How to Identify Aging Roofs in a Neighborhood (A Roofing Contractor's Field Guide)

Most roofing crews canvass a neighborhood the same way they did twenty years ago: pick a subdivision that "looks old," park the truck, and start walking down the street knocking every door. By the end of the day they've talked to forty homeowners, half of whom have roofs with another decade of life left, and they've skipped the three houses on the next street over whose roofs are visibly cooked. The roofs were aging out the whole time. The crew just couldn't see it from where they were standing.

Identifying aging roofs in a neighborhood is a skill, and like any skill it has a ground-level version and a systematized version. The ground-level version is reading shingles with your eyes and a pair of binoculars. The systematized version is using install dates, aerial imagery, and per-roof data to rank an entire territory before anyone leaves the shop. Good contractors do both, and they know when each one is worth the time.

This guide walks through the whole thing: the visual cues that actually predict a near-term replacement (and the ones that fool people), how to estimate the install year of a roof without ever climbing it, how to read a neighborhood's age profile from the development pattern, the aerial and records workflows that let you cover a thousand homes in an afternoon, and how to turn all of it into a ranked door list your reps will actually work. The goal isn't to admire old roofs. It's to find the ones that are close enough to the end that the homeowner is ready to act, and then to spend your knocking hours on those doors instead of the whole street.

Why "aging out" beats "looks old"

Before the cues, get the target right. The roofs worth your time are not simply the oldest roofs. They're the roofs near the point where the homeowner is ready to replace, which is a narrower and more valuable set.

A roof that's two years from failure is a far better conversation than one that failed five years ago and the owner already lived with it, patched it, or sold the house. And a roof that's brand new but took a beating in a hailstorm can be a better conversation than a 22-year-old roof that's somehow still holding. "Aging out" is about the gap between a roof's condition and the end of its service life. "Looks old" is just a vibe, and vibes send crews to the wrong doors.

What drives a roof toward the end of its life:

• Material and its rated service life. A 3-tab asphalt shingle is a different clock than an architectural (laminate) shingle, which is a different clock than a metal panel or a tile field. The asphalt-shingle industry generally describes field service lives in ranges, not guarantees, and a printed warranty length is a marketing number, not a failure date.
• Climate and sun exposure. Asphalt shingles age primarily through thermal cycling and UV exposure. A south- and west-facing slope in a hot, high-UV climate ages faster than a north slope in a mild one. Two identical roofs installed the same week in the same subdivision can be years apart in remaining life depending on orientation and shade.
• Ventilation and the deck below. A poorly ventilated attic bakes shingles from underneath and shortens their life. You often can't see this from the street, but you can sometimes infer it.
• Storm history. Hail bruises the mat and knocks granules loose; wind lifts and creases shingles and breaks seal strips. A single significant event can move a roof years closer to replacement, and repeated marginal events add up.
• Install quality. Over-driven nails, missing starter strip, reused flashing, and short nailing all shorten real-world life regardless of the shingle's rating.

You can read most of these from the ground or from the air with practice. The clock you can't see directly is the calendar age, and that's where records and imagery come in. Hold both ideas at once: a roof is "due" when its accumulated wear plus its calendar age put it near the end, and your job is to spot the roofs sitting in that window across an entire neighborhood without inspecting every one.

Part 1: Reading a single roof from the ground

This is the foundational skill. Even if you later run aerial and records workflows over a whole city, you still need to read a roof with your eyes, because that's what you'll do standing in the driveway and what you'll teach every new canvasser. Carry a decent pair of binoculars (8x42 is a good all-around choice) or use the zoom on a phone. Most of these cues are visible from the street or the sidewalk.

Granule loss and a thinning, bald look

Asphalt shingles are protected by mineral granules. As the shingle ages, granules shed, exposing the asphalt mat to UV. From the ground, heavy granule loss reads as a roof that looks darker, shinier, or patchy compared to its neighbors, with the slope losing its uniform color.

Two tells:

1. Shiny or black streaking down the slope where the granules have worn off and bare asphalt is showing. Don't confuse this with algae streaking (more on that below).
2. Granules in the gutters and at downspout splash blocks. From a driveway you can often see a buildup of sand-like granules where the downspout dumps. A handful of granules after a new roof's first rain is normal break-in; a steady accumulation under an older roof is a roof shedding its protection.

Granule loss is one of the better visual predictors because it tracks the actual failure mechanism. A roof that's gone matte and patchy across whole slopes is genuinely near the end, beyond merely cosmetic wear.

Curling, cupping, and clawing

As shingles lose their oils and the mat dries out, the edges deform. You'll hear three words for it:

• Curling: edges turn up.
• Cupping: the center of the tab dishes down while edges rise (often tied to attic heat/ventilation).
• Clawing: the middle rises while the edges stay down.

From the ground, all three create a rough, shadowed, "textured" look across the slope instead of clean flat courses. Late-day, low-angle sunlight is your friend here: raking light throws shadows off every curled edge and a deformed slope lights up like sandpaper. A roof that looks crisp and flat at noon can look visibly shot at 6 p.m. Train reps to drive territories in the golden hour for exactly this reason.

Missing, cracked, and creased shingles

Missing tabs and visible gaps mean wind has already broken seal strips and the roof is shedding pieces. Creased shingles, a horizontal line across a tab where it folded up and slapped back down, are a classic wind-damage signature and are often visible with binoculars. A field of scattered creases across windward slopes after a storm is a strong canvassing signal in its own right.

Cracking shows up as a network of fine lines, sometimes called crazing or alligatoring, as the asphalt embrittles with age. You usually need binoculars or a zoom photo to confirm it, but a roof that's cracking across the field is at or past the end of its useful life.

Sagging, waviness, and an uneven deck

Look at the ridgeline and the planes of the roof against the sky. A ridge that dips in the middle, slopes that look wavy or rippled rather than flat, or a deck that seems to undulate can indicate long-term moisture, rot in the decking, or structural issues underneath. This is a more serious, higher-urgency signal than worn shingles, and it usually means the homeowner has a real problem whether they know it or not. A sagging deck reads from the street if you look at the lines instead of the surface.

Flashing, penetrations, and the details

Older roofs often show their age first at the details: rusted or lifting step flashing along walls and chimneys, deteriorated pipe-jack boots (the rubber collar around plumbing vents cracks and is a common leak source on roofs around the 10-to-15-year mark), and tired valley metal. From the ground you can often spot rust streaks running down from a chimney or a vent, or a visibly cracked boot if you zoom in. These details age on their own clock and frequently fail before the field shingles do, which makes them a useful early flag.

Algae streaks: a trap, not a signal

The black streaks you see running down many roofs are usually Gloeocapsa magma, a blue-green algae, not roof failure. Algae streaking is largely cosmetic and is heavily influenced by shade, humidity, and whether the shingle has copper/zinc granules, not by age alone. Plenty of 8-year-old roofs in humid, shaded lots are badly streaked while sun-baked 20-year-old roofs nearby are clean.

The mistake new canvassers make constantly is equating black streaks with an old, dying roof. Don't. Algae tells you about the microclimate, not the remaining service life. Use the wear cues above (granule loss, curling, creasing, cracking) to judge age, and treat streaking as, at most, a conversation opener about appearance.

A quick ground-read scorecard

When you're scanning a street, a simple mental tally keeps reps consistent. Count how many of these are present on a given roof:

Three or more high-weight cues on a single roof means it's worth a knock today. One or two means flag it and check the install year. Zero means move on, unless records or storm data tell you otherwise.

Part 2: Estimating the install year without climbing

Visual wear tells you condition. Install year tells you the calendar. Put them together and you get a real read on remaining life. Here's how pros estimate the install year of a roof without ever getting on it, and why the answer should always be a range, not a date.

Read the neighborhood's build vintage first

The single most useful number when sizing up a subdivision is its build era. Tract developments get roofed in waves: the builder shingles a whole phase within a year or two of construction. So in a neighborhood built in, say, 2004, a large share of homes still wear roofs that are either the original (now around 20+ years old and well into replacement territory) or a single previous re-roof.

This is why "roof age by neighborhood" is a real planning concept and not a fuzzy one. Within one development phase, the original-roof cohort moves through its service life together. When the first wave starts replacing, the rest of the phase is usually close behind. Find the development that was built right at the front edge of an asphalt roof's typical service window and you've found a neighborhood where a big fraction of homes are due around the same time.

How to get the build vintage:

• County assessor / property appraiser records list the year built for nearly every parcel, free, online, in most U.S. counties. This is your anchor.
• Plat and subdivision records show when phases were platted and developed.
• The architecture itself, window styles, garage placement, siding materials, and street layout cluster by decade once you've trained your eye on local stock.

Year built is not roof age, but it's the floor: a roof can't be older than the house, and on a never-re-roofed home it's the same age as the house.

Spot the re-roof: has it been done already?

The big question on any individual home is whether the original roof is still up there or it's already been replaced once. Tells that a roof has been redone:

• Color and granule freshness that's noticeably newer than the house's vintage and cleaner than neighbors of the same build year.
• Modern architectural (laminate) shingles on an older home. If the neighborhood was built when 3-tab was standard and a house wears thick, dimensional shingles, it's been re-roofed.
• Crisp ridge caps, new pipe boots, new drip edge against an otherwise dated house.
• Permit records (next section) that show a reroof permit pulled in a given year.

If a home in a 2002 subdivision is wearing what's clearly a 3-tab roof with heavy wear, you're likely looking at the original, around two-plus decades old. If it's wearing crisp architectural shingles, someone replaced it and you should find out when before you spend a knock on it.

Pull permit history

Many jurisdictions require a permit for a roof replacement, and many publish permit records online or release them on request. A reroof permit gives you something close to an actual install date for that address. Permit data is uneven, plenty of work gets done without a permit, especially in rural areas and for "repairs", so absence of a permit doesn't prove the roof is original. But a present reroof permit is a strong, dated signal that a home has already been done and isn't your target for years.

Worked into a territory plan, permit history lets you subtract the recently-reroofed homes from your knock list, which is often a meaningful slice of an older subdivision and pure wasted effort if you don't filter it out.

Match material and style to an era

The shingle type itself dates a roof:

• 3-tab asphalt dominated residential roofing for decades and is now largely displaced by laminate shingles on new and re-roof work. A 3-tab roof skews older or budget.
• Architectural / laminate (dimensional) shingles became the mainstream choice over the past couple of decades; their thickness and shadow lines are easy to spot from the street.
• Older heavyweight materials, certain tiles, slates, and early metal, carry their own much longer service clocks and should not be judged on an asphalt timetable at all.

Knowing your local stock, what builders installed in which years around you, turns a glance at the slope into a date estimate.

Always express age as a range

Here's the discipline that separates pros from guessers: the install year you derive is an estimate with error bars, and the condition is what actually matters. A responsible read sounds like "this roof is most likely 18 to 23 years old based on the build year and the original-looking 3-tab, and it's showing heavy granule loss and edge curl, so it's near the end," not "this roof was installed in 2003." You almost never know the exact date from the outside, and pretending you do gets people in trouble. A range plus a condition read is honest and is all you need to decide whether to knock.

Part 3: Reading the whole neighborhood

Now zoom out from the single roof to the street and the subdivision. The pattern of how neighborhoods are built and re-roofed is what lets you target a whole territory instead of one house at a time.

The replacement-wave pattern

Because tract builders roof in phases, a subdivision's roofs don't age randomly. They move in cohorts. Picture a development built across 2000 to 2003 in four phases. The Phase 1 originals hit the front edge of the asphalt service window first; a handful of homeowners replace; then as the cohort all crosses the same threshold, replacements accelerate across the phase, and the later phases follow a couple of years behind.

For a contractor this means two things. First, you can predict which neighborhoods are entering their replacement window this year by their build vintage. Second, once you see a few fresh roofs popping up on a street of otherwise-original homes, that street is heating up, the cohort is crossing the line, and the un-replaced neighbors are prime, ready-to-act conversations. A cluster of new roofs is a buy signal for the block, not a sign you missed it.

Walk the street for relative wear

Within a block of same-vintage homes, the eye is great at relative comparison even when it's bad at absolute dating. Drive or walk a street and rank roofs against each other:

• Which roofs look matte, dark, and worn next to neighbors that still look uniform?
• Which ones already got replaced (crisp, fresh) so you can subtract them?
• Which slopes facing south and west are visibly more cooked than the north slopes on the same house, confirming sun-driven aging?

After one pass you'll have the street sorted into "already done," "clearly worn, knock now," and "holding, check later." That relative sort is more reliable than trying to put an exact age on any single roof, and it's exactly what you want a canvasser doing on foot.

Microclimate and orientation across the neighborhood

The same neighborhood ages unevenly. Lots with mature tree cover on the south side shade their roofs and slow UV aging (while sometimes worsening algae). Homes on a ridgeline or in an open field with full sun and wind exposure age faster than homes tucked into a sheltered cul-de-sac. Knowing these patterns lets you weight your read: two homes of the same build year aren't the same risk if one bakes in full afternoon sun and the other sits under oaks.

Don't forget the storm layer

Neighborhood age tells you about wear-out. Storm history tells you about damage, and the two combine. A subdivision that's only 8 years old, too young to be aging out, can still be full of due roofs if a significant hail or wind event rolled through and bruised mats and lifted shingles across the area. Conversely, an older neighborhood that's never taken a real hit may be aging slowly and uniformly.

The trap is treating a storm as uniform across a neighborhood. It isn't. Hail swaths are narrow and patchy; wind funnels and gusts hit some slopes and skip others; a single street can have homes that were pounded and homes two doors down that were barely touched. Where the storm passed is a fat region on a map. Where it actually landed hard enough to matter on a given roof is far more specific, and that distinction is the difference between knocking a whole zip code and knocking the right forty houses. Important honesty here: a storm having passed over an address is not proof that a particular roof is damaged. It raises the odds. The roof still has to be inspected and documented before anyone says a word about a claim.

Part 4: The aerial and records workflow (covering a thousand homes before lunch)

Ground reading is essential but it doesn't scale. You can't drive every street in a metro at golden hour with binoculars. The systematized version of identifying aging roofs uses overhead imagery and public records to triage a huge area, then sends crews to the doors that survive the filter. Here's the workflow pros use.

Step 1: Define the territory and pull build years

Start with a polygon, a subdivision, a zip, a storm swath, whatever you're working. Pull year-built for every parcel from the county assessor. This alone segments the area into build-vintage cohorts so you can see which neighborhoods are sitting in their replacement window right now. Many counties offer this as a downloadable dataset or via a parcel-data provider; otherwise it's a per-address lookup you can batch.

Step 2: Layer in permit history

Where available, overlay reroof permits to subtract homes that have already been replaced. This is the single highest-value filter for an older neighborhood, because the recently-reroofed homes look fine and are dead doors for years, and you want them off the list before a rep ever drives by.

Step 3: Read roofs from aerial and oblique imagery

High-resolution aerial imagery, both top-down (orthographic) and angled (oblique), lets you read roof condition at scale. From good overhead imagery you can often see:

• Color and tonal uniformity (worn, patchy slopes vs. uniform fresh ones)
• Obvious missing sections, tarps, and patches
• Streaking and staining patterns
• Recently-replaced roofs that stand out as crisp and bright against aged neighbors

Obliques are especially useful because the angled view reveals slope texture, curling, and surface deformation that a straight-down shot flattens out. Imagery freshness matters: a two-year-old image misses two years of aging and any recent storm, so know how current your source is. Top-down imagery alone won't catch subtle curling, so it's a triage tool, you still confirm marginal calls on the ground or with a closer inspection.

Step 4: Add the storm layer per roof, not per region

If you work storm restoration, overlay hail and wind history. Standard practice is to pull historical event data and hail-swath maps (NOAA's Storm Prediction Center publishes storm reports, and several commercial providers sell refined hail-swath and wind data). The upgrade that matters is going from "this zip got hail" to an estimate of what each individual roof likely experienced, accounting for the storm's path, intensity, and the roof's own exposure. The narrower you can make that, the less time your crews waste on roofs the storm didn't actually wear out.

Step 5: Score and rank every address

Now combine the layers into a single per-address priority:

• Build vintage / estimated roof-age range (older originals score higher)
• Already-reroofed flag from permits or fresh imagery (drops the score hard)
• Visible wear from imagery (worn slopes score higher)
• Storm exposure modeled for that specific roof (recent significant exposure scores higher)
• Orientation and shade where you can get it

The output is a ranked knock list: the addresses most likely to be aging out or storm-worn, sorted, with the dead doors already removed. That's the artifact that changes a canvassing day. Instead of "walk every door on Maple," it's "these 40 addresses, in this order, on this route."

Step 6: Confirm on the ground and document

The list gets you to the right doors; it doesn't replace an inspection. A rep still verifies condition at the house and, when there's a roof to look at, documents what's actually there, photos, measurements, the specific wear and any storm damage, before anyone discusses options. The data narrows where you spend attention. The inspection and the documentation are what stand behind any conversation with the homeowner.

Part 5: Where RoofPredict fits

Everything in Part 4 is doable by hand. The reason most contractors don't run it well is that assembling assessor data, permit history, fresh imagery, and storm layers, per address, across a whole territory, and then keeping it current, is a lot of work to stand up and maintain. That's the gap RoofPredict is built to close.

RoofPredict does two things that map directly onto this workflow:

1. It estimates roof age house-by-house from aerial imagery and gives you a range per address, not a single date you'd have to defend. That's the honest form of the answer, the same range-plus-condition read a good estimator gives from the driveway, produced across an entire neighborhood at once so you can see which homes sit in their replacement window without driving every street.
2. It models storm physics per roof, not only where the storm passed. Instead of painting a whole zip as "hit," it estimates what each individual roof likely experienced from a given hail or wind event based on the storm and that roof's exposure. The line we use internally is, we model the storm on each roof, not only where it passed, and that's the part that keeps crews off the doors a storm map would have wrongly flagged.

Put together, you get the ranked knock list from Step 5 without building the pipeline yourself: the roofs that are aging out and the roofs a storm actually wore out, sorted so your reps knock those first and skip the rest.

The honest limits, because you should know them before you rely on it:

• Roof age is a range, not a date. Aerial imagery can place a roof in a likely age band; it can't read a hidden install receipt. You'll still find the occasional surprise re-roof. Treat the range as a strong prioritization signal, not a guarantee about any single home.
• A modeled storm exposure is odds, not proof. That a roof probably saw damaging hail is a reason to go look, not evidence of damage. The roof still has to be inspected and the condition documented before any homeowner conversation about a claim. RoofPredict points the crew at the door; it doesn't decide what's on the roof.
• It's a targeting tool, not a lead-buying service. It ranks the doors and routes in territory you choose to work. It doesn't hand you a homeowner who's already raised their hand; your reps still earn the conversation. For a lot of owners that's the point, the work and the relationship stay yours.

Used the way it's meant to be used, it does the unglamorous part, turning a whole subdivision into a sorted, de-duplicated, storm-aware knock list, so your people spend their hours in driveways instead of in spreadsheets, and in the right driveways.

Part 6: Turning identification into a working canvass plan

Identifying the roofs is half the job. Here's how the better operators convert a read on a neighborhood into a day that actually produces inspections and signed work.

Build the route around the ranked list

Don't hand a rep a neighborhood; hand them a route. Order the ranked addresses into an efficient walking or driving path so they hit the highest-priority doors with the least backtracking. A rep working 40 pre-qualified doors in a tight loop will out-produce one wandering a whole subdivision knocking everything, by a wide margin, because every door has a reason behind it.

Tier the doors and script accordingly

Not every door on the list deserves the same approach. A simple tiering:

• Tier 1, clearly aging out and/or strong storm exposure. Lead with a specific, honest observation ("I was looking at roofs on this street and yours is showing a lot of wear, granules in the gutters, curling on the south slope, mind if I take a closer look?"). These convert to inspections fastest.
• Tier 2, in the age window but holding, or marginal storm exposure. Softer, education-first approach; offer the free inspection and let condition decide.
• Tier 3, recently reroofed or clearly fine. Usually skip, or a quick referral ask. Don't burn a pitch here.

The ranked data is what lets a rep walk up already knowing which tier they're in, which is the difference between a confident, specific opener and a generic one.

Train the eye, then trust the data, then verify on the roof

The sequence that works:

1. Train every canvasser on the ground-read cues (Part 1) so they can confirm what the data flagged and read the houses between the flagged ones. A rep who can't tell granule loss from algae streaking will misjudge doors no matter how good the list is.
2. Use the ranked list to decide where they spend the day so judgment is applied to the right 40 doors, not diluted across 400.
3. Verify on the roof and document before any options or claims conversation. The inspection is the source of truth; the targeting just got you there efficiently.

Track outcomes and feed them back

Log what you find behind each door: was the roof actually due, already done, storm-damaged, fine? Over a few neighborhoods this tells you how well your identification is calling it and where it's off, maybe your imagery is stale, maybe a local builder used a longer-life shingle than you assumed, maybe a storm swath was narrower than the map showed. Tighten the inputs and the next territory's list gets sharper. Identification isn't a one-time read; it's a loop that gets better the more honestly you log results.

Part 7: Material-specific aging clocks

Almost everything above assumes asphalt shingles, because that's the bulk of residential stock in most U.S. markets. But you'll work neighborhoods with other materials, and judging them on an asphalt timetable produces bad reads. Here's how the major materials age and what to look for on each.

Asphalt 3-tab vs. architectural

Within asphalt there are two clocks. Classic 3-tab is thinner and generally ages out sooner than the thicker laminate (architectural) shingles that dominate current work. On the street, 3-tab reads as flat, uniform tabs with visible cutout lines; architectural reads as thick, staggered, dimensional shadow lines. When you see 3-tab on a home old enough that the neighborhood was built in the 3-tab era and it's showing wear, you're very likely looking at an original roof deep into its service window. Architectural on an older home usually signals a prior re-roof, check the date before you spend a knock.

The failure cues are the same for both, granule loss, curling, cracking, lost tabs, but 3-tab tends to show them earlier and more uniformly, while architectural can hold appearance longer and then go quickly once the mat starts breaking down. Don't let an architectural roof's good looks fool you on a home that's clearly old enough to be near the end; check the details (boots, flashing, ridge) which often give it away.

Metal panels and standing seam

Metal roofs run on a much longer clock than asphalt and age differently. Instead of granule loss and curling, you're watching for fastener and seam issues, backing-out exposed fasteners and failed neoprene washers on screw-down panels, oil-canning, finish chalking and fade, rust at cut edges and scratches, and loosened or separated seams on standing seam. A metal roof that's faded and chalky may still have years of structural life, so judge it on the seams, fasteners, and any actual perforation rather than on color. Hail dents metal cosmetically far more readily than it functionally damages it, which matters for storm conversations: cosmetic denting is real but is a different discussion than a compromised roof.

Tile (concrete and clay)

Tile fields can last a very long time, but the tiles aren't usually the failure point, the underlayment beneath them is. From the ground you're looking for cracked, slipped, or broken tiles, sections that have been walked and shattered, and any sign of underlayment failure (leaks reported, staining at eaves). A tile roof can look intact from the street while the underlayment that actually keeps water out has aged past its life. Storm-wise, hail can crack tiles and the cracks aren't always obvious from below. Tile reads require more on-roof confirmation than asphalt, so weight ground reads cautiously and lean on records for the install era.

Wood shakes and shingles

Wood ages through splitting, curling, cupping, rot, and moss/algae growth, and is far more sensitive to moisture and shade than asphalt. A wood roof going gray and splitting with moss in the shaded sections is a strong age signal. Wood also carries fire-code considerations in many jurisdictions, which sometimes forces a material change at replacement, useful context for the homeowner conversation.

Slate

Genuine slate is a multi-generational material; the slates outlast the fasteners and flashing that hold them. On a slate roof you're watching for slipped or missing slates, delamination and flaking (more common with softer slates), and failed flashing in valleys and at penetrations. Never judge slate on an asphalt clock and never let an untrained canvasser onto a slate roof, it's brittle, dangerous to walk, and easily damaged.

The practical takeaway: identify the material first, then apply the right clock. A neighborhood read that lumps a tile or metal roof in with the asphalt cohort will misjudge it badly. Most of your volume will be asphalt, but knowing the other clocks keeps you from embarrassing misreads on the mixed streets.

Part 8: The data sources that power this, and what each one is good for

If you're going to systematize identification, it helps to know exactly which public and commercial sources feed each layer, what they cost in effort, and where each one lies to you.

County assessor / property appraiser (year built, parcels)

The backbone. Nearly every U.S. county publishes parcel records with year built, free, online or by request. It's your build-vintage anchor and the input that segments a territory into aging cohorts. Caveats: year built reflects the structure, not the roof, so it's a floor age for never-reroofed homes and tells you nothing about a home that's been done. Data quality and access vary by county; some offer bulk downloads, others only per-parcel lookups.

Building permit records (reroof dates)

The best signal for subtracting already-done homes. Where jurisdictions publish reroof permits, a present permit gives you a near-actual install date for that address. The weakness is coverage: unpermitted work is common, especially repairs and rural jobs, so a missing permit doesn't prove a roof is original. Use permits to confidently remove homes, not to confidently keep them.

Aerial and satellite imagery (condition, re-roof, patches)

Top-down (orthographic) and oblique (angled) imagery let you read condition at scale. Public options exist (USGS EarthExplorer and various government orthoimagery programs), but the freshest, highest-resolution imagery usually comes from commercial providers. The two things to check on any imagery source are resolution (can you actually see granule wear and patches?) and capture date (how many years of aging and how many storms are you missing?). Obliques beat top-down for reading slope texture and curling. Treat imagery as triage; confirm marginal calls on the ground.

Storm and hail data (NOAA SPC, commercial swaths)

NOAA's Storm Prediction Center publishes storm reports and the NWS documents severe events, free historical baselines for where storms occurred. Commercial providers refine this into hail-swath and wind maps with finer resolution and per-property estimates. The universal caveat: any region-level storm layer overstates how many roofs were actually affected, because hail and wind are patchy. The value is in narrowing from region to roof, and even then it's odds, not proof.

Putting the layers together

No single source identifies aging roofs on its own. Year built without permits over-counts (it keeps the already-done homes). Permits without imagery under-counts (it misses unpermitted reroofs and can't read condition). Imagery without records can't tell a fresh-looking new roof from a fresh-looking recent reroof you should still avoid. Storm data without per-roof modeling paints whole zips. The identification works because the layers cover each other's blind spots, and that's also exactly why assembling and maintaining all of them by hand is the part most contractors never quite get to. It's a real project to stand up and keep current, which is the gap a tool like RoofPredict exists to fill.

Common mistakes that cost crews their day

A short list of the errors that show up again and again, even with experienced teams:

1. Knocking every door because the neighborhood "looks old." A subdivision can be old and still have half its roofs already replaced. Without subtracting the reroofs, you spend the day on dead doors.
2. Treating algae streaks as age. The most common rookie misread. Streaking is microclimate, not service life.
3. Calling a single install date instead of a range. You don't know the exact date from outside, and acting like you do erodes trust and leads to bad prioritization. Range plus condition.
4. Treating a storm as uniform across a zip. Hail and wind are patchy. "The storm passed over here" is not "this roof is damaged," and canvassing a whole swath as if it were uniformly hit wastes most of the effort.
5. Skipping the golden-hour drive. Curling and surface deformation hide at noon and jump out in low-angle light. A daytime scan undercounts worn roofs.
6. Ignoring orientation. South and west slopes age faster. Judge a roof by its worst slope, and don't assume a clean north slope means a healthy roof.
7. Letting imagery go stale. A two- or three-year-old aerial misses recent aging and any recent storm. Know how fresh your source is before you trust it.
8. Confusing targeting with proof. Age ranges and modeled storm exposure tell you where to look. They never substitute for the on-roof inspection and documentation that stand behind a homeowner conversation.
9. No de-duplication against prior work. Knocking a homeowner you (or a competitor) just reroofed last year is a credibility hit. Filter what's already been done.
10. Working without a route. Even a perfect list wastes hours if reps backtrack across a subdivision. Sequence the doors.

A field checklist you can hand a crew

Print this and put it in the truck.

Before the day:

• Territory defined and build-vintage cohorts identified
• Recently-reroofed homes subtracted (permits / fresh imagery)
• Storm layer reviewed per roof where relevant, rather than per zip
• Addresses ranked and sequenced into routes
• Reps briefed on tiers and openers

At each door (ground read):

• Granule loss: matte/bald slopes, granules at downspouts?
• Curling / cupping / clawing across the field (check in raking light)?
• Missing, cracked, or creased shingles, especially windward slopes?
• Sagging ridge or wavy deck?
• Rusted flashing, cracked pipe boots, rust streaks?
• Patches / mismatched shingles (prior repairs)?
• Algae streaks present, noted but NOT counted as age
• South/west slopes judged as the worst case

Decision:

• 3+ high-weight cues, or strong storm exposure, knock now
• 1-2 cues, in age window, offer inspection, let condition decide
• Clearly fine / recently done, skip or referral ask

After contact:

• Roof inspected and condition documented before any options/claims talk
• Outcome logged (due / already done / storm-damaged / fine) for feedback

Safety note

When identification turns into inspection, the work moves onto ladders and roofs, where falls are the leading cause of death in construction. Follow fall-protection rules, use proper ladder setup, stay off roofs in wet, windy, or icy conditions, and never let a canvasser get on a roof they're not trained and equipped to be on. A great knock list isn't worth an injury. Much of the read in this guide is deliberately from-the-ground for exactly that reason: you can qualify the overwhelming majority of doors without leaving the driveway.

Putting it together

Identifying aging roofs in a neighborhood comes down to two reads working together. The condition read, granule loss, curling, creasing, cracking, sagging, the failing details, tells you how worn a roof is, and you can do most of it from the ground with binoculars and good light. The calendar read, build vintage, re-roof tells, permits, material era, tells you where the roof sits on its clock, always as a range, never a single date. Layer storm history on top, per roof rather than per zip, and you can sort an entire territory into the roofs that are aging out, the roofs a storm actually wore out, and the roofs you should leave alone for now.

The crews that win don't knock harder. They knock the right doors. They subtract the homes that have already been done, they weight the worn and the storm-worn to the top, they sequence the route, and they send a trained eye to confirm and document what the data flagged. That's the whole discipline: spend your knocking hours where the roofs are genuinely due, and skip the rest.

If you'd rather not stand up the assessor-permit-imagery-storm pipeline yourself, that's exactly the part RoofPredict handles, an age range per address from aerial imagery, storm modeled on each roof rather than just where it passed, sorted into a knock list you can work, in territory you choose. It points your crews at the right driveways. What's on the roof, and what you do about it, is still yours to inspect, document, and earn. See how it ranks a neighborhood at roofpredict.com.

FAQ

How can I tell how old a roof is without getting on it?

You can't read an exact install date from the ground, so estimate a range instead. Start with the home's year built from county assessor records (a roof can't be older than the house). Then look for re-roof tells, fresh color, modern architectural shingles on an older home, crisp ridge caps and new pipe boots, and check permit records for a reroof permit. Combine that with visible wear (granule loss, curling, cracking) to land on something like '18 to 23 years old and near the end of its life' rather than a single date.

What are the clearest signs from the ground that a roof needs replacing?

The strongest visual cues are heavy granule loss (matte or bald patches and granules piling at downspouts), curling/cupping/clawing across whole slopes, cracked or creased shingles, and any sagging ridge or wavy deck. Failing details, rusted flashing and cracked pipe boots, are a useful early flag. Look in low-angle golden-hour light, which throws shadows off curled edges that hide at midday. Three or more of these high-weight cues on one roof means it's worth a closer look.

Do black streaks mean a roof is old and needs replacing?

Usually not. Black streaks are typically Gloeocapsa magma algae, which is largely cosmetic and driven by shade and humidity, not age. You'll see badly streaked 8-year-old roofs in damp, shaded lots and clean 20-year-old roofs in full sun. Treat streaking as information about the microclimate, not the remaining service life, and judge age by wear cues like granule loss and curling instead.

How does the build year of a neighborhood help me find old roofs?

Tract builders roof whole phases within a year or two of construction, so a subdivision's original roofs age as a cohort and reach the end of their service window together. If you find a development built right at the front edge of an asphalt roof's typical service life, a large share of homes that were never re-roofed are due around the same time. Pull year built from the county assessor to segment a territory into build-vintage cohorts and spot which neighborhoods are entering their replacement window now.

Why does a storm map alone send crews to the wrong houses?

Hail swaths are narrow and patchy and wind hits some slopes while skipping others, so 'this zip got hail' covers a lot of roofs that weren't meaningfully affected. Canvassing a whole swath as if it were uniformly hit wastes most of the effort. The fix is going from where the storm passed to an estimate of what each individual roof likely experienced based on the storm's path and that roof's exposure, then confirming with an on-roof inspection. A storm passing over an address raises the odds of damage; it isn't proof of it.

What is the fastest way to find old roofs across a large area?

Use a records-and-aerial workflow instead of driving every street. Define the territory, pull year-built data to find aging cohorts, overlay reroof permits to subtract homes already replaced, read roof condition from high-resolution aerial and oblique imagery, add storm history per roof, and score every address into a ranked knock list. Then send trained reps to confirm the top doors on the ground. This lets you triage a thousand homes before lunch and spend knocking hours only where roofs are genuinely due.

How accurate is estimating roof age from aerial imagery?

Aerial imagery can reliably place a roof in a likely age band by reading color uniformity, wear, patches, and whether it's a fresh re-roof, and obliques add slope texture that top-down shots miss. It's a strong prioritization signal, not a guarantee, you'll still find occasional surprise re-roofs, so the honest output is an age range per address rather than an exact date. Imagery freshness matters too: a two-year-old image misses recent aging and storms, and marginal calls should be confirmed on the ground.

What does RoofPredict actually do, and what doesn't it do?

RoofPredict estimates roof age house-by-house from aerial imagery (as a range per address) and models storm physics per roof rather than just where the storm passed, then sorts a neighborhood into a ranked knock list of the roofs aging out and the roofs a storm likely wore out. What it doesn't do: hand you pre-qualified leads, give an exact install date, or prove a specific roof is damaged. The age is a range and the storm exposure is odds, so your reps still inspect, document, and earn the conversation. It points crews at the right driveways in territory you choose to work.

Should I knock every door in an old neighborhood?

No. An old subdivision often has a large share of homes that were already re-roofed and look fine, and knocking those is wasted time, or a credibility hit if you reroofed one recently. Subtract the recently-done homes using permits and fresh imagery, weight the visibly worn and storm-exposed roofs to the top, sequence the survivors into an efficient route, and work that list. A rep working 40 pre-qualified doors in a tight loop out-produces one knocking a whole subdivision.

Why express roof age as a range instead of a specific date?

Because from the outside you genuinely don't know the exact install date, and acting like you do leads to bad prioritization and erodes homeowner trust. Build year, material era, and wear give you an error-banded estimate, and the roof's condition is what actually determines whether it's due. A read like 'most likely 18 to 23 years old and showing heavy wear, so it's near the end' is honest and is all you need to decide whether to knock. Save exact dates for permits or receipts, the only places they actually live.