How to Spot an Old Roof From the Street (a Roofer's Read)

You can read a roof from the street if you know what you are looking at. Most people can't. A homeowner sees "the roof" as one gray thing that either leaks or doesn't. A good roofer pulls up to a curb, looks for maybe eight seconds, and walks away with a working estimate of how old the covering is, whether it has taken a beating, and whether the house is worth a knock. That skill is not magic and it is not a gift. It is a checklist that has been run so many times it stopped feeling like a checklist.

The problem is that the checklist most reps carry around is half wrong. They look for the dramatic stuff — a tarp, a visibly caved valley, three-tab shingles curling like potato chips — and they miss the quiet tells that separate a roof that is genuinely worn out from one that just looks tired in bad light. They knock the obvious houses and skip the ones that are actually due. Worse, they confuse "old-looking" with "old," which are not the same thing, and they get burned both directions: they pitch a homeowner whose roof was replaced four years ago, and they drive past a fifteen-year-old laminate that is two summers from failing.

What follows is the full read. Every visual sign of an aging roof that you can actually catch from a truck or a sidewalk, ranked by how reliable it is, with the failure modes spelled out so you don't get fooled. Then the part nobody writes down: how to turn a curbside read into a route, how to tell age from damage, and how to stop running the whole street when only a third of it is worth your time.

Why the street read matters more than anything else you do

Start with the economics, because they explain why this skill pays. A door-knocking rep working a residential street is spending the most expensive thing a roofing company has, which is a closer's hours, on doors. Payroll, gas, the opportunity cost of the conversations that didn't happen — every door you knock has a real per-door cost, and most companies have never actually calculated theirs. Run the math once and it changes how you work.

Say a rep knocks forty doors in a working afternoon. Industry door-to-conversation rates on cold residential are brutal — a large share of homes have nobody home, and of the ones that answer, most are not in the market. If even a few of those forty roofs are genuinely due, you have spent the entire afternoon's payroll finding them by brute force. Now flip it. If you could look at the same forty houses and know that eleven of them have a covering old enough to be a real conversation, you knock eleven and spend the rest of the afternoon on the next street. Same payroll, three to four times the at-bats on roofs that can actually become jobs.

The curbside read is the cheapest, fastest filter you own. It costs you eight seconds and zero dollars per house. It is not perfect — you will misjudge some — but it is the difference between working a street and working a list. Every other targeting tool, every piece of software, every mailing campaign, is ultimately trying to reproduce at scale what a sharp rep does instinctively from the truck. So learn to do it instinctively first.

There is a second reason it matters that has nothing to do with knocking. When you can read a roof from the street, you can read it while the homeowner is standing next to you, and you can narrate what you see in plain language that builds trust instead of triggering the "here comes a sales pitch" reflex. "See how the edges of those shingles are starting to lift and the granule's washing into your gutters? That's the roof telling you it's getting close." That is not a pitch. That is a man reading a roof, out loud, and being right. It closes.

The first thing to settle: age versus damage

Before any signs, get this distinction straight, because almost every rep blurs it and it costs them.

A roof can look bad for two completely different reasons. It can be old — the asphalt shingle has simply run out its service life, the binders have dried, the granules have shed, the mat underneath is going brittle. Or it can be damaged — a hailstorm bruised it, a windstorm creased and lifted it, a tree limb gouged it. These produce different visual signatures, they happen on different timelines, and they lead to different conversations.

Age is gradual and uniform. The whole roof field degrades together, roughly evenly, because the whole roof saw the same sun and the same years. Damage is patchy and directional. Hail hits hardest on the slopes facing the storm's approach and on the soft accessories — vents, the metal, the ridge. Wind lifts and creases the edges, the rakes, the field shingles that caught an updraft, and it leaves the protected slopes alone.

Why this matters on the street: an old roof gives you an age conversation ("this covering is at the end of its run, let's talk about a replacement plan"). A damaged roof gives you a storm-and-documentation conversation, which runs through the homeowner's insurer and follows very different rules. You can spot both from the curb, but you pitch them differently, and you never want to walk up to a freshly storm-bruised roof and start talking about it like it's just old, or walk up to a tired old roof and start implying a storm did it when it didn't. Read which one you're looking at before you open your mouth.

The cleanest tell: is the wear even or is it directional? Even, all-over wear is age. Concentrated damage on the storm-facing slopes with clean slopes elsewhere is impact. You can often see the difference from the street by walking to where you can view two different slopes.

The reliable signs, ranked

Not all visual cues are worth the same. Some are nearly diagnostic on their own; some only mean something stacked with two or three others. Here they are, ordered by how much weight a single sighting deserves.

Tier 1: signs that, by themselves, tell you a lot

Cupping and curling shingle edges. This is the single most reliable age tell on an asphalt roof you can read from the ground. As asphalt shingles age, they lose the oils that keep the mat flexible, and they start to deform. Curling is when the edges of each shingle lift up and away from the deck — the tab corners turn up. Cupping is the opposite warp, where the center rises and the edges hold down, often from moisture cycling. Both throw shadows. On a sunny day with the sun at a low angle, a curling roof looks like the whole field is covered in tiny shadow-lines that a flat new roof doesn't have. The field loses its smooth, sheet-like look and gets a texture, almost a shimmer of little raised edges. Once you've seen it on a known-old roof you can't unsee it. Curling is a late-life sign. A roof that is visibly curling across the field is usually past 15 years and frequently past 20, and it is a roof to knock.

Bald spots and color mottling from granule loss. Asphalt shingles are protected by a layer of mineral granules — that's the colored grit. The granules are the sunscreen; they block UV from cooking the asphalt underneath. As a roof ages, it sheds granules, and where they're gone you see the darker, shinier asphalt mat showing through. From the street this reads as mottling: the roof field looks blotchy and uneven in color, with darker patches, instead of a clean uniform tone. Bad cases look almost two-toned. New roofs are uniform. Splotchy, patchy, two-tone color is a roof shedding its protection. (Be careful: not all dark streaking is granule loss — see the algae trap below.)

A bowed, wavy, or rippling roofline. Sight along the ridge and the eaves like you're checking a board for straightness. A healthy roof deck is flat and the lines are crisp. When you see waves, dips, sagging between rafters, or a ridgeline that bows like a swayback horse, you're looking at something deeper than worn shingles — that's the deck, the structure, or long-term moisture, and it usually rides along with an old covering. A sagging roofline is a Tier 1 "this house is a real conversation" sign. It also flags a bigger job (possible decking replacement), which is worth knowing before you knock.

Tier 2: strong signs that confirm when stacked

Granules in the gutters and at the downspout splash. You often can't see this from the street directly, but you can frequently see the result: a dark, gritty wash stain on the driveway or sidewalk below a downspout, or a visible line of granule sediment in an open gutter when you're at the right angle. Where the downspout dumps, look for a little delta of black sandy grit. That grit is the roof's sunscreen, washed off and piled up. A roof actively shedding measurable granules is a roof in the back third of its life.

Worn, lifted, or missing accessories. The roof field is only part of the story. Look at the accessories: the metal flashing in the valleys and around the chimney, the plumbing vent boots (the rubber or lead collars around the pipes sticking out of the roof), the ridge cap, the drip edge. These age faster and uglier than the field. Rubber pipe boots dry-rot and crack — from the street a failed boot reads as a dark, cracked collar or a visibly slumped fitting around a vent pipe. Rusted, streaked flashing tells you the metal's been up there a long time. A roof with three sun-rotted vent boots is an old roof, full stop.

Patchwork and color mismatch. Scan for sections that don't match. A rectangle of slightly different-colored shingles, a repaired valley, a stretch of newer cap over older field — these are patches, and patches tell a story: this roof has already started failing somewhere, somebody's been nursing it, and the homeowner has been spending money to avoid the conversation you're about to have. Patchwork is a green light. It means the roof has a known history of problems and the owner is already half-convinced.

Exposed or deteriorating underlayment / fastener heads showing. On a really tired roof you'll sometimes see shingles that have eroded enough that the nail line or the underlayment beneath is starting to show as a different texture or sheen. Hard to catch from far, easy to confirm with a zoom on your phone camera. Means the covering is failing outright, beyond simply aging.

Tier 3: weak signals — only count them when they pile up

Dark streaks and stains. This is the famous trap. Those black or dark-green streaks running down a roof, especially on the north-facing slopes, are usually algae (commonly Gloeocapsa magma), not age and not damage. Algae is mostly cosmetic. A roof can be black with algae streaks and still have years left, and a clean roof can be at death's door. Reps lose all the time by reading algae as "old." Use streaking only as a faint hint that the roof is not new, never as proof it's done. The exception: when streaking sits on top of curling, mottling, and bald spots, it's part of a consistent old-roof picture.

Moss. Moss is a step beyond algae. Where algae is a surface stain, moss is a plant with a root structure that holds moisture against the shingle and lifts the edges, which accelerates real damage. Heavy moss, especially in damp and shaded climates, is a stronger signal than algae because it both indicates age and actively worsens the roof. Still, treat it as Tier 3 on its own and Tier 2 when it's clumping thick and lifting shingles.

General "tired" appearance. Faded color, a flat dull finish instead of the slight sheen new shingles have, a roof that just looks dusty and dark. Real, but soft. Don't knock on vibes alone.

Why two identical roofs age at different speeds

Before the material guide, internalize this, because it's why "the calendar" alone fools people: two roofs of the same shingle, installed the same year, two streets apart, can be years apart in real condition. The drivers:

Sun exposure and slope orientation. UV is what cooks asphalt. A south- and west-facing slope in a hot, sunny climate bakes for more hours a day than a north slope, and it ages visibly faster — you'll often see one slope of a roof curling and mottled while another slope of the same roof still looks decent. When you read a house, read the slope you can see and remember it may not represent the whole roof. This is also why a desert or southern roof at year twelve can look like a northern roof at year twenty.

Ventilation. A poorly ventilated attic traps heat against the underside of the deck, and that heat ages shingles from below, fast. You can't see ventilation from the street, but you can sometimes infer it: ridge vents, gable vents, and roof vents tell you the system breathes; their absence on a hot-climate home is a reason an otherwise mid-life roof looks older than its years.

Install quality and shingle grade. A roof nailed correctly with quality laminate over good underlayment outlasts a builder-grade three-tab slapped on to flip a house — by a wide margin. You can't grade the install from the curb, but the shingle type is a proxy: three-tab skews shorter life, premium laminate longer.

Trees and debris. Overhanging limbs drop debris that holds moisture, feed algae and moss, and physically abrade shingles in wind. A roof under heavy canopy ages rougher and looks older, sometimes misleadingly so.

The practical upshot: treat every life-span number as a range that shifts with climate, slope, and ventilation, and let the visual signs override the calendar in both directions. A "should-be-fine" twelve-year-old south-facing roof in Phoenix that's already mottling is old; a twenty-year-old shaded north roof in Oregon that's still crisp may have life left.

A field guide by roofing material

The street read changes completely depending on what the roof is made of, and a rep who only knows asphalt will misjudge half the houses on an older block. Here's how to age the common materials by eye.

Asphalt shingles (the majority of American homes)

Everything above applies. The two sub-types matter:

• Three-tab shingles (flat, uniform, with the two cutouts giving three "tabs" per strip) are the older, cheaper style. If a roof is three-tab, you're already looking at either an older home or a builder-grade installation, and three-tabs typically carried shorter service ratings. Three-tabs that are curling are gone. The presence of three-tab shingles at all skews the odds toward "older roof, knock it."
• Architectural / laminate / dimensional shingles (thicker, with a layered, shadowed, shake-like look) are the modern standard and last longer. They hide age better — they can look decent from the street longer than a three-tab — so you lean harder on granule mottling and edge lift, and you respect the calendar more. A laminate roof that's clearly shedding granules and mottling is genuinely old even if it still reads "okay" at a glance.

Rough service-life framing to carry in your head: typical asphalt shingle roofs in the field are generally cited in the 15-to-30-year range depending on type, quality, ventilation, and climate, with three-tabs at the lower end and premium laminates at the upper. Treat these as ranges, not promises — a south-facing slope in a hot, sunny climate ages years faster than the same shingle on a shaded north slope. Never state a roof's age as a fact from the street. You're estimating a window.

Wood shake and shingle

Splitting, cupping, dark gray weathering, missing pieces, and moss are the tells. Wood goes gray and brittle with age, individual shakes split and slide, and moss thrives on it. A wood roof that's gone uniformly dark gray with visible gaps and missing shakes is old and likely a fire-code and insurance headache for the owner — which is a conversation. Wood is a smaller slice of the market and skews older homes.

Tile (clay and concrete)

Here's the trap that costs reps real jobs: tile itself can last 50-plus years, but the underlayment beneath it does not. The clay or concrete tiles you see from the street might be a century from failing while the felt or membrane under them — the actual waterproofing layer — has dried out and is leaking. So you can't read a tile roof's true condition from the color of the tile. What you can spot from the street: cracked, slipped, or missing tiles (dark gaps in the field), and a visibly aged, faded finish. A tile roof with missing pieces and an obviously dated look on an older home is worth a knock not because the tile is shot but because the system underneath probably is.

Metal

Standing-seam and metal panel roofs age by finish failure and fastener issues: chalky, faded paint; rust streaks; oil-canning waviness; lifted seams; and on screw-down panels, the rubber washers under the exposed fasteners dry out and the panels start to lift or leak at the screws. Metal lasts a long time, so age alone is rarely your angle; look for visible corrosion and seam problems.

Flat / low-slope (membrane)

On homes with flat sections, you're looking at modified bitumen, EPDM rubber, or TPO. From the ground you often can't see the membrane field at all, but you can sometimes catch ponding stains, patched areas, blistering, and aged, cracked perimeter metal. Flat roofs have shorter typical lives and fail at the seams and penetrations.

Reading age you can't see: the house tells on the roof

Some of the best clues to roof age aren't the roof at all. They're the rest of the house, and they let you estimate even when the roof itself is hard to read.

The age of the home. A roof's life is finite, so the home's vintage sets the baseline odds. A house built in the 1990s on its original roof is overdue. A neighborhood of homes built in the same era will tend to need roofs in clusters, because they were all built — and last re-roofed — around the same time. This is why whole streets "come due" together. You can often eyeball a home's era from its style, windows, and trim; you can confirm it from public records (more on that below).

Tract neighborhoods age in waves. A subdivision built out over two or three years will hit roof-replacement age over the same two or three years, fifteen to twenty-five years later. Drive a tract where one or two roofs have obviously already been redone and you're looking at the leading edge of a wave — the rest of that street is right behind. That clustering is one of the most useful patterns in residential roofing and most reps never name it.

The neighbors already re-roofed. If three houses on a block have crisp, new, uniform roofs and the one in the middle still has the original, the middle one is both the most likely to be due and the most likely to convert, because the owner has watched neighbors do it and the social proof is already built. A new roof next door is a sales asset for the old roof beside it.

Other deferred maintenance. Peeling paint, an old fence, a dated garage door, a yard that's maintained but not lavish — the roof is part of a maintenance pattern. A house where everything is fifteen years past its refresh probably has a roof to match. Conversely, a freshly renovated house often has a freshly done roof; don't waste the knock.

Don't get fooled: the false positives that burn reps

Reading roofs is as much about not knocking as knocking. Here are the traps that send reps to the wrong doors.

Algae streaking on a young roof. Covered above — the number one false positive. Dark streaks alone mean almost nothing about age. Many modern shingles even have algae-resistant granules that streak less, so a clean roof can be old and a streaked roof can be new.

Bad light and bad angles. A roof in flat overcast light or backlit against a bright sky can look uniformly dark and "tired" when it's fine, or hide curling that a low-angle sun would reveal. Reps misjudge roofs at 2 p.m. under high sun and at dusk constantly. When in doubt, note the house and re-look from a different angle or time, or zoom in.

A new roof that looks old because it's a cheap shingle or a dirty one. Builder-grade three-tabs and roofs under heavy tree cover collect debris and look rough fast. Dirty is not old.

An old roof that looks new because it was just cleaned or is premium laminate. A pressure-washed or soft-washed roof loses its algae and reads younger than it is. Premium laminates hide age. This is the false negative — the roof you wrongly skip — and it's why the calendar (home age, records) matters as a backstop to the eyeball.

Solar panels. Panels cover a chunk of the field and make the roof hard to read, and they complicate any re-roof (panels have to come off and go back on). Note them; they change the job.

Recent partial work. A roof with one new slope and three old ones — common after a storm claim that only covered the damaged elevation — can read as "newish" from the wrong angle and "old" from another. Walk the corner.

The curbside read, as a repeatable workflow

Turn the signs into a routine you run the same way every time. Here's the sequence a sharp rep runs in well under a minute per house, often in eight to ten seconds once it's automatic.

1. Material and type. Asphalt three-tab, asphalt laminate, tile, metal, wood, flat? This sets which signs apply and the baseline life. (2 seconds.)
2. Field texture and color. Smooth and uniform, or textured with curling shadow-lines and mottled, patchy, two-tone color? Curling + mottling = old. (3 seconds.)
3. Lines. Sight the ridge and eaves. Straight and crisp, or wavy, sagging, bowed? Wavy = deeper problem, bigger job. (2 seconds.)
4. Accessories. Vent boots, valley and chimney flashing, ridge cap, drip edge — crisp or sun-rotted, rusted, cracked? Rotted accessories = old. (3 seconds.)
5. Patches and mismatch. Any sections that don't match? Patchwork = known history, warm door. (2 seconds.)
6. House context. Home vintage, neighbor roofs (any freshly redone?), overall maintenance level. Sets the odds. (3 seconds.)
7. Age vs. damage check. Is the wear even (age) or concentrated on storm-facing slopes (damage)? Decides which conversation. (Variable.)

Then the call: knock, skip, or flag for a second look. Knock the clear olds and the warm-context houses. Skip the obvious news and the renovated ones. Flag the ambiguous ones — bad light, premium laminate, can't see two slopes — for a record check or a return pass.

Here's the thing reps undervalue: keep notes. A rep who walks a street and logs "knocked the curled-laminate ones, skipped the eight new roofs, flagged four I couldn't read" has built a real working list of that block. Do that across a few streets and you have a route. Most reps run the same street twice because they kept nothing.

A worked example

You pull onto a 1998-built cul-de-sac, twenty-two homes. From the truck: four houses have obviously new, uniform, crisp laminate roofs — skip, recently redone. Twelve have original-looking architectural shingles. Of those twelve, seven show clear granule mottling and edge curl with a couple of sun-rotted vent boots — these are at the front of your list, knock them. Three are streaked with algae but otherwise smooth-fielded and crisp-lined — flag, possibly just dirty, re-look in better light or check records. Two are heavily mossed with visible patchwork in a valley — top of the list, knock first, the owner's already been spending money. The remaining six are tile — note that the tile read is about underlayment age, and given a 1998 build the underlayment is plausibly original and due; warm, knock the ones with any slipped tiles.

You just turned a "knock all twenty-two" Saturday into roughly nine strong knocks and a handful of flags, on a street whose 1998 vintage already told you a wave is breaking. That's the whole skill, in one cul-de-sac.

From eyeball to records: confirming the age window

The street read gives you a window. Public and semi-public records can tighten it, and you should know what each source actually tells you — because the most common mistake reps and homeowners both make is treating "year built" as "roof age."

Year built is not roof age. A county assessor record or a real-estate listing tells you when the house went up, not when the roof was last replaced. A 1985 house could be on its third roof or its first. Year built sets a ceiling on roof age and a baseline for the odds — useful — but a re-roof is usually invisible to these records. Don't pitch off year built as if it were roof age.

Permit records sometimes catch re-roofs. Many jurisdictions require a permit to re-roof, and where they do, a permit search can reveal the date of the last re-roof — the closest thing to a real "roof birthday" you'll find in public data. Coverage is wildly inconsistent: some areas pull permits for everything, some homeowners and even some contractors skip the permit, and the records aren't always online or searchable. When it's there, it's gold. When it's not, you're back to the window.

Listing photos and history. If a house sold recently, the listing history and old photos can show what the roof looked like at the time of sale, sometimes with the agent noting a "new roof." Past listing photos from a few years back, compared to today, can show whether a re-roof happened in between.

The homeowner is the real record. Nothing beats "when did you last do the roof?" at the door. Half of them don't precisely know, which is itself useful, but they'll often say "the previous owners did it" or "right after we moved in, that was 2011." Your street read plus that one answer pins the window.

The honest takeaway: there is no clean public database of roof ages. You triangulate — eyeball window, home vintage, any permit you can find, listing history, neighbor pattern — into a confident range. Anyone who tells you they can give you an exact roof age from a desk is overselling. A tight range is what's real and a tight range is enough to decide whether to knock.

Doing this at scale without burning your week

Everything so far is the manual craft, and you should know it cold. But there's an obvious ceiling: a human can only drive so many streets, and reading roofs one truck at a time doesn't scale past a small crew working nearby blocks. The skill that wins on a cul-de-sac doesn't, by itself, tell you which cul-de-sacs across a whole service area are full of due roofs. You end up driving good neighborhoods and bad ones alike to find out.

This is the gap where aerial imagery and modeling earn their keep. The same visual signals you read from a truck — and the home-vintage and storm-history context you triangulate by hand — can be estimated from overhead imagery across an entire area at once, so instead of driving to find the wave, you start the day already knowing which streets the wave is breaking on.

A quick, honest note on what the off-the-shelf tools each actually do, because reps conflate them:

• Zillow, county records, Google give you year built, not roof age. Re-roofs are invisible to them, as covered above.
• EagleView, HOVER, Roofr and the measurement tools measure a roof — squares, pitch, facets — for estimating and ordering. They're excellent at that. They don't tell you which roof is old or which house to knock. Different job: "measure this house" versus "which house."
• Hail and storm maps show you where a storm passed — a swath on a map. They don't tell you which individual roofs in that swath actually got worn out, because a storm's energy lands unevenly house to house.

Where RoofPredict fits

RoofPredict sits in the gap those tools leave: it estimates, from aerial imagery, a roof-age range per address across your whole area, and it models storm physics — hail and wind — on each roof, rather than only where the storm passed, so you can see which specific houses are most likely worn out and rank a street before anyone gets on a ladder. It's the curbside read, run across thousands of roofs at once, with the storm history layered on top.

Be clear about what that is and isn't. It is a way to decide which doors to knock and which to skip — to point a finite crew at the roofs that are actually due instead of the whole ZIP. It is not a lead service; it doesn't sell you a homeowner who raised their hand, and it doesn't resell the same contact to five competitors. The roof age is a range, the same window you'd build by eyeball and records, not an exact date — because an exact date isn't real, and we won't pretend it is. The storm modeling gives you odds that a given roof took a beating, which is a reason to go look, not a substitute for getting up there and documenting what's actually there.

And it stays in its lane on storms and claims: the model and the imagery help you decide where to knock and give you something concrete to talk through at the door, but the roofer is the one who documents the conditions and writes the estimate, the insurer decides coverage, and the homeowner owns their claim. The data points your crew at the right houses faster than driving every street; it doesn't file anything, promise anything, or get on the roof for you. Used that way — as a sharper version of the read you already do — it turns "knock the whole subdivision" into "knock the eleven roofs the imagery and the storm history flagged," which is the same move the cul-de-sac example made, at the scale of a service area.

Turning a read into a conversation that closes

A street read is only worth something if it changes what happens at the door. Tie the two together.

When your read says a roof is genuinely old, lead with what you saw, not with what you sell. "I do roofs in the neighborhood and I noticed the shingles on your place are starting to curl along the edges and you've got some granule washing down by the downspout — that's pretty normal for a roof this age, but it means you're getting close. Mind if I take a quick look so you actually know where you stand?" You've demonstrated competence, you've been specific, and you've offered information before asking for anything. That is a different reception than "we're doing roofs in the area."

When the read is a storm-damage pattern — directional, concentrated on the storm-facing slopes — the conversation shifts to documentation. You're there to look at the roof, photograph the conditions, and put together an honest assessment of what's there; what happens with the insurer is the homeowner's process, on their claim. Keep that line clean. Reps who blur it — implying they'll "get the roof approved" or talking deductibles — create problems for the homeowner and the company. Document, estimate, hand it over. Stay in your lane and you stay trusted.

When the read is ambiguous or the homeowner insists the roof is fine, the inspection is the close. "Totally fair — from the street I can only see so much. The only way either of us knows for sure is to get up there. No charge to look, and if it's got years left I'll tell you that and you can stop thinking about it." You've reframed the inspection as settling the question, which a homeowner with a fifteen-year-old roof actually wants settled.

Common mistakes pros still make

Even experienced reps leave money on the table here. The recurring ones:

Reading algae as age. Said three times now because it's that common. Dark streaks are mostly cosmetic. Stop knocking streaked young roofs and stop skipping clean old ones.

Knocking off "year built." Year built sets odds, not roof age. The re-roof is invisible. Confirm the window before you treat a house as due.

Working the whole street. The entire point of the read is to not knock everything. Reps who knock all forty doors because reading roofs feels like extra work are spending three times the payroll for the same jobs.

Only catching the dramatic. Tarps and caved valleys are obvious and rare. The money is in the quiet curl-and-mottle roofs that look passable but are two summers from failing, and in the warm-context houses — patchwork, mossy, neighbors-already-redone — that most reps drive past.

Pitching age as damage or damage as age. Read which one you're looking at first. Even wear is age. Directional wear is storm. They're different conversations with different rules.

Trusting a single sign. One curling shingle, one streak, one dark patch proves little. The read is a stack — material, field texture, lines, accessories, context — and the confidence comes from several signs agreeing.

Keeping no notes. A read you don't record is a street you'll re-walk. Log knock/skip/flag per house and you build a route instead of repeating a Saturday.

A printable checklist

Run this from the truck, top to bottom, per house:

• Material/type identified (asphalt three-tab / asphalt laminate / tile / metal / wood / flat) — sets baseline life
• Field texture: smooth & uniform vs. curling/cupping shadow-lines (curling = old)
• Field color: uniform vs. mottled/blotchy/two-tone from granule loss (mottled = old)
• Granule wash: black grit stain under downspouts / on driveway (present = back-third of life)
• Roofline: crisp & straight vs. wavy/sagging/bowed (wavy = deck issue, bigger job)
• Accessories: vent boots, flashing, ridge cap, drip edge crisp vs. cracked/rusted/slumped (rotted = old)
• Patchwork/mismatch: any non-matching sections (present = known history, warm door)
• Streaks/moss: noted but weighted low (algae ≈ cosmetic; thick lifting moss = stronger)
• Home vintage: era of the house sets the odds
• Neighbor roofs: any freshly redone nearby (wave-breaking / social proof)
• Age vs. damage: even all-over wear (age) vs. directional storm-facing wear (impact)
• Verdict: KNOCK / SKIP / FLAG-for-second-look
• Logged the verdict for the route

If three or more of the wear signs stack on one house, that's a knock regardless of how it photographs. If the only thing you've got is a streak and a vibe, that's a flag, not a knock.

The bottom line

Spotting an old roof from the street is a learnable, repeatable read: identify the material, look for curling and mottling in the field, sight the roofline for sag, check the accessories for rot, scan for patchwork, and weigh the house's vintage and its neighbors — then separate even age-wear from directional storm-damage so you know which conversation you're walking into. Do that and you stop knocking the whole street. You knock the roofs that are actually due, you walk up able to narrate exactly what you see, and you spend your most expensive hours on the doors that can become jobs.

The eyeball is the foundation, and you should be able to do every bit of it without a single tool. Records tighten the window; a tool that estimates a roof-age range and models the storm on each roof across a whole area lets you skip the part where you drive every street to find the wave. But it all rests on the same read — the one you can now run from a truck in eight seconds, and be right.

FAQ

What is the single most reliable sign a roof is old from the street?

Curling and cupping of the shingle edges in the field. As asphalt shingles lose the oils that keep them flexible, the tab corners lift and warp, throwing little shadow-lines across the roof in low-angle sun. A roof visibly curling across the whole field is usually past 15 years and often past 20. It's more reliable than dark streaking, which is mostly cosmetic algae, and it's a stronger single sign than color alone because it reflects the actual breakdown of the shingle mat rather than mere surface staining.

Are the black streaks on a roof a sign it's old?

Usually not. Those dark or greenish streaks, most common on north-facing slopes, are typically algae (often Gloeocapsa magma), which is largely cosmetic. A roof can be heavily streaked and still have years of life, and a clean roof can be at the end of its run. Many modern shingles even use algae-resistant granules. Treat streaking as a faint hint that a roof isn't brand new, never as proof it's done. It only carries weight when it stacks with real age signs like curling, granule mottling, and bald spots.

Can you tell how old a roof is just by looking at it?

You can estimate a range, not an exact age. The visual signs (curling, granule loss, accessory rot, sag) place a roof in a window, and the home's vintage and neighbor patterns sharpen the odds. But shingles age at different rates depending on type, install quality, ventilation, climate, and which direction a slope faces, so a precise age from the curb isn't real. Anyone claiming an exact roof age from a glance is overselling. A confident range is what's achievable and it's enough to decide whether to knock.

What's the difference between an old roof and a storm-damaged roof from the street?

Age wear is gradual and uniform: the whole roof field degrades roughly evenly because it all saw the same sun and the same years. Storm damage is patchy and directional: hail and wind hit hardest on the slopes facing the storm and on soft accessories like vents and metal, leaving protected slopes relatively clean. From the curb, walk to where you can see two different slopes. Even all-over wear means age (an age conversation). Concentrated wear on storm-facing slopes with clean slopes elsewhere means impact (a documentation conversation that runs through the homeowner's insurer).

Does the year a house was built tell you the roof's age?

No. Year built from county records, Zillow, or a listing tells you when the house went up, not when the roof was last replaced, and a re-roof is usually invisible to those records. A 1985 home could be on its first roof or its third. Year built sets a useful ceiling and baseline odds, but you confirm the actual age window with the visual read, any available re-roof permit, listing history, the neighbor pattern, and ideally the homeowner's own answer at the door.

How do you tell the age of a tile or metal roof from the street?

Differently than asphalt. Tile itself can last 50-plus years, so its color tells you little; what fails is the underlayment beneath it, which you can't see, so you look for cracked, slipped, or missing tiles and an obviously dated home where the underlayment is plausibly original. Metal ages by finish failure (chalky, faded, rust-streaked paint), oil-canning waviness, lifted seams, and dried-out washers under exposed screws. For both, age alone is rarely the angle; visible system failures are.

What roof signs mean a bigger, more expensive job?

A wavy, sagging, or bowed roofline is the big one. Sight along the ridge and eaves; waves and dips between rafters point to deck or structural moisture problems, which can mean decking replacement on top of the covering. Patchwork in valleys, exposed underlayment or fastener heads, and multiple rotted accessories also flag a roof that's failing rather than merely aging. Solar panels add cost too, since they must come off and go back on. Catching these from the street lets you size the conversation before you knock.

How can I find old roofs across a whole area instead of driving every street?

The curbside read doesn't scale past a small crew working nearby blocks, so to find which neighborhoods the replacement wave is breaking on, roofers use aerial imagery and modeling. Tools like RoofPredict estimate a roof-age range per address from overhead imagery across an entire service area and model storm physics (hail and wind) on each individual roof, rather than only where a storm passed, so you can rank a street and pick which doors to knock before anyone climbs a ladder. It's the same read you'd do from a truck, run across thousands of roofs at once. It's not a lead service, the age is a range rather than an exact date, and the storm modeling gives odds a roof was worn out, which is a reason to go look and document, not proof on its own.

Why do whole neighborhoods seem to need new roofs at the same time?

Because tract subdivisions are built out over a short span and then re-roofed on similar timelines, so a neighborhood tends to hit replacement age in clusters 15 to 25 years after construction. If you drive a tract and see one or two roofs already redone with crisp uniform shingles, you're looking at the leading edge of a wave and the rest of that street is right behind. Spotting a freshly re-roofed house next to original roofs is one of the strongest signals on a block, because the neighbor's roof is both a timing clue and built-in social proof.

What false positives make reps knock the wrong doors?

The big ones: reading algae streaks as age (mostly cosmetic), judging roofs in bad light or from a single backlit angle, mistaking a dirty or cheap roof under tree cover for an old one, and missing premium laminates or freshly cleaned roofs that look younger than they are. Recent partial work (one new slope, three old) can read new from one angle and old from another, so walk the corner. The fix is to never trust a single sign; confidence comes from several age signs agreeing, backed by home vintage and records.