How to Know Which Roofs Are Due for Replacement (A Field Guide for Roofing Crews)

Every roofing company has the same finite asset, and it is not trucks or crews or even capital. It is daylight. A canvasser has roughly 40 productive door-knocking hours a week. A sales rep can run maybe six to eight quality inspections a day if the routes are tight and the appointments hold. The companies that win are rarely the ones with the slickest pitch. They are the ones who point those finite hours at the roofs that are actually due, and quietly skip the ones that are not.

That is what knowing "which roofs are due" really means. It is not about predicting the future of a single shingle. It is about ranking a street, a subdivision, or a whole storm footprint by the odds that the roof behind each door is at or near the end of its service life, or has been beaten up enough by weather that an insurer would reasonably pay to fix it. Get the ranking right and your close rate climbs, your cost per acquired job drops, and your crews stop burning afternoons on 9-year-old roofs that have another decade in them.

What follows is the working playbook a sharp estimator or sales manager uses to make that call: what to read from the street, what to confirm on the roof, how to estimate age when nobody knows when the roof went on, how storms change the math, and how to turn all of it into a route your crew can actually run. None of it is magic. It is pattern recognition, a little building science, and a refusal to waste a knock.

What "due for replacement" actually means

Before you can rank roofs, you have to be honest about what you are ranking them against. A roof becomes a replacement candidate for one of three reasons, and they are not the same conversation at the door.

1. It is aging out. The roofing material has reached or passed the point where it no longer reliably keeps water out. For a typical three-tab asphalt shingle roof in a four-season climate, that window is roughly 15 to 20 years. Architectural (dimensional) shingles run longer, often 22 to 30 years in practice, though the printed warranty number is almost always optimistic compared to real-world performance. This is the slow, predictable bucket.

2. It got hurt. A hail event, a windstorm, a fallen limb, or years of poor ventilation cooking the deck from below pushed an otherwise mid-life roof past the point of economical repair. This is the storm-restoration bucket, and it is where age and weather intersect. A 12-year-old roof that ate golf-ball hail is a very different prospect than a 12-year-old roof on a calm street.

3. It is failing for a reason that has nothing to do with age. Bad installation, the wrong product for the slope, chronic ventilation problems, a manufacturing defect, or ice-dam damage in a cold climate. These roofs fail early and they fail visibly. They are gold for a crew that knows how to read them, because the homeowner often has no idea why their five-year-old roof is leaking.

Most of the rest of this comes back to telling these three apart from the curb and confirming them on the deck.

The honest disclaimer up front

Nothing you read from the street or even from the roof tells you what an insurance carrier will decide. The roofer's job in a storm claim is to document the conditions, measure them honestly, and write a fair estimate to repair to a pre-loss condition. The insurer decides coverage. The homeowner owns the claim and the decision. Keep that lane clean in every conversation. You are an expert on roofs, not an adjuster, and the fastest way to lose a market is to start promising outcomes you do not control.

A quick worked example

Picture a 1997 tract subdivision of 180 homes. The builder roofed them in batches over about ten months with the same three-tab product. Fast-forward to today and the original roofs are pushing 28 to 29 years, well past the 15-to-20-year window for that material. Some have been replaced once already by previous owners; many have not. A hailstorm with quarter-size stones clipped the east third of the subdivision two springs ago.

Now you have three distinct conversations on three blocks of the same neighborhood. On the east blocks, you have old roofs that took hail: the strongest storm-restoration bucket on the map, because age and storm multiply. On the untouched blocks, you have old roofs that are simply aging out: a slower, age-based sale, but a real one, because nobody buys a 28-year-old three-tab roof another decade. And scattered throughout you have homes a prior owner reroofed five years ago, which you skip unless the install was bad. One subdivision, three plays, and you know which is which before you knock because you read the vintage, the storm footprint, and the surface. That is the entire discipline in miniature.

Estimating roof age without a paper trail

The single most useful number when ranking roofs is age, and the frustrating reality is that the homeowner usually does not know it. They bought the house six years ago. The roof was "already on." The permit history is a coin flip. So you learn to estimate a range, because a range is honest and a single date is a guess dressed up as a fact.

Here is how seasoned reps triangulate age.

Read the shingle granules

Asphalt shingles are protected by a layer of mineral granules. Those granules are sacrificial; UV and weather knock them loose over the roof's life, and the rate of loss accelerates near the end. From the ground with binoculars, or up close on the deck, you are looking for:

• Granule thinning that exposes the asphalt mat. Shiny, darker, smoother patches where the colored granules are gone. Early life: none. Mid-life: scattered on south- and west-facing slopes. End of life: widespread, with the black mat clearly visible.
• Granules in the gutters and at downspout splash blocks. A handful after a fresh storm is normal on any roof. A consistent sandy load, especially years after install, says the roof is shedding its protection faster than it should.
• Color fade, unevenly. The slopes facing the equator (south in the U.S.) and the prevailing weather take the most UV and fade first. A roof that is two distinct colors front-to-back is telling you its age and its exposure at the same time.

Read the shingle shape

Asphalt cups, curls, and clamps as it dries out and loses oils over the years.

• Curling at the edges ("fishmouthing") and cupping of the tab corners are classic late-life signals. The shingle is no longer lying flat, which means wind can get under it and water can find the seams.
• Cracking that runs with or across the shingle, especially in cold climates where thermal cycling is brutal, points to an embrittled mat.
• Bald, brittle shingles that crumble when you flex a corner are end-of-life, full stop.

Read the neighborhood

This is the trick that separates pros from amateurs, and it is the foundation of smart canvassing. Tract-built subdivisions were roofed in batches. The builder put the same product on the same vintage of homes within months of each other. So if you find one house on the street with a roof that hailed out or aged out, the houses around it are statistically the same age and the same product, installed by the same crews, with the same flaws.

That is why a single confirmed replacement on a block is a signal to work the whole block, not a one-off. It is also why pulling the construction era of a neighborhood, from county records or the assessor's data, gives you a baseline age for the original roofs. A 1998 subdivision on original architectural shingles is squarely in the replacement window in 2026 even before anyone climbs a ladder.

Read the layers

Look at the rake edge and the eave. If you can see two distinct layers of shingle, the roof has been reroofed over once already. That matters for two reasons: a second layer traps heat and ages faster, and most codes and manufacturers will not allow a third layer, so the next event is a full tear-off. A double-layer roof is closer to "due" than its visible surface alone suggests.

A practical age-estimate worksheet

None of these alone is decisive. Three or four pointing the same direction give you a defensible range. And a range, not a date, is what you write down: "roof reads 16 to 20 years, original to the subdivision, sun slopes near end of life."

Why ventilation tells you the real age

Two identical roofs installed the same week on the same street can be years apart in effective age, and the difference is almost always ventilation. A roof needs balanced intake (at the soffits) and exhaust (at the ridge or upper field) so that air moves continuously through the attic and carries heat and moisture out. When that loop is broken, the attic turns into an oven in summer and a condensation chamber in winter, and both conditions cook the shingles from below and rot the deck from the inside.

You can read ventilation problems without a meter. From the street, look for the absence of ridge vents or roof vents on a large roof, painted-shut or stuffed soffit vents, and shingles that look prematurely curled or blistered on the slopes over the living space rather than the garage. In the attic, the tells are obvious once you know them: a wall of heat when you open the hatch on a mild day, frost or water stains on the underside of the deck in cold climates, rusted nail tips poking through the sheathing, and matted or moldy insulation. A roof on a badly ventilated attic is effectively older than its calendar age, sometimes by five or more years, and it will fail early no matter how good the shingles were. When your curbside age read and your attic read disagree, the attic wins.

Documenting the age read so it holds up

Whatever range you land on, write it down the same way every time so your notes are usable later and defensible if anyone questions them. A good age note names the evidence, not only the conclusion: "original to 1999 subdivision per assessor; south slope granule-thin with exposed mat; tabs curling on west exposure; single layer; pipe boots cracked. Reads 22 to 26 years, end of service life on sun slopes." That sentence tells the next person on your team exactly what you saw and why you concluded what you did, and it never overstates a date you cannot know. Sloppy notes ("old roof, looks done") are worthless a week later and indefensible in front of a homeowner or an adjuster.

Reading a roof from the street (the 30-second curbside triage)

When you are canvassing, you cannot inspect every roof. You triage. A good rep can sort a street into knock / skip / maybe in seconds per house, from the sidewalk, before ever ringing a bell. Here is the curbside checklist, in priority order.

1. Surface texture and color uniformity. A roof that is one even color and lies flat is probably mid-life or younger. Skip or maybe. A roof that is blotchy, faded unevenly, and visibly rough is your knock.
2. The lines. Sight down the ridges and hips. Wavy, sagging, or dipping lines mean deck problems underneath, which means the conversation is bigger than shingles. Strong knock.
3. Missing, lifted, or patched shingles. Even one obvious bare spot or a tarp is an invitation. Wind damage shows up as creased or torn tabs and missing shingles, usually concentrated on the windward slope.
4. The accessories. Look at the pipe-boot rubber, the ridge cap, and any visible flashing. Cracked boots and curled ridge caps are early, cheap-to-see signals that the whole assembly is drying out.
5. The neighbors' work. Dumpsters, new shingle bundles on driveways, fresh-and-bright roofs next to weathered ones. A street being reroofed is a street where the rest of the homes are the same age.
6. Trees and exposure. Heavy tree cover means shade (slower UV aging) but also constant debris, moss in wet climates, and limb-strike risk. Full southern exposure means faster aging. Note it; it changes your age read.

The goal of curbside triage is not to be right about every house. It is to spend your knocks on the top third of the street and not waste breath on the bottom third.

What hail and wind actually do (and how to read storm damage)

Storm work is where age and weather collide, and it is the highest-value bucket for most restoration crews. But it is also where reps get sloppy, oversell, and burn their reputation. Read the damage honestly.

Hail

Functional hail damage to asphalt shingles looks like bruises: round, soft spots where the impact fractured the mat and knocked granules loose, often with a darker center where the asphalt is exposed. Press one and it may feel soft, like the give in a bruised apple. Key tells:

• Randomness and density. Real hail hits are randomly distributed across slopes, with higher counts on the sides that faced the storm. A test square (commonly a 10-foot-by-10-foot area) is used to count hits and judge density.
• Collateral, soft-metal evidence. Hail never hits only the roof. Check the gutters, downspouts, gutter aprons, metal vents, the AC condenser fins, mailboxes, and window screens. Dents in soft metals around the property corroborate that hail of damaging size fell here. This is the single most persuasive, hard-to-fake evidence on a property.
• Granule loss exposing fresh, dark mat. Fresh hits expose mat that has not yet weathered, so it reads darker than the surrounding aged surface.

What is NOT hail: blistering (a manufacturing/heat artifact, usually pops up in clusters and exposes lighter, un-fractured asphalt), mechanical scuffs from foot traffic, and normal granule loss from age. Calling blisters hail is how you get a reputation with adjusters that follows you around a market.

Hail size matters. Generally, marble-sized hail and smaller rarely does functional damage to asphalt; quarter-size and up starts to matter; and the bigger the stones and the higher the density, the more likely an event produced a payable loss. You can corroborate that a damaging storm hit an address using public storm records, which brings us to a major point below.

Wind

Wind damage is more obvious from the street than hail. Look for:

• Creased shingles. Wind lifts a tab and folds it back; when it lays back down it leaves a crease across the shingle. Even if it is not missing, a creased shingle has a broken seal and a compromised mat.
• Missing shingles and exposed nail heads or felt. Concentrated on the windward slopes, ridges, hips, and rakes where uplift is highest.
• Lifted seals. A shingle that lifts easily by hand because the factory seal-down strip let go. Once seals break, the next wind event takes the shingle.

Wind damage tends to be directional and concentrated, where hail is more random and distributed. Reading which one you are looking at tells you what kind of claim and conversation you are walking into.

Age plus storm is the real signal

Here is the part most reps miss. A storm does not damage every roof equally. An old, brittle, granule-thin roof takes far more functional damage from the same hailstones than a fresh, pliable roof next door. So the roofs most likely to be "due" after a storm are the ones that were already aging. Age and storm exposure multiply. When you can layer where the storm actually hit on top of how old each roof already was, you are no longer guessing. You are ranking.

A note on climate and exposure

The same roof ages at wildly different rates depending on where it sits. In the hail-heavy plains, storm exposure dominates the math and a roof's effective life is whatever the next severe season decides. In the deep South and Southwest, relentless UV and heat are the killers; sun slopes there can be a decade ahead of their north-facing counterparts on the same house. In cold, wet climates, freeze-thaw cycling and ice damming punish the eaves and valleys first, and moss and algae thrive on shaded north slopes, holding moisture against the shingles and shortening their life. In coastal zones, salt air and high design-wind exposure mean seals fail and edges lift sooner.

The practical upshot is that you adjust your age ranges to your market. A 15-year-old asphalt roof in a mild, dry, low-storm climate may have real life left, while a 15-year-old roof in a hail alley or a brutal-sun market is squarely a replacement candidate. Reps who learn their local climate's failure pattern read roofs faster and more accurately than any generic chart, because they know which slope and which detail to look at first.

Using storm data instead of chasing rumors

The old way to find storm work is to hear a town got hit, drive there, and knock randomly until you find damage. It works, barely, and it wastes enormous amounts of windshield time on streets the storm missed by a mile. Hail in particular is famously patchy; one side of a subdivision can be totaled while the other side took nothing.

The better way is to start from data about where damaging weather actually fell.

• Public storm reports. The National Weather Service and the Storm Prediction Center publish severe weather reports, including hail size reports and wind reports, by date and location. The Storm Events Database is a searchable archive going back decades. These tell you which events were real and roughly where they landed.
• Radar-derived hail estimates. Beyond ground reports, radar products estimate maximum hail size over a grid, which fills in the gaps between sparse human reports. This is how you find the streets that hailed but where nobody happened to file a report.
• Your own confirmed inspections. Every roof your crew confirms as damaged is a data point. Plotted on a map, your confirmations draw the real footprint better than any forecast.

The point is to knock where the weather was, on the roofs that were already old enough to be vulnerable, and to stop treating an entire county as one undifferentiated target.

Where RoofPredict fits in the ranking

Most of what is above is work you can do by hand. The reason it rarely gets done well at scale is that it is enormously time-consuming to estimate roof age house-by-house and overlay storm exposure across thousands of addresses before your crews hit the street. That is the specific gap RoofPredict is built to close.

The platform reads aerial and street-level imagery to produce a roof-age range per address, not a fabricated install date, a range, the same way a sharp rep would read granules and shingle lay from the curb, but across an entire territory at once. Then it models storm physics per roof: it layers where damaging hail and wind actually fell on top of each roof's estimated age and condition, because, as covered above, age and storm exposure multiply. The output is a ranked list of doors and routes, so your canvassers and reps spend their finite hours on the roofs the storm most likely wore out and the roofs that are simply aging out of their service life.

A few honest limits, because the tool is only useful if you trust it:

• It is a ranking, not an X-ray. RoofPredict tells you which roofs are most likely due and should be looked at first. It does not replace the ladder, the test square, or the human eye on the deck. Nobody should write an estimate off a screen.
• Age is a range, on purpose. Imagery cannot read a permit. A range that is honest beats a date that is invented. Treat "16 to 20 years, original to the subdivision" as a prioritization signal, not a fact to repeat to a homeowner as gospel.
• Storm modeling is odds, not proof. Showing that damaging hail probably fell on a street tells you where to look. It is not evidence of damage on any one roof, and it is never a substitute for documenting the actual conditions on that roof. The forecast points the truck; the inspection makes the case.
• You still own the conversation. The data ranks doors. It does not knock them, build rapport, or write the estimate. Crews still close jobs the old-fashioned way.

Used the way it is meant to be used, it turns the curbside triage and storm-overlay work into something you do before you ever leave the office, so the day's route is already sorted knock-first by the time the trucks roll. That is the whole value: more of your daylight aimed at roofs that are actually due.

The economics of a wasted knock

It helps to put a number on why ranking matters, even a rough one you build for your own shop. Say a canvasser costs you a loaded rate per hour and knocks ten doors an hour, with maybe three answering and one agreeing to an inspection on a good street. On a poorly targeted street, that conversion can fall by half or more because most of the roofs are too young to need anything. Run the math across a 40-hour week and the gap between a well-ranked route and a random one is not a rounding error; it is the difference between a canvasser who pays for themselves several times over and one who barely breaks even.

The same logic applies to your sales reps and the windshield time between inspections. A rep who runs eight tight, pre-qualified inspections in a day closes more total jobs than a rep who runs five spread across a county chasing rumors, even if the second rep is the better closer. Targeting beats talent more often than salespeople like to admit. This is why the disciplined shops obsess over which doors, more than over the pitch at the door, and why getting the ranking right upstream quietly outperforms every sales tactic downstream.

Build versus buy your ranking

You can assemble a coarse version of address-level ranking by hand: pull subdivision vintages from assessor data, overlay storm reports from public sources, drive the streets for curbside signals, and score each block. For a single neighborhood or a small market, that is entirely doable and worth the afternoon. The wall you hit is scale. Doing it by hand across thousands of addresses, refreshing it after every storm, and keeping the age estimates current is a full-time job nobody on a roofing crew has time for, which is exactly why most shops never do it and fall back to knocking randomly. The choice is not whether ranking helps; it always does. The choice is whether you build it by hand for a few streets or let a system produce it across the whole territory so your crews start every day already sorted.

The on-roof inspection: confirming what the street suggested

Curbside triage and a ranked list get you to the right doors. They do not write the estimate. Once you are invited up, the job is to confirm or rule out, honestly and systematically, so your documentation holds up and your homeowner trusts you. Safety first: follow fall-protection basics, never walk a roof that is too steep or too wet or too brittle to bear weight, and know that occupational fall rules exist because roofing is one of the most dangerous trades there is.

Work the roof in a consistent order so you never miss a plane:

1. Walk the perimeter and sight the planes. Note slope direction, exposure, and which slopes faced the last storm. Photograph the whole roof in context before you touch anything.
2. Chalk and count a test square. On a representative slope, mark a 10x10 area, count hail hits, and photograph with the chalk visible. Repeat on the windward slope and a leeward slope to show distribution.
3. Inspect the field shingles. Check for bruising, granule loss, mat fractures, creasing, lifted seals, and curling. Lift a few tabs gently to check the seal-down strip.
4. Inspect every penetration and transition. Pipe boots, vents, skylights, chimney and wall flashing, valleys, ridge and hip caps. These leak first and they are where age shows fastest.
5. Check the soft metals and accessories. Gutters, downspouts, drip edge, vent caps, and the AC unit. Corroborating hail dents here are your most persuasive evidence.
6. Get in the attic if you can. Decking moisture, daylight through the boards, rusted nail tips, inadequate ventilation, and active leak staining tell you whether the problem is surface-only or structural. Poor ventilation is the quiet killer that ages decks and shingles from below; if the attic is an oven, the roof above it is aging faster than its years.
7. Measure. Get accurate squares, pitch, and accessory counts so the estimate is real. Many crews use aerial measurement reports to speed this up, then verify on site.

Document everything with date-stamped, address-identifiable photos. The estimate you write should restore the roof to its pre-loss condition, no more and no less. That discipline is what keeps you on the right side of adjusters and homeowners both.

What pros get wrong on the roof

• Calling blisters hail. The fastest way to lose credibility with a desk adjuster. Learn the difference cold.
• Inspecting one slope. Damage distribution is the argument. One slope is not a story.
• No collateral evidence. Roof hits with zero dents in soft metals around the property are a weak claim. Always photograph the gutters, screens, and AC.
• Overstating to win the door. If you tell every homeowner their roof is totaled, you are not an inspector, you are a salesperson with a ladder, and it catches up with you.
• Ignoring the attic. The deck tells you whether this is a reroof or a rebuild. Skipping it costs you on the estimate and the warranty.

Turning the read into an honest homeowner conversation

Knowing a roof is due is only useful if you can explain it to the person who owns it without overselling. The homeowner has heard the horror stories about storm-chasers, and the fastest way to lose them is to walk up acting certain about things you cannot be certain about. The reps who convert at the door do the opposite: they show their work and they stay in their lane.

Lead with what you actually observed, framed as evidence rather than verdict. "Your roof reads as original to the neighborhood, which puts it in the mid-twenties for age, and the slopes facing the afternoon sun are losing their granules, see the darker patches there." That is true, it is specific, and it invites the homeowner to look with you instead of being sold at. Then offer the inspection as the next honest step, not as a foregone conclusion. You are documenting conditions so the homeowner has the facts; you are not pronouncing a sentence on their wallet.

When storm work is on the table, the discipline tightens further. You can say that records show damaging hail or wind affected the area and that you would like to document whether this specific roof took functional damage. You cannot tell them their roof is covered, promise an outcome, or treat a storm record as proof of damage on their particular roof. Those are not your calls. The roofer documents conditions and writes a fair estimate to restore the roof to its pre-loss condition; the insurer decides coverage; the homeowner owns the claim and the decision. Reps who blur that line win a few extra doors in the short run and lose the market's trust, and the adjusters', over a season.

And sometimes the honest read is that the roof is fine. A homeowner with a seven-year-old roof in good shape should hear "you've got several good years left, here's what to watch for." That five-minute act of leaving money on the table is the single best marketing a roofing company can do, because that homeowner remembers who told them the truth, and they call that company, and tell their neighbors, when the roof actually is due.

Material-specific lifespans and tells

Asphalt dominates U.S. residential roofing, but you will run into other materials, and "due" means something different for each. Use these as ranges, not promises; ventilation, install quality, and climate move every number.

Two notes that catch reps off guard. On tile and slate, the visible material can outlive the building, but the underlayment and flashing fail decades sooner, so a 40-year-old tile roof can absolutely be "due" for a tear-and-reset even though the tiles look fine. And on metal, the failure is usually at the fasteners and coatings, not the panels, so you read the screws and seams, not the field.

Turning all of this into a route your crew can run

Knowing which roofs are due is worthless if it does not change where the trucks go. Here is how to convert reads and data into a workday.

Step 1: Define the target footprint

For aging-out work, that is the subdivisions built in the right era for the material on them, roughly 15-plus years ago for asphalt, adjusted by climate. For storm work, that is the confirmed hail and wind footprint from public storm data and your own confirmations, not the whole county the news mentioned.

Step 2: Rank the addresses

Score each address on the factors you can know before knocking: estimated roof-age range, whether damaging weather actually fell there, exposure, subdivision vintage, and any visible curbside signals. This is exactly the per-address ranking RoofPredict produces at territory scale, but a disciplined sales manager can do a coarse version by hand for a few streets with assessor data and storm reports.

Step 3: Sequence the route knock-first

Order the day so canvassers hit the highest-probability doors during the best knocking hours (late afternoon and early evening on weekdays, mid-morning to afternoon on weekends), and fill the dead hours with lower-probability streets or follow-ups. Tight geographic clustering matters as much as scoring; a high-probability door 20 minutes away loses to two good doors next to each other.

Step 4: Track the confirmations

Every inspection result, due / not due / damaged / declined, goes back into your map. Over a season, your own confirmations become the most accurate footprint you own, and they sharpen every future route. The companies that compound advantage are the ones that treat each knock as a data point rather than a coin flip.

A simple scoring rubric

Give each address a quick 0-3 on each factor and add them up. Work the highest totals first.

A street where most homes score 8 or higher is a street worth a full day. A street averaging 3 is a street to skip until it ages or a storm finds it.

Edge cases and judgment calls

The rubric handles the common roofs. The money is often in the exceptions, where a rep with judgment beats a rep with a checklist.

• The young roof that is failing. A five-year-old roof with curling and granule loss is not aging out, it is failing, usually from bad ventilation, a bad install, or a defect. The homeowner is confused and frustrated and ripe for an honest expert. Read the attic, document the cause, and you have a job a checklist would have skipped.
• The shaded roof that reads young but is not. Heavy tree cover slows UV aging, so a 22-year-old north-facing slope under oaks can look 12. Trust the subdivision vintage and the accessories (boots, flashing, seals) over the field color here.
• The reroofed-over double layer. Surface looks mid-life, but it is a second layer trapping heat. It is closer to due than it looks, and the next event is a full tear-off because a third layer is off the table.
• The storm that missed by a block. Hail patchiness is real. Confirm the actual footprint before you commit a crew; do not let one totaled roof convince you the whole zip code hailed.
• The homeowner who "just had it done." Sometimes the previous owner did, sometimes a fly-by-night did it poorly five years ago. "Recently replaced" is not the same as "done right." A quick look tells you which.
• The cosmetic-only roof. A roof with scattered granule loss and one cracked boot is a repair, not a replacement, and selling it as a replacement is how you lose the market's trust. Sometimes the honest answer is "you have several good years left," and saying it earns you the call in five years.

A field checklist you can hand to a crew

Print this. It is the whole playbook compressed to what a rep needs in their hand.

From the street:

• Color uniform and flat? (younger) or blotchy and rough? (older)
• Ridge/hip lines straight, or wavy/sagging? (deck issue)
• Missing, creased, lifted, or patched shingles?
• Cracked pipe boots or curled ridge cap?
• Neighbors being reroofed? Dumpsters, fresh bundles?
• Full sun or heavy shade? (adjusts age read)

Estimating age:

• Granule coverage: full / thinning / bald patches?
• Shingles flat / lifting / curling?
• One layer or two at the rake?
• Subdivision vintage from records?
• Write a RANGE, never a date.

Storm check:

• Did damaging hail/wind actually hit this address per storm records?
• Hail: random bruises, soft spots, fresh dark mat?
• Wind: creased/missing shingles on windward slopes?
• Collateral dents in gutters, screens, AC, soft metals?

On the roof:

• Chalk and count test squares on multiple slopes.
• Inspect every penetration and flashing.
• Get in the attic: moisture, daylight, ventilation, leaks.
• Photograph everything, date- and address-stamped.
• Measure for a real estimate.

Before you write it:

• Is this a repair or a replacement? Be honest.
• Does the documentation tell the whole story?
• Am I staying in my lane: document conditions, write the estimate, let the insurer decide coverage?

Putting it together

Knowing which roofs are due is not one skill, it is a stack of them: reading granules and shingle lay to estimate an age range, reading the neighborhood to know that roofs come in batches, reading hail and wind honestly to tell damage from artifacts, and reading storm data so you knock where the weather actually was instead of where the news said it might have been. The reps who do all of this by instinct are the ones whose close rates make the rest of the company wonder what their secret is. There is no secret. They just stopped wasting knocks.

The leverage comes from doing that ranking before the trucks roll, across a whole territory, instead of one street at a time from the sidewalk. That is the work RoofPredict automates: a roof-age range per address from imagery, storm physics modeled per roof, and a route sorted knock-first by the odds each roof is due, so your finite daylight lands on the doors most likely to convert. It will not climb the ladder or close the job. It will make sure your crew spends the day on roofs that are actually worth the climb. Point your hours at the right roofs and everything downstream, close rate, cost per job, crew morale, gets easier.

FAQ

How can I tell how old a roof is if the homeowner doesn't know?

You estimate a range, not a date. Read granule coverage (full, thinning, or bald patches exposing the mat), shingle lay (flat, lifting, or curling), and cracking. Cross-reference the subdivision's construction era from county or assessor records, since tract homes were roofed in batches. Check the rake edge for a second layer. Three or four signals pointing the same way give you a defensible range like '16 to 20 years,' which is honest and useful, where a single invented date is neither.

What are the clearest signs a roof needs replacing rather than repair?

Widespread granule loss with exposed asphalt mat, shingles that curl, cup, or crack across most slopes, sagging or wavy ridge and hip lines (a deck problem), multiple failed pipe boots and flashings, and an attic showing chronic moisture or ventilation failure. Scattered, isolated issues, one cracked boot, a few missing shingles in one spot, are usually repairs. Replacement is the call when the failures are widespread or structural, not local.

How do I know if a roof has real hail damage versus normal wear?

Functional hail damage shows as random, soft bruises where the impact fractured the mat and knocked granules loose, exposing fresh dark asphalt, with higher density on storm-facing slopes. The decisive corroboration is collateral: dents in soft metals like gutters, downspouts, vent caps, window screens, and AC condenser fins. Blistering (clustered, lighter, un-fractured), foot-traffic scuffs, and age-related granule loss are not hail. Count hits in a chalked test square on multiple slopes.

How long do asphalt shingles actually last?

In real-world conditions, three-tab asphalt typically runs about 15 to 20 years and architectural (dimensional) shingles about 22 to 30 years. Printed warranty numbers are usually optimistic versus field performance. Ventilation, install quality, sun exposure, and climate move these ranges significantly: a poorly ventilated, full-sun, double-layer roof ages far faster than a single-layer, well-ventilated, shaded one of the same age.

Why do some roofs in a neighborhood need replacing while others don't?

Two reasons. First, hail is patchy; a storm can total one side of a subdivision and miss the other by a block, so confirmed storm footprint matters more than the zip code the news named. Second, age and storm exposure multiply: an older, brittle, granule-thin roof takes far more functional damage from the same hailstones than a fresh, pliable one next door. The roofs most likely to be due after a storm are the ones that were already aging out.

How can I find which streets actually got hit by a storm?

Start from data, not rumors. The National Weather Service and Storm Prediction Center publish hail-size and wind reports by date and location, and the Storm Events Database archives them. Radar-derived maximum hail-size estimates fill in gaps between sparse ground reports. Then layer your own confirmed inspections onto a map; over a season they draw the real footprint better than any forecast. Knock where damaging weather actually fell, on roofs already old enough to be vulnerable.

Does RoofPredict tell me exactly which roofs need replacement?

No, and it shouldn't claim to. It produces a roof-age range per address from aerial and street imagery, then models storm physics per roof to rank which doors are most likely due, so your crew works them first. It is a prioritization signal, not an X-ray: age is an honest range, not an install date, and storm modeling is odds, not proof of damage on any one roof. The ladder, the test square, the attic, and the human eye still make the actual call and write the estimate.

Can I tell a homeowner their roof is covered or that a storm damaged it?

Stay in your lane. The roofer's job is to document the actual conditions, measure them, and write a fair estimate to restore the roof to its pre-loss condition. The insurer decides coverage and the homeowner owns the claim. Showing that damaging hail probably fell on a street tells you where to inspect; it is not proof of damage on a specific roof and never a substitute for documenting real conditions. Overstating to win a door catches up with you fast.

How do material types change when a roof is 'due'?

Service life and failure mode differ. Asphalt fails in the field (granules, curling, cracking) at roughly 15 to 30 years depending on type. Metal usually fails at fasteners, seams, and coatings, not the panels, often at 40-plus years. Tile and slate can outlast the building, but their underlayment and flashing fail decades sooner, so a 40-year tile roof can be due for a tear-and-reset even with sound tiles. Read the right failure point for each material.

What's the biggest mistake crews make when qualifying roofs?

Wasting knocks. Inspecting every house instead of triaging the street, calling blisters hail, inspecting only one slope, skipping the collateral soft-metal evidence and the attic, and overstating damage to win a door. The fix is discipline: triage from the curb, estimate age as a range, confirm storm footprint from data, document multiple slopes and collateral honestly, and be willing to tell a homeowner they have good years left. Honesty earns the callback in five years.