer. What you cannot do is act as the homeowner's insurance representative by negotiating, adjusting, or handling the claim, which can cross into unlicensed public adjusting that is illegal in most states and has been enforced against roofers. Never guarantee an approval and never offer to absorb or cover a deductible, which is treated as fraud in many states. The homeowner owns the claim, the insurer decides coverage, and you document and build. This is not legal advice; verify your state's rules and get counsel sign-off on any claims script.

Do I need a permit to knock doors for roofing?

Often yes. Many municipalities require a solicitation or peddler's permit to canvass, and many neighborhoods and HOAs post no-soliciting rules. Some cities maintain do-not-knock registries you must honor. Check the local ordinance for every city you work, carry the permit and visible ID, and honor posted and registered no-soliciting requests. If a homeowner asks you not to return, log it as do-not-contact and never come back. Several states also have storm-restoration laws covering contract cancellation windows and deductible handling, so know those before working a disaster area.

Does software like RoofPredict replace door knocking?

No. RoofPredict is not a lead service and it does not knock for you. It sharpens the outbound you already do by estimating roof age as a range per address and modeling storm physics on each roof, so it ranks which doors are most likely due and helps you build routes that skip the new roofs. Your crew still has to walk up, build rapport, inspect, and earn the job. Treat it as a sharper version of manual targeting, not a magic button: roof age is an estimate range, not an install date, and a storm model is odds that point you to likely-worn roofs, not proof of damage. Only an inspection proves that.