Door-knocking gets dismissed as the grunt work of roofing sales, the thing you do when you can’t afford ads. That’s backwards. Standing on someone’s porch is still one of the highest-converting things a roofing company can do, because you’re in front of the exact house, at the exact address, with a face and a business card. No shared lead form comes close. The roofers who win storms now pair it with the phone: they call the same streets they knock, so no homeowner slips through just because they were out when the rep came by.
The knocking works. Calling the same list at the same time works even better, and the waste in both comes down to which doors and numbers you pick. A rep who walks a random subdivision burns the day on new roofs, empty houses, and neighbors who signed with the other guy last Tuesday. This is a playbook, drawn from running these routes inside a roofing company, for picking the streets likely to buy and working them by phone and at the door: how to choose streets, cut territories, dial alongside the knock, and measure whether it’s working.
Why random canvassing burns the day
Walk a street cold and here’s what eats your hours:
- New and recently replaced roofs. A five-year-old roof isn’t a customer for years, and on a normal block that’s a big share of the doors.
- Empty homes and no-answers. Knock at the wrong time and half the street isn’t home, so you log steps and get nothing to show a manager.
- Already-signed neighbors. After a storm the fastest company works a neighborhood in days. Show up late and you’re knocking doors with a contract already taped to the fridge.
- Wrong fit. Rentals, homes that can’t finance the job, HOAs that block your sign: all real doors, all dead ends.
The door-to-door approach is fine. The inputs are what’s broken. Same rep, same script, same hours, pointed at the right 40 houses instead of a random 200, is a completely different day.
How to pick high-probability streets
A high-probability street is one where a lot of homeowners have a real reason to say yes right now. Four signals find them, and the more that overlap, the better the block.
1. Storm and hail footprint
The single strongest signal in roofing. When hail or wind moves through, a whole neighborhood needs a roof at once and urgency is already on the homeowner’s mind. The trick is knowing the actual footprint, not the rumor of “somewhere off the highway it got bad.” Storm paths are narrow. One subdivision gets hammered while the one a mile over is untouched. Work the streets genuinely hit and your knock lands on someone already thinking about their roof. Our live hail maps by city are built for exactly this.
2. Roof age
Away from storm season, age is your best proxy for need. Neighborhoods built 18–25 years ago are hitting the end of a typical asphalt shingle’s life all at once. The whole subdivision aged together, and most homeowners don’t know they’re due yet. You’re getting there before the leak does, not manufacturing urgency out of nothing. (This is your own read of the local building boom: pull up when the subdivision went up and knock it a couple decades later.)
3. Income and ownership fit
A roof is a five-figure decision. Streets where homeowners can actually finance the job convert better and stall less in the “let me think about it” stage. Owner-occupied beats rentals almost every time, since renters can’t authorize the work. The point is simple: keep your best rep’s afternoon off doors that structurally can’t close, rich ZIP or not.
4. Proximity to a job you just landed
The strongest single street you can knock is the one where your crew is already working. Neighbors watch the dumpster, they trust you’re real, and “we’re already doing your neighbor’s roof at number 14” is the most honest opener in sales. Any time you sign a job, canvass the surrounding blocks that week, before the crew packs up and the social proof leaves with them.
The magic is in the overlap. A block inside a recent hail footprint, built 20 years ago, owner-occupied, and two doors from a job you just signed is about as close to a sure thing as canvassing gets. Send your best rep first.
Build tight, non-overlapping territories
Once you know the good streets, the operational job is making sure two reps never knock the same house and no promising block gets skipped.
- Assign by boundary, not by vibe. Draw each rep a clear area: streets, not “the north side.” Overlap means two people annoy the same homeowner; gaps mean a good block never gets worked.
- Size the route to the day. If you’ve pre-filtered to high-probability houses, a tight route of 40–60 good doors beats a sprawling 200-door slog.
- Rotate the no-answers. A “not home” isn’t a “no.” Track it and swing back rather than burning the address forever.
- Keep a record of every door. Signed, not interested, not home, new roof: logging outcomes lets you re-work an area intelligently instead of starting cold.
Dial the street while you knock it
Here’s the upgrade most crews miss, and the one that has moved results the most for the teams we work with: don’t just knock a route, call it at the same time. Once you have the highly accurate name and phone for every house on a block, the phone becomes a second rep working the same street in parallel.
- Call ahead of the knock. A quick “we’ll have a rep on your street Thursday” warms the door, and a live call books inspections with homeowners who would never answer a cold knock.
- Reach the doors feet can’t. Half a block isn’t home when you knock. A call catches them at work, on a break, or that evening, so no-answers stop being dead weight.
- Cover more homes, faster. A caller can work a hundred numbers in the time a rep walks forty doors. Run both at once and you touch the whole neighborhood in a fraction of the time, while the storm is still fresh.
- Use the name, then follow up. “Hi, is this the Hernandez residence?” beats “hey there,” and the maybes get a same-day text while the conversation is fresh.
This is the model that wins storms now: reps knock the high-probability streets while an inside caller works the phone list for the same blocks. Two channels, one targeted route, the whole neighborhood reached before the competition finishes lacing up. A mailer to the same addresses a few days later reinforces both.
A door-knocking script that respects their time
Keep it short, honest, and specific to the street. A frame that works:
- Name the reason you’re there. “We’re doing a roof two doors down at number 14. After last month’s hail a lot of homes on this block took damage.”
- Lower the stakes. “I’m not here to sell you anything today, just offering a free look at your roof while we’re on the street.”
- Make the ask small. A free inspection, not a signature. The close comes after you’re on the roof.
- Set the follow-up. If they’re busy, get a time (“I’ll swing back around six”) and actually come back.
That neighbor-referencing opener only works when you’ve picked the right street. You can’t say “a lot of homes on this block took hail damage” on a block that didn’t take any.
Measure doors-to-deals, not doors
Canvassing dies when nobody tracks it, because “I knocked a lot” isn’t a result. Watch the funnel, not the effort:
- Doors knocked → conversations (did anyone answer and talk?)
- Conversations → inspections booked
- Inspections → signed jobs
Knowing your true doors-to-deal ratio lets you forecast (X signed jobs means roughly Y doors) and compare streets. A fresh storm footprint should crush a random subdivision on the same rep. It’s also the honest way to weigh canvassing against paid channels: a rep’s day has a cost, and doors-to-deals tells you whether it beats a $30 shared lead.
Where the targeting comes from
Everything above depends on knowing which houses to work before anyone knocks or dials. That’s a data problem, and it’s exactly what Overview is built to solve. Draw the neighborhoods you want and pull highly accurate homeowner contacts (name, phone, email, mailing address) for every house inside the boundary. Start with where hail actually hit, then add the older streets and the blocks near your last job you already know. The hail footprint is a live overlay you draw right on top of; the rest is your local knowledge deciding where to draw.
The economics make this easy. Highly accurate records run around $0.25 each versus $15–$40 for a shared digital lead four other roofers also bought. Arming a route with the full contact list, phone numbers included, costs a few dollars, and one signed job pays for a thousand records. So you can afford to both call and knock every high-probability home on it. The throughline: stop overpaying for shared, resold leads, target the exact neighborhoods that will buy (starting with the streets a storm actually hit), and reach those homeowners first, by phone and at the door. Overview gives you the contacts. How you reach them, by dialer, on foot, or both, is your call.
For where canvassing fits among your other channels, start with how to get roofing leads. And if storms are your bread and butter, the Roofer’s Guide to Storm & Hail Lead Generation goes deep on working a footprint fast.
Want to reach the homeowners likely to buy, by phone and at the door? Draw your first neighborhood free or book a 15-minute demo.
Frequently asked questions
Is door-to-door roofing sales worth it? Yes. It’s still one of the highest-converting roofing channels because you’re physically in front of the exact house, which no shared lead form can match. The catch is efficiency: random canvassing wastes the day on new roofs, empty homes, and already-signed neighbors. It’s worth it when you pre-select high-probability streets (storm footprint, roof age, proximity to a recent job) instead of walking blocks cold. The fastest teams pair it with the phone, calling the same streets they canvass so the homeowners who never answer a knock still get reached.
What’s the best time to knock doors for roofing? Late afternoon into early evening on weekdays (roughly 4:30 to 7:00) catches the most people home, plus weekend mornings. But timing matters less than street selection: a well-chosen block at a decent hour beats a perfect time slot on a random subdivision. Track your no-answers and rotate back rather than writing the address off.
How do you find the best neighborhoods to canvass? Stack four signals: where a storm actually hit, neighborhoods old enough that roofs are due (18–25 years), owner-occupied streets that fit the job financially, and blocks near a job you just landed. The more that overlap on one street, the higher the odds. Overview lets you draw a boundary over the streets you pick and pull highly accurate contacts, and the live hail maps show the real storm footprint to draw on.