How Roofing Contractors Find Homeowners with Old Roofs — Without Door-Knocking

Five methods roofing contractors use to find homeowners with aging roofs — permit pulls, storm data, direct mail, paid search, and satellite roof scoring — with real cost benchmarks and conversion tradeoffs for each.

Door-knocking closes deals. That part is still true. But it doesn't scale — and in 2025, it's also getting legally complicated. Dallas requires a solicitation permit and a $50 fee per canvasser. Chicago has a city-wide ordinance that lets residents post "no soliciting" signs with legal teeth. Aurora, Colorado added permit requirements in 2023 that carry a $500 fine for violations. The list of municipalities moving in this direction is long and getting longer.

The contractors reading this already know the real problem isn't the permit fee. It's that door-knocking is a blunt instrument. You're paying a rep $25–$35/hour to walk up to doors where the roof might be fine, the homeowner might be at work, and the neighborhood might have a HOA that filed a complaint last month. The conversion rate on cold doors runs 1–3% in most markets. That's a lot of windshield time for a thin pipeline.

The alternative isn't to stop selling in person. It's to only show up at doors where you already know the roof is likely to need work — and to build that list before you deploy any field rep. The five methods below do exactly that. They range from free-but-manual to paid-and-automated, and the best ones let you filter by roof age, material, and damage probability before you spend a dollar on outreach.

Why Door-Knocking Is Losing Ground (and What Contractors Are Replacing It With)

Three forces are pushing roofing contractors toward desk-based prospecting:

Solicitation ordinances are spreading. Beyond Dallas and Chicago, cities like Naperville IL, Henderson NV, and Frisco TX have added permit requirements or outright restrictions on door-to-door solicitation in residential zones. Enforcement is inconsistent but the liability is real — one complaint can cost more than a month of leads.

Canvassing crews are expensive. A two-person door-knocking crew running 8 hours costs $400–$560 in labor before you count gas, vehicle wear, or the rep who burns out after three weeks of rejections. At a 2% conversion rate on cold doors, you're paying $200–$280 per appointment set — before the estimator shows up.

The better roofers have already made the shift. The shops with the best margins in 2026 aren't knocking fewer doors because they're lazy. They're knocking fewer doors because they're targeting better. They identify which homes need work first, then send a rep — or a mailer, or a text — to that specific address. The door-knock, when it happens, is a warm follow-up, not a cold pitch.

The methods below are how they build that list.

Method 1: Pull Building Permit Data for Expired or Overdue Roofs

Cost: free to ~$20/month. Scale: low. Proactive: partial.

Most counties publish building permit records online. A roof permit pulled in 2004 means that roof is now 21 years old — well past the 20–25 year lifespan of standard 3-tab asphalt shingles. You can use this to build a targeted list of addresses where a replacement is statistically overdue.

Here's the exact process: Go to your county assessor or building department website and search for permit type "roofing" or "re-roof." Filter by year range (2000–2007 for roofs hitting end-of-life now). Export the results, cross-reference with current ownership data from the same assessor portal, and you have a working list.

A few counties with clean GIS portals worth knowing: Harris County TX at hcad.org, Maricopa County AZ at mcassessor.maricopa.gov, and Arapahoe County CO at co.arapahoe.co.us/assessor. Most let you export to CSV without a FOIA request.

The honest limitation: permit data is incomplete everywhere. Homeowners replace roofs without pulling permits more often than the industry likes to admit — some estimates put unpermitted replacements at 20–30% of all residential re-roofs. You're also doing manual data cleaning, which caps useful output at around 300–500 records per week without a dedicated data person.

Good for small markets and rural counties. Doesn't scale in dense suburban metros without automation.

Method 2: Use Storm Event Data + Insurance Claim Proxies

Cost: free to $40/lead. Scale: medium. Proactive: reactive.

After a hail or wind event, NOAA's Storm Prediction Center publishes reports at spc.noaa.gov that identify affected zip codes by date, hail size, and storm path. A contractor can cross-reference storm-hit zips with property age data — homes built before 2005 in a zip that took 1.5"+ hail last Tuesday are high-probability targets.

The free US hail map overlays NOAA hail reports on a county-level map updated daily, which makes the geographic targeting step a lot faster than parsing SPC raw data.

The limitation here is competition. Storm-chaser roofers flood impacted markets within 48 hours of an event. If you're the fifth contractor to knock on a door in a hail-hit neighborhood, the homeowner's already been pitched and is either annoyed or already signed. Margin compresses fast in heavily-chased markets.

Enterprise data vendors like CoreLogic and Verisk sell storm-claim propensity scores — they're good, but pricing starts around $5,000/month, which is enterprise-tier for a 3-person crew.

Method 3: Direct Mail to Homes by Roof Age (Property Data Lists)

Cost: $25–$80 per lead. Scale: high. Proactive: partial.

Data brokers sell property records filterable by year built, roof material, and last permit date. ATTOM Data, DataTree, and PropStream are the three most contractor-accessible options. A typical workflow: pull single-family residential homes built before 2005 in your target zip codes, filter out condos and commercial, export, and hand the list to a direct mail vendor.

Cost benchmarks that are actually accurate: ATTOM list pulls run $0.10–0.25 per record depending on volume. A standard direct mail piece — postcard, full-color, addressed — runs $0.50–1.00 all-in with printing and postage through vendors like Every Door Direct Mail (EDDM) or Click2Mail. At a 0.5–2% response rate, you're paying $25–$200 per inbound response. Closer to $25–$80 for a qualified appointment.

The core limitation: year-built is not roof age. A 1998 house may have had its roof replaced in 2015 after a hail event. Property records don't capture that unless a permit was pulled. You're mailing to a list that includes a meaningful percentage of homes with roofs that are actually fine — which drives response rates down and cost-per-lead up.

Method 4: Google LSA + Paid Search Targeting Storm-Affected Areas

Cost: $80–$150 per lead. Scale: high. Proactive: reactive.

Google Local Services Ads let you geo-target by zip code and appear at the top of results when homeowners search "roof repair near me" or "roofing contractor [city]." After a significant storm event, search volume in affected zips spikes 3–10x within 72 hours. Contractors who pre-fund LSA campaigns and have their geo-targeting set up capture this surge without scrambling.

The cost benchmark: roofing LSA leads averaged $80–$150 per lead in competitive markets in 2024 per Google's own benchmarking data. That's per lead — not per close. At a 15–20% close rate on LSA leads (they're inbound and warm), you're looking at $400–$1,000 per acquired customer. That's better than Angi's shared-lead model in most markets, but it's still reactive — you're waiting for homeowners to search, not finding them before they know they have a problem.

The other issue: LSA requires consistent budget management. Underfund it and your ads go dark during the surge window you were trying to capture.

Method 5: Satellite Roof Scoring (The Proactive Approach)

Cost: $15–$50 per lead. Scale: high. Proactive: yes.

This is the only method on this list that lets you build a ranked list of specific addresses by re-roofing likelihood before any storm, before any permit, before any homeowner searches for anything.

Here's how it works in plain language: AI models trained on satellite and aerial imagery assess each parcel in a target geography for roof age, material type (asphalt shingle, metal, tile, TPO), visible damage indicators, and replacement probability. The output is a scored list — address, estimated roof age, material classification, damage probability score, and homeowner contact data — ranked from highest to lowest re-roofing likelihood.

A contractor targeting a suburban zip code gets back something like: "412 Maple St — estimated roof age 19–23 years, asphalt 3-tab, high replacement probability." That's a specific, actionable lead. Not a zip code. Not a year-built range. An address with a score.

The conversion advantage is significant. Outreach to pre-scored high-probability homes converts at 3–5x the rate of cold canvassing because the pitch is specific. "Our records show your roof may be 18–22 years old — we're in your neighborhood this week and can do a free inspection" lands differently than "we're a local roofer, can we look at your roof?" The homeowner doesn't know you ran satellite imagery. They just know you seem to know something about their house.

Modern AI roof models trained on millions of parcels achieve 85–92% accuracy on age classification. Damage detection is directional — it flags probability, not a guarantee, and a physical inspection is still required before quoting. Satellite scoring is a prospecting tool, not an adjuster replacement.

Roofbird is built specifically for this use case — satellite imagery scored at the address level, exclusive leads, exportable to CSV or CRM. See how the scoring works →

Comparing All Five Methods

Method	Avg. Cost per Lead	Scalability	Proactive?	Requires Field Time?
Permit data pulls	$5–$20	Low	Partial	No
Storm event data	$10–$40	Medium	Reactive	No
Direct mail (property lists)	$25–$80	High	Partial	No
Google LSA / paid search	$80–$150	High	Reactive	No
Satellite roof scoring	$15–$50	High	Yes	No

The best contractors don't pick one method and ignore the rest. They layer. Satellite scoring tells you which homes to target. Direct mail or a targeted call campaign tells you how to reach them. LSA captures the homeowners who are already searching. The combination produces a pipeline that doesn't depend on any single channel going dry.

Roofbird is designed to be the identification layer in that stack — start your first zip code at $199 with no shared leads →.

FAQ

Q: How accurate is satellite imagery for detecting old or damaged roofs?

Current AI roof models trained on large parcel datasets achieve 85–92% accuracy on age classification. Damage detection is probabilistic — the model flags homes with elevated damage probability, not a confirmed damage assessment. Always do a physical inspection before quoting. Satellite scoring is a prospecting filter, not a replacement for an adjuster or an estimator on a ladder.

Q: Is it legal to use satellite imagery and property data to prospect for roofing leads?

Yes. Satellite imagery of properties is publicly permissible — the same imagery is available on Google Maps and used by insurance companies, municipal assessors, and appraisers. Property records are public data in all 50 states. Outreach using that data must comply with CAN-SPAM for email and TCPA for phone and text — standard compliance requirements that apply to any outreach campaign.

Q: How is satellite roof scoring different from buying leads on Angi or HomeAdvisor?

Angi and HomeAdvisor sell the same lead to 3–7 contractors simultaneously. You're racing to be the first callback on a prospect who's already been pitched by your competitors before you finish dialing. Satellite scoring is the opposite model — you identify the home, you own the lead, and you control which homes you target based on roof age, material, and geography. There's no shared-lead race because no one else has the same list.

Q: What markets does satellite roof scoring cover?

The continental U.S. is covered by the major aerial and satellite imagery providers that feed AI roof models. Verify current coverage for your specific market at roofbird.ai before purchasing — rural markets with lower imagery resolution may have reduced scoring accuracy compared to dense suburban areas.

Q: How many leads should I expect from a starter plan?

It varies by zip code density. A dense suburban zip in the DFW or Atlanta metro might return 400–800 scored addresses in a single pull. A rural zip might return 80–150. The scoring filters that down to high-probability targets — typically 15–30% of parcels in a given area score in the top tier for re-roofing likelihood. Run the numbers for your target zip before committing to a coverage area.

What to Do This Week

Pick one method and run a real test — not a mental model of a test, an actual one.

If you're in a market that just had a hail event, pull the US hail map, identify the affected zips, and cross-reference with your county assessor's permit data for homes built before 2005. That's a free list you can have in 90 minutes.

If you're in a stable market without recent storm activity, pull a PropStream trial ($99/month, cancel anytime), filter your highest-volume zip for single-family homes built 2000–2007, and price out a 500-piece EDDM postcard run. That's a real direct mail test for under $600 total.

If you want to skip the manual work entirely, run a satellite score on one zip code. You'll have a ranked address list with homeowner contact data before the end of the day — no permit hunting, no data cleaning, no door-knocking required.

The contractors who build consistent pipelines aren't doing anything exotic. They're running one of these methods every week, tracking cost-per-lead honestly, and doubling down on whatever produces the best margin in their specific market. Start with one. Measure it. Adjust.