Roof Age Lookup by Address: 4 Methods That Actually Work
Four ways to estimate a roof's age before you knock — from county permit records to satellite imagery. Accuracy bands and cross-check workflow included.
The single best predictor of whether a homeowner will replace their roof in the next 18 months isn't visible damage. It's age. Roofs in their 20-25 year window get replaced. Younger ones get patched. Older ones — by then — already have a tarp or have already been done.
Which means if you can find out the age of every roof in your service area before you knock, you can prioritize ruthlessly. This post is about the four legitimate ways to do that — three free, one paid — and how to combine them for the highest-confidence pre-knock list possible.
Why roof age beats condition as a sales-conversion predictor
Most roofers default to "find the worst-looking roofs and knock those." That works, but it skews you toward urgent-repair conversations rather than full-replacement quotes. The roofers I know with the best margins focus on roofs in their 18-22 year window — old enough that the homeowner is starting to think about it, but not so degraded that they've already accepted a temporary fix.
The pattern in DFW residential data:
- Roofs 0-10 years old: 1% annual replacement rate (warranty work mostly)
- Roofs 10-18 years old: 3-5% annual replacement rate (storm-driven)
- Roofs 18-25 years old: 15-20% annual replacement rate (the "right time" window)
- Roofs 25+ years old: 8-12% annual replacement rate (deferred, often distressed)
That 18-25 year window is your highest-margin segment. Identifying it before you knock means your day's first 5 conversations are all in the right age band — not random guesses.
Method 1: County permit records (most accurate, slow)
Every roof replacement in most U.S. counties requires a building permit. Those permits are public records — which means the date of the last roof permit at any address is recoverable for free, if you know how to find it.
For DFW specifically:
- Dallas County — search at
dallascityhall.com → building inspection → permit search. Free, web-based, but slow to navigate. - Collin County (Plano/Frisco/Allen/McKinney) — each city runs its own permit system. Plano:
plano.gov/permitsearch. Frisco:friscotexas.gov/permits. - Tarrant County (Fort Worth/Arlington/Grand Prairie) — Fort Worth at
fortworthtexas.gov/permits. Arlington atarlingtontx.gov/city-hall/departments/community-development-and-planning. - Denton County — Denton city portal + smaller cities each have their own.
What you can extract:
- Date of the most recent roof-related permit (re-roof, repair, replacement)
- Permit value (rough proxy for job size)
- Contractor name (competitive intel — who did the last job)
- Material change (asphalt → metal upgrades are red flags for upsell timing)
The math for roof age:
Estimated roof age = today's date − date of last roof permit
If there's NO roof permit on file, the roof is likely original to the home. Cross-reference with the home's year-built (also in public records) — original asphalt on a 1998 home is 27 years old in 2026. That's a strong prospect.
The downsides:
- Not every replacement is permitted. Some homeowners skip permits, especially for cosmetic re-roofs. Permit data tends to under-count by 15-20%.
- Coverage varies. Some counties expose data via API; others require web scraping or in-person visits.
- The data is always 30-90 days stale (permits get filed during work, not before).
When to use: when you're building a target list for a specific zip and have time to script the lookup. Best for shops with a sales ops function.
Method 2: Building age + typical replacement cycle math
If permit records aren't accessible for your area, you can fall back to build year + asphalt lifespan math.
The premise: most homes had their original roof installed within 12 months of completion. Asphalt shingle roofs last 20-25 years before replacement. If a home was built in 2003 and there's no evidence of replacement, the roof is approximately 22 years old.
The math:
Estimated roof age = (today's year) − (year built) − adjustment factor
The adjustment factor accounts for typical replacement cycles:
- If the home is 25+ years old AND has no permit on file: roof is likely 15-25 years old (one prior replacement)
- If the home is 25+ years old AND has a permit from 12-18 years ago: roof is 12-18 years old
- If the home is 20-25 years old AND has no permit: roof is original, 20-25 years
Year-built is in property tax records — free at most county appraisal districts (e.g., Dallas Central Appraisal District at dallascad.org).
Accuracy: roughly ±5 years for asphalt. Tile, slate, and metal hold longer (40+ year lifespan), so this method is less useful for those materials.
When to use: as a fast first-pass filter on a large list. Combine with Method 3 or 4 to refine.
Method 3: Google Earth historical imagery comparison
Google Earth Pro (the desktop app, free download) has historical satellite imagery for most U.S. metros going back 10-20 years. You can pull up any address, scroll through the timeline, and visually see when a roof changed color or texture — which usually means a replacement.
The workflow:
- Open Google Earth Pro → enter the address
- Click the historical imagery toolbar icon (looks like a clock with an arrow)
- Scroll the timeline back to the earliest available image
- Compare the roof's appearance across years
- A clean color/texture change between two adjacent years = replacement event
- The most recent visible change = the current roof's installation year
What you're looking for:
- Sudden color shift between two years (asphalt fades gradually, so a sudden change means new material)
- New ridges or seams appearing
- Visible work signs in any single year (tarps, dumpsters, debris around the house)
Accuracy: very high for a positive signal (if you see a replacement event, it's almost certainly real). Lower when imagery gaps span 3+ years.
The downside: doesn't scale. You can do 5-10 addresses an hour manually. Best used to verify high-value prospects from Method 1 or 2.
Method 4: AI-estimated age from satellite condition signals
This is what Methods 2 and 3 collapse into when automated. Modern AI vision models can estimate roof age within a ±3-5 year band by analyzing visible condition signals from a single satellite image:
- Granule loss progression (early/mid/late stage)
- Color fading patterns (intensity loss correlates with UV exposure years)
- Shingle curl onset (typically begins at year 15-18 on asphalt)
- Texture mottling (uniform → patchy as roof ages)
- Algae onset patterns (algae appears 8-15 years in, depending on humidity)
- Neighborhood cohort signals (when 8 adjacent homes have similar-aged roofs, install cohort context strengthens the estimate)
The accuracy of the latest models on asphalt roofs in our internal validation: ±4 years at 80% confidence across 200 ground-truth comparisons. Tile and slate are harder (longer lifespans, more variation) — those land at ±7 years.
Importantly, this method gives you an age band for every roof in your service area in one pass — not just the ones you've manually queried. That's the scaling unlock that makes large-area prospecting feasible.
Roofbird gives you the AI-estimated age band on every property in your service area, along with the visible signals that contributed to the estimate. Open the DFW sample dashboard to see what the output looks like — 10 unlocked leads with full age + condition breakdown.
Cross-checking the four methods before you knock
The roofers with the best pre-knock conversion rates use 2-3 methods in combination, not just one. The cross-check workflow:
Pre-knock checklist:
- AI age estimate (Method 4): is this roof in the 18-25 year window?
- County permit cross-check (Method 1): does the permit history confirm or contradict?
- Year-built sanity check (Method 2): does the home's age make the AI estimate plausible?
- Visual confirmation (Method 3): does Google Earth show the visible signals the AI flagged?
When 3 of 4 agree, you have a high-confidence prospect. When they disagree (e.g., AI says "20 years" but permit says "8 years ago"), the AI is usually wrong — the recent permit is ground truth.
A practical example: I scored a Plano home as vision-9 / age 22 years from satellite. The permit record showed a re-roof 4 years ago. The Google Earth timeline confirmed: the prior roof was clearly degraded; the current roof was replaced in 2022 but installed with a low-quality 3-tab product that aged 18 years in 4. Cross-checking saved me from a useless knock — the homeowner just paid for a roof, even if it looks bad.
What homeowners search for vs. what roofers need
Most "roof age lookup by address" search traffic is homeowners trying to figure out when their own roof was installed. Roofers searching the same query usually want this for prospecting.
For homeowners reading this: Method 1 (county permits) is your fastest path. Plug your address into your county building department's permit search and the most recent re-roof permit tells you the year.
For roofers reading this: Method 4 (AI scoring across your whole service area) is the scaling unlock. The other three are for verification.
What to do this week
If you've never built a roof-age-prioritized prospect list before, the lowest-friction starting point:
- Pull a list of 50 homes in your top zip from county property records (free, ~15 minutes)
- Filter by year-built: only keep homes 20+ years old
- For each, check the permit record: drop any with a re-roof in the last 15 years
- The remaining list is your high-priority age cohort — knock these first
If you want to skip the manual zip-canvassing step, Roofbird scans every roof in your service area and gives you the top 200-500 prospects ranked by AI-estimated age + condition. Free 25-lead trial.
— Jake
Written by
Jake Thompson
Have a question about anything in this post? Reach the Roofbird team at support@roofbird.ai.
Try Roofbird — 25 free leads in your area
See a sample dashboard for DFW first, no signup needed. Trial loads 25 free pre-scored leads in your own service area.
More for roofers
How to Generate Roofing Leads Without Cold Calling: 7 Methods Ranked by Cost and Close Rate
Seven proven ways to generate roofing leads without cold calling — ranked by realistic cost-per-lead and close rate. From free GBP optimization to satellite roof scoring. No shared leads, no phone room.
How Solo Roofing Contractors Get Leads Without Angi or HomeAdvisor (7 Methods That Actually Work)
Seven proven ways to generate roofing leads without paying Angi or HomeAdvisor's marketplace tax — ranked by cost, effort, and lead quality for solo contractors and small crews.
Best Alternatives to EagleView for Roofing Measurements and Lead Generation (2025)
Honest breakdown of EagleView alternatives for roofers — Hover, Roofr, Nearmap, GAF QuickMeasure, and AI prospecting tools. Covers both measurement accuracy and lead generation, because most comparison articles only cover one.