How to Choose a Roofing Service Area That Pays the Bills

Most roofing companies pick a service area by accident. You started in the town you live in, you took the jobs that called, and over a few years a blob formed on the map. It works until it doesn't. Then a crew spends ninety minutes a day in the truck instead of on a roof, your best closer drives past four good neighborhoods to reach one bad one, and your marketing dollars scatter across zip codes that will never buy a re-roof because every house out there is eight years old.

Choosing a service area is one of the few decisions in this business that quietly sets the ceiling on everything else. It decides your close rate before a single door gets knocked, because it decides who answers the door. It decides your gross margin, because it decides how far the truck rolls and how many production hours get eaten by windshield time. It decides whether your marketing spend compounds or evaporates. And unlike hiring or pricing, you usually only revisit it after something breaks.

So treat it like the capital decision it is. Below is the way an operator draws the boundary: which signals actually predict revenue, how to weigh drive time against density, how to read storm exposure without fooling yourself, how to size up the competition honestly, and how to turn a fuzzy region into a ranked, knockable list of streets. There is real math, worked examples with real numbers, and the mistakes that cost the most. By the end you should be able to take a county map and a few public datasets and say, with a straight face, exactly where you will and won't sell, and why.

What a service area actually is (and the three you really run)

People say "service area" like it's one thing. It's three, and conflating them is the first mistake.

Your sales area is where you actively prospect, market, and knock. It's the smallest and most deliberate of the three. This is the set of streets where you spend money to create demand.

Your install area is where you'll send a crew if a good job lands. It's usually wider than your sales area because a referral or a commercial bid is worth the drive even when prospecting there isn't.

Your storm area is the elastic one. After a verified hail or wind event you'll chase exposure well outside your normal footprint, run it hard for a season, and then retreat. Storm work has different economics, a different sales motion, and a different shelf life, so it gets its own boundary that expands and contracts.

The trap is letting the storm area silently become the sales area. A company chases a hailstorm two counties over, does well for a summer, keeps a few customers there, and now "the service area" includes a place forty-five minutes away where they have one past job and no density. Every new lead from that zip costs a half-day of production to service. Draw the three boundaries separately and label them, even if it's just three colors on the same map.

The rest of this is mostly about the sales area, because that's the one you control with a pen. We'll come back to the storm area, because reading exposure correctly is where a lot of money is won and lost.

The five signals that actually predict roofing revenue

Forget vanity geography. A zip code being "nice" or "close" tells you almost nothing. Five signals do most of the predictive work, and they're all knowable before you spend a dollar.

1. Roof age distribution

A re-roof happens when a roof reaches the end of its serviceable life or gets damaged. Asphalt shingle, which covers the overwhelming majority of US steep-slope residential roofs, typically serves somewhere in the 15 to 30 year range depending on product, ventilation, slope, and climate. That's a range, not a number, and you should hold it loosely. But the implication is blunt: a subdivision built three years ago is dead inventory for a retail re-roof for at least a decade. A neighborhood where the original roofs went on 18 to 24 years ago is a field that's ripening.

This is the single most under-used signal in the trade, mostly because roof age is hard to see from a spreadsheet. You can infer it from the year a house was built for the original roof, but houses get re-roofed, so the build year decays as a proxy the older the home gets. A 1998 house might be on its second roof from a 2011 hailstorm, in which case it's not due, it's freshly covered. More on resolving that below.

2. Housing stock and ownership

Owner-occupied single-family homes convert on retail re-roofs at a far higher rate than rentals or large multifamily, because the decision-maker lives under the roof and carries the insurance. You want the share of owner-occupied detached single-family housing, and you want the absolute count of those homes per square mile. The US Census Bureau publishes both through the American Community Survey at the tract and block-group level for free. A tract that's 85% owner-occupied detached homes is a different animal than one that's 60% renters in townhomes, even if the median income is identical.

3. Density of qualifying homes

Density is the quiet multiplier on every other number. Two areas can have the same count of due roofs, but if one packs them into four square miles and the other spreads them across forty, your cost to knock, your fuel, and your crew utilization differ by a factor of three or more. A canvasser working a dense neighborhood might hit 60 to 80 doors in a shift; the same canvasser in a rural spread hits 25 and burns gas between them. Always think in qualifying roofs per square mile, not raw population.

4. Storm exposure history and forward odds

Hail and high wind are the two events that turn a slow retail market into a busy one and put insurance on the table. You want two things: the historical frequency of significant events for the area, and an honest read on whether a specific recent storm actually hit the roofs you care about. The NOAA Storm Prediction Center and the National Weather Service publish storm reports and archives; the Insurance Institute for Business and Home Safety (IBHS) publishes regional hail and wind research. Use history to rank a market's baseline; use per-storm verification to decide whether to mobilize.

5. Insurance and rebuild economics

This one is subtle and it's where you must stay disciplined. You are not adjusting claims or interpreting anyone's coverage. But a market's baseline economics still matter: average home values, prevailing roof sizes and pitches, and local labor and material costs all shape the size of a typical job. A 30-square steep cut-up roof on a custom home is a different ticket than a 14-square walkable ranch. You can read typical roof size straight off aerial imagery and parcel data. Bigger, more complex roofs mean larger average jobs, which changes how far the drive is worth it.

Drive time is the hidden tax: do the math

Here is the number almost nobody calculates and everybody pays.

A crew that loses 90 minutes a day to travel, versus 30, loses an hour of production per crew per day. Over a 250-day working year, that's 250 hours. If a productive crew-hour is worth, say, 350 dollars in installed revenue, that single hour a day is 87,500 dollars of capacity per crew per year that you torched in the truck. With three crews, you're past a quarter million in lost throughput, and you paid fuel and wages for the privilege.

That's the case for thinking in drive-time radius, not mileage radius. Twenty miles on an interstate is twenty-five minutes; twenty miles through town with eleven lights and a school zone is fifty. Draw your boundary in minutes from your shop or your material yard, not in a clean circle.

A simple worked drive-time model

Let's put numbers on a sales-area decision. Say you're weighing whether to add a neighborhood that's a 35-minute drive each way versus doubling down on one that's 12 minutes away.

The far neighborhood isn't automatically wrong. If it just took verified hail and the near one didn't, the far one wins this season easily. But for steady retail, the near neighborhood lets a canvasser knock 75% more doors per shift and lets a crew waste 46 fewer minutes per service trip. You'd need the far area to convert meaningfully better, or carry much bigger average tickets, to justify making it core sales territory rather than opportunistic install territory.

The 30-45-60 rule of thumb

A workable default for a single-shop residential operation:

• 0 to 30 minutes: core sales area. Market here, knock here, this is where your density should be highest and your CAC lowest.
• 30 to 45 minutes: secondary. Service good inbound leads and referrals, run targeted campaigns only where the data is strong, don't blanket-market.
• 45 to 60 minutes: opportunistic and storm only. Take the big job, chase the verified event, but don't build steady demand-gen here unless you're standing up a second location.
• Beyond 60 minutes: you're not extending a service area, you're opening a branch. Treat it like one, with its own yard, crews, and break-even plan.

Adjust the bands for your geography. A dense metro might compress everything inward; a rural market might need to stretch them or you'll never have enough roofs. The point is to pick the bands deliberately and let them gate marketing spend.

Reading roof age without guessing

Roof age drives retail demand, and getting it roughly right is worth more than any other single input. The problem is you can't see it from the street at scale and the public proxies are imperfect. Here's how the pieces actually behave.

Year built gets you the original roof date and nothing after. It's accurate for newer homes and decays steadily as homes age and accumulate re-roof events. For a 2015 subdivision, year built is a great proxy. For a 1985 neighborhood, it tells you almost nothing about the current roof.

Permit records are the gold standard when they exist and are digitized, because a pulled re-roof permit timestamps the current roof. Coverage and quality vary wildly by jurisdiction; many re-roofs were never permitted at all. Where your county or city has a searchable permit portal, mine it. Where it doesn't, you're back to inference.

Aerial and satellite imagery is where the real progress is. Current high-resolution imagery shows granule loss, color fade, patching, tarps, missing shingles, and prior repairs. Time-stacked historical imagery can bracket when a roof changed, because you can often see the before-and-after across image dates. From imagery you also get roof size, pitch class, complexity, and material, which feed your average-ticket math. None of it gives you a birth certificate. What it gives you is a defensible range, and a range is exactly what you should be working with.

The honest framing, and the one that keeps you out of trouble in front of a homeowner, is this: "Based on the imagery and the records, this roof is most likely in the 18 to 24 year band." Not "your roof was installed in 2003." You don't know that, and pretending you do erodes trust the moment a homeowner remembers their 2014 replacement. A range is more honest and, used well, more persuasive, because you can pair it with what you can actually see: "the granule loss and the fade in the imagery are consistent with a roof near the end of its serviceable life."

A practical roof-age workflow for a target neighborhood

1. Pull the build-year distribution for the tract from Census or county assessor data. If most homes went up 19 to 25 years ago, flag the tract as ripening.
2. Check the permit portal for re-roof permits in that tract over the last 12 years. Heavy recent permitting means a chunk of the inventory is freshly covered, lower your expectations. Light permitting on old housing stock means more original roofs are likely still up there.
3. Cross-reference any known storm dates. If a verified hail event hit eight years ago, assume a meaningful share got replaced then and aren't due now.
4. Spot-check imagery on a sample of streets to confirm the inferred condition. Are you seeing faded, worn roofs, or recent replacements?
5. Assign a confidence band, not a date, to each street or block. Work the high-confidence-ripe blocks first.

This is slow to do by hand across a whole county, which is exactly why it usually doesn't get done and why the company that does it has an edge.

A worked roof-age inventory estimate

Put numbers on a single tract to see why ranges beat dates. Say a tract has 1,000 owner-occupied detached homes, mostly built in a 1999-2002 wave, so the original roofs are roughly 24 to 27 years old. Year-built alone would tag all 1,000 as overdue. But the permit portal shows 280 re-roof permits pulled in the last 12 years, and you know a verified hail event passed through about 9 years ago. Reading those together, you'd estimate that something like a quarter to a third of the original roofs were already replaced, many of them in that storm year, and those homes are now on roofs 9-ish years old, not due. That leaves on the order of 650 to 700 homes most likely still on aging original roofs, your real working inventory in that tract. Imagery spot-checks then sort those into 'clearly worn, work first' versus 'ambiguous, lower priority.' The output isn't 'this tract has 1,000 leads.' It's 'this tract has roughly 650-700 likely-due roofs, concentrated on these blocks, with a confidence band, not a date, on each.' That estimate is what you market against, and it's dramatically more accurate than treating build year as gospel.

Storm exposure: read it without fooling yourself

Storms are the biggest swing factor in roofing demand, and they're the easiest thing to be wrong about with confidence. Two failure modes dominate.

The first is chasing the headline, not the hail. A storm makes the news, the radio crackles with hail reports, and three counties light up with door knockers. But hail is brutally local. A core can drop golf-ball stones on one subdivision and leave the neighborhood a mile away untouched. "There was a storm in the county" is not actionable. "This specific grid of streets saw hail large enough and dense enough to damage shingles" is. If you mobilize on the headline, you'll knock a thousand doors where nothing actually got hit, train homeowners to distrust roofers, and burn your crews on inspections that find clean roofs.

The second is assuming wind and hail behave the same. Wind damage tracks differently, concentrates on exposures and roof edges, and is shaped by terrain and gust structure. A market with a high baseline of damaging wind but little hail is a real opportunity, but it's a different sales and documentation motion than a hail market.

How to verify a storm actually hit before you mobilize

• Pull the event reports from the NOAA Storm Prediction Center and your local National Weather Service office. Look at reported hail sizes and wind gusts mapped to actual coordinates rather than county-level callouts.
• Look at the swath, not the point. Hail and wind events have shapes. You want the footprint of damaging intensity, then intersect that footprint with where your qualifying roofs are.
• Confirm intensity thresholds. Small hail dents soft metals and stresses old shingles but often doesn't create a clean functional-damage story on a sound roof. Know roughly what size and density actually marks shingles in your market.
• Verify on a roof before you scale. Inspect a handful of properties across the suspected swath and confirm real, documentable damage before you commit a sales team and a marketing burst to the area.
• Time-box it. Storm areas have a shelf life. Decide up front how many weeks you'll run it and what close rate makes you pull out.

The forward-odds view

Beyond any single event, some markets simply sit in higher-frequency hail and wind corridors. Knowing a market's baseline event frequency helps you decide where to plant a flag for the long run rather than only where to chase this week. A territory that takes a damaging hail event every few years on average will generate recurring storm seasons; one that gets a real event once a decade won't carry a storm-dependent model. Use historical frequency to rank markets for permanence, and use per-event verification to decide each mobilization. One is strategy, the other is tactics, and people constantly mix them up.

A note you have to internalize: a forecast or a storm history gives you odds, not proof. "This area is high-exposure and a verified core passed over it" raises the probability that a given roof has real damage. It does not promise that any specific roof is damaged, that any specific homeowner has a claim, or what any insurer will decide. Keep your language probabilistic and your inspections honest, and you'll stay on the right side of both ethics and the law.

Staying legal when the market is insurance-driven

A lot of roofing demand is tied up with insurance, and a lot of roofers talk themselves into serious trouble around it. Where you choose to work doesn't change the rules, but storm-heavy areas put you closer to the line, so internalize it before you draw a boundary around hail country.

Here's the clean version of what a roofing contractor may do. You may inspect a roof. You may thoroughly document damage with photos and measurements. You may prepare an accurate, itemized estimate to repair the roof, ideally aligned to standard estimating practice like Xactimate line items, reflecting your real scope of work. You may state facts about your scope to the carrier. Then you hand that documentation and estimate to the homeowner. The homeowner files. The insurer decides coverage.

Here's the do-not-say list, and it's worth teaching every salesperson you hire word for word, because crossing it can constitute unlicensed public adjusting in many states and is regulated by state departments of insurance:

• Don't, for a fee, negotiate, adjust, or "handle" the claim on the homeowner's behalf.
• Don't interpret the homeowner's policy or tell them what is or isn't covered.
• Don't promise a specific payout, approval, or that the claim will go through.
• Don't promise the deductible will be waived, absorbed, eaten, or made to disappear. The deductible is the homeowner's legal obligation.
• Don't advertise a "free roof."
• Don't represent the homeowner against their insurer.

The safe frame is simple and it actually sells better because it's credible: document thoroughly, write an accurate repair estimate, hand it over, and let the homeowner and the insurer do their parts. Your value is the quality of your inspection and documentation and the soundness of your scope, not a promise about someone else's money. Pick your service area to put you in front of roofs that likely qualify on the merits, age and verified storm exposure, and then win on documentation rigor.

Sizing up the competition honestly

A neighborhood full of ripe roofs is worthless to you if six established crews already own it. Market selection is partly a contest, and you should scout it like one.

What to actually measure

• Density of established competitors. How many real, crewed roofing companies operate this area as core territory? Not how many show up in a search, but how many have yard signs, trucks you see, and a referral base here.
• Their position. Are they the cheap volume player, the premium installer, the storm chaser who blows through and leaves? There's usually an open lane. If everyone's fighting on price, the documentation-and-service play is open. If it's a sleepy retail market with no storm-savvy operator, you can own the storm response.
• Saturation signals. A neighborhood with a roofing yard sign on every third lawn just got worked. The inventory of un-pitched homeowners is thin for a while. Conversely, an old neighborhood with ripe roofs and almost no signs is under-served, and that's where you want to be.
• Switching cost and loyalty. Referral-driven markets are sticky and hard to crack cold. Storm markets churn faster because the buying event is external and urgent.

The under-served pocket play

The best territory is frequently not the obvious one. It's the aging neighborhood that's a little awkward to reach from where the big competitors are based, so they under-work it. Find the pockets of ripe, owner-occupied, decent-density housing that sit in the seam between two competitors' cores, or just past the edge of where the dominant player bothers to drive. Those pockets convert better and cost less to win because nobody's saturating them. The data work above, roof age plus ownership plus density, is exactly how you find them before a competitor does.

Putting it together: a territory scoring model you can run

You don't pick a service area on vibes. You score candidate areas, sort the list, and draw the boundary around the top. Here's a model you can build in a spreadsheet this week.

Work at the block-group or zip-code-tabulation level to start, then drill into streets for the winners. For each candidate area, score these factors 1 to 5 and weight them.

A worked example: three candidate zips

Say you're a single-shop retail-plus-storm operator choosing between three areas at the edge of your map.

Area A is a 2004-2007 subdivision belt, 14 minutes out, 82% owner-occupied detached, dense, moderate storm history, and two competitors already running it hard.

Area B is a mixed older neighborhood, 22 minutes out, 78% owner-occupied, decent density, sits in a higher hail-frequency corridor, and is oddly under-served because it's across a river from where the big competitor is based.

Area C is a newer 2018 development, 18 minutes out, high owner-occupancy, dense, but the roofs are seven years old and nothing's due.

Score them:

Area C falls out immediately, dead inventory, no matter how nice it is. A and B are close, and that's the interesting result. Area A scores highest on paper but two crews already own it, so your real-world CAC there will run high and you'll be the third sign on the lawn. Area B is nearly tied, sits in a better storm corridor, and is under-served because of a geographic quirk. For a company trying to build durable, lower-cost demand, B is very likely the better plant-a-flag choice even though A edges it on the raw score. The model gets you to the short list; judgment about competition and durability picks the winner. That's how it's supposed to work. The spreadsheet ranks, you decide.

From region to a knockable list: street-level prioritization

A scored zip code is still too coarse to deploy a canvasser against. The last step is turning the winning areas into a ranked, street-by-street list so your reps spend their hours on the highest-probability doors.

Within a winning area, rank streets and blocks by a simple composite: inferred roof-age band, condition signals from imagery, owner-occupancy, and whether any verified storm swath touches them. Then route the canvass so reps work the ripe blocks first and don't zigzag. A few operational rules that pay off:

• Cluster, don't scatter. A rep should finish a block before moving on. Density of effort beats coverage.
• Lead with the strongest blocks. Your first impression of a neighborhood is set by your first conversations. Start where you're most likely to find real, documentable need.
• Tag and skip the fresh roofs. If imagery or permits show a recent replacement, mark it and move on. Knocking a three-year-old roof wastes the rep and annoys the homeowner.
• Feed results back. Track close rate and damage-found rate by block and re-rank. The map should get smarter every week.

Where RoofPredict fits

Everything above, roof-age ranges across a whole county, storm physics modeled per roof, owner-occupancy and density layered in, then ranked into a knockable list, is real work to assemble by hand from a half-dozen public sources. That's the gap RoofPredict is built to close. It tells roofing contractors which roofs are due, house by house: a roof-age range per address inferred from aerial imagery, plus storm exposure modeled per roof rather than per county, so you can see which specific roofs the storm most likely wore out and which are simply aging out. It enriches your own CRM or mailing list with those roof-age and storm signals, and ranks doors, routes, and lists so crews target the roofs that are actually ripe.

It is not a lead-buying service, and it's worth being clear about the limits. The roof age is a range, not a date, because that's what the data honestly supports. The storm read is odds, not proof, it raises or lowers the probability that a given roof has real damage; it does not promise damage on any specific roof or anything about a homeowner's claim. You still send a person up a ladder to verify before you scale, and you still win on the quality of your documentation. What it buys you is the part that's hard to do by hand: turning a fuzzy region into a sorted list of the addresses most worth your crew's time, so your service-area decision becomes an operational advantage instead of a line on a map.

Special cases: commercial, rural, and multi-location

The retail single-shop model above is the common case, but three situations change the math.

Commercial and low-slope

Commercial roofing flips the density logic. Jobs are larger, rarer, and worth a much longer drive, so your "service area" for commercial can be far wider than your residential sales area, because one TPO or modified-bitumen project justifies the windshield time that no single re-roof would. Density of qualifying roofs matters less; relationships, building age, and the property-manager network matter more. If you run both, draw two separate maps. The mistake is letting the commercial radius set your residential ambitions, or vice versa.

Rural and low-density markets

In a rural market you may not be able to hit the density targets that make tight drive-time bands work. The honest answer is that your numbers will be different: higher travel cost per job, lower doors-per-shift, and a service area that has to stretch geographically to contain enough roofs. The lever that saves rural economics is average ticket and routing discipline. Bigger rural roofs, batched scheduling so a crew does several jobs per trip out, and a referral-heavy sales motion that lowers your cost to create demand. Don't import metro density assumptions into a rural plan; you'll either price wrong or starve the crews.

Standing up a second location

Once a target area sits beyond 60 minutes, you're not extending, you're opening a branch, and it needs its own break-even analysis: a yard or storage, crews who live near the work, separate marketing, and enough qualifying-roof density and storm exposure within 30 minutes of the new base to support it on its own. The same scoring model applies, you're just choosing a new center point. Companies get hurt when they treat a far market as a stretch of the existing service area, eat the drive time for a year, and never build the local density that would have made it profitable. If it's far enough to need a branch, plan a branch.

Matching marketing channels to the area you chose

The service-area decision and the marketing decision are the same decision wearing two hats. Once you've picked the ground, the channels that work are dictated by the area's makeup, and spending the wrong way on the right area still wastes money.

In a dense, ripe, owner-occupied core, the cheapest demand is physical and local: door knocking, yard signs on every completed job, neighbor letters around an active install, and tight-radius direct mail to the streets your data flagged as ripe. Density is what makes these channels cheap, because the cost per touch drops when the touches are clustered. A canvasser in a dense core is the lowest-CAC tool you own. Direct mail to a hand-picked street list of likely-due roofs beats a blanket carrier-route blast by a wide margin, because you're paying postage only on doors that can plausibly buy.

In a storm area, speed and presence beat polish. The buying event is external and urgent, so the channels that win are rapid canvassing of the verified swath, a visible truck-and-sign footprint, and fast scheduling of inspections while the event is fresh in homeowners' minds. The window is short; the company that's documenting roofs the week after a verified event captures the demand the company that shows up a month later won't.

In a referral-driven, sticky retail market, paid acquisition is expensive and slow to crack, so you lean on past-customer referrals, online reviews, and reputation. Here the service-area logic is about being present and recommendable over years, not about a burst.

The practical rule: don't pick a uniform marketing playbook and apply it everywhere. Read each chosen sub-area's density, age, and storm profile, then assign the channel that's cheapest for that profile. A single company commonly runs canvassing in the dense core, targeted mail in the secondary band, and a storm-response motion in the elastic storm area, all at once, because the areas are different and demand different tools.

The CAC and payback math that should gate every area

Every candidate area has a cost to acquire a customer (CAC) and a contribution margin per job, and the relationship between them is what actually tells you whether to plant a flag. The scoring model ranks; this math sanity-checks the winner before you fund it.

Work a rough example. Suppose in a dense, lightly-competed core a canvasser knocks 70 doors a shift, sets inspections on a small fraction of them, and the season nets one signed retail job for every, say, 120 doors knocked. If your fully-loaded canvassing cost runs a certain dollar amount per door, your CAC is that cost times 120, plus whatever closing and inspection labor rides on top. Against a retail job carrying a healthy contribution margin, that CAC is a small fraction of the margin, the area pays back fast, and you scale spend.

Now run the same math in a saturated version of that area, where you're the third sign on the lawn. The doors-per-signed-job number gets worse, maybe one in 200 instead of one in 120, because the ripe homeowners were already pitched. Your CAC nearly doubles on the same margin. That's how a high-scoring-on-paper area quietly loses money: the demographics are great, but the competitive density wrecked the conversion denominator. The under-served seam, even with slightly weaker demographics, can carry a far better doors-per-job ratio and therefore a better payback.

The discipline is simple: estimate doors-per-signed-job and CAC per candidate area, compare CAC to contribution margin, and require the payback to clear a threshold before you commit steady spend. Far areas have to overcome higher travel cost in their margin; saturated areas have to overcome a worse conversion ratio; storm areas can carry a higher CAC for a short window because the volume and urgency are temporarily high. Put numbers on it and the right boundary usually picks itself.

What pros get wrong

A few patterns show up over and over, expensive every time.

Drawing a circle, not a drive-time shape. A clean radius on a map looks tidy and lies to you. Real travel time is shaped by highways, rivers, and traffic. Map minutes, not miles.

Confusing the storm area with the sales area. Chasing a storm is fine. Letting last year's storm permanently inflate your steady footprint into places with no density is how you quietly destroy crew utilization.

Using year-built as if it were roof age. On older housing stock, build year is a weak proxy. Without checking permits, storm history, and imagery, you'll market hard to neighborhoods that re-roofed five years ago.

Ignoring competition density. The ripest neighborhood on paper can be the worst place to spend a marketing dollar if it's already saturated. Look for the under-served seam.

Marketing everywhere at once. Thin spend across thirty zip codes builds awareness nowhere. Concentrate spend where the data is strongest until you own it, then expand.

Never re-scoring. Roofs age, neighborhoods get re-roofed after storms, competitors come and go. A service area is not a one-time decision; re-run the scoring at least annually and after every significant storm season.

Promising things you can't, around insurance. In storm-heavy territory the temptation to over-promise on claims is strongest exactly where the legal exposure is highest. Sell the documentation and the estimate, never the outcome.

A 10-step playbook to choose your service area

Put it all together into a sequence you can run:

1. Set your center point(s). Shop, yard, or the home base of your crews. This anchors all drive-time math.
2. Draw drive-time bands, 30 / 45 / 60 minutes, using real travel time, not radius circles.
3. Pull housing data from the Census ACS at tract or block-group level: owner-occupancy, detached single-family share, counts, density.
4. Estimate roof-age ripeness from build year, permit records, storm history, and imagery spot-checks. Work in ranges and confidence bands.
5. Layer storm exposure, baseline historical hail and wind frequency for the long view, plus any verified recent swaths for the near term.
6. Scout competition in each candidate area: density, positioning, saturation, and open lanes.
7. Score and rank candidate areas with the weighted model. Cut the dead-inventory and saturated zones.
8. Choose your three boundaries, sales, install, and storm, and put them on one map in three colors.
9. Drill the winners to street level and build a ranked, clustered, knockable list. Tag and skip fresh roofs.
10. Measure and re-score. Track close rate and damage-found rate by block, feed it back, and re-run the whole thing annually and after major storms.

Do this once with discipline and your service area stops being the blob that formed by accident. It becomes the highest-leverage operating decision you make: the boundary that puts your crews on the most roofs, your reps in front of the readiest homeowners, and your marketing dollars where they compound. Everything downstream, close rate, margin, utilization, gets easier because you chose the ground instead of inheriting it.

FAQ

How big should a roofing service area be?

Think in drive time, not miles. For a single-shop residential operation, a workable default is a core sales area inside 30 minutes of your shop or yard, a secondary band from 30 to 45 minutes for good inbound leads, and a 45-to-60-minute band reserved for big jobs and verified storms. Beyond 60 minutes you're opening a branch, not extending an area. Size the bands so you have enough qualifying owner-occupied roofs at decent density to keep crews busy without long windshield time. Rural markets may need wider bands; dense metros can compress them.

What data should I use to pick a roofing territory?

Five inputs do most of the work: roof-age distribution (which roofs are nearing the end of serviceable life), owner-occupied detached single-family share, density of qualifying roofs per square mile, storm exposure (both historical frequency and verified recent events), and competitive saturation. Most of it is public: the US Census ACS for housing and ownership, NOAA Storm Prediction Center and the National Weather Service for storm events, county assessor and permit portals for build years and re-roof dates, and aerial imagery for roof condition and size.

How do I estimate roof age across a whole neighborhood?

Combine sources and work in ranges, not exact dates. Build year from assessor data gives the original roof date but decays as a proxy on older homes. Permit portals timestamp re-roofs where they exist. Known storm dates tell you which homes likely got replaced in a past event. Aerial imagery shows fade, granule loss, patching, and prior repairs, and time-stacked historical imagery can bracket when a roof changed. The output should be a confidence band per block, such as 'most likely 18 to 24 years,' never a precise install date you can't actually prove.

Should roof age be a date or a range?

A range. The data, build year, imagery, and partial permit records, supports a band, not a birthday. Telling a homeowner 'your roof was installed in 2003' invites them to correct you the moment they remember a later replacement, and it erodes trust. Saying 'the imagery and records put this roof most likely in the 18-to-24-year band, and the fade and granule loss are consistent with a roof near the end of its serviceable life' is both more honest and more persuasive, because it pairs a defensible range with what you can actually see.

How do I know if a storm actually hit an area worth working?

Don't trust the headline; verify the swath. Pull event reports from the NOAA Storm Prediction Center and your local National Weather Service office, and look at reported hail sizes and wind gusts mapped to real coordinates rather than county-level callouts. Hail is intensely local, so intersect the damaging-intensity footprint with where your qualifying roofs sit, then inspect a handful of properties across the suspected area to confirm real, documentable damage before you commit a sales team. Time-box the campaign and set a close rate that tells you when to pull out.

What can a roofing contractor legally say about insurance claims?

You may inspect a roof, document damage thoroughly with photos and measurements, write an accurate itemized estimate to repair your own scope (ideally aligned to standard estimating line items), and state facts about your scope to the carrier. Then you hand it to the homeowner, who files, and the insurer decides coverage. You may not, for a fee, negotiate or 'handle' the claim, interpret the policy or what's covered, promise a specific payout or approval, promise the deductible will be waived or absorbed, advertise a 'free roof,' or represent the homeowner against their insurer. Those activities can constitute unlicensed public adjusting, which is regulated by state departments of insurance.

Is it better to pick a dense area with competition or an open area with fewer roofs?

It depends on saturation and durability, which is why you score rather than guess. A dense, ripe neighborhood that already has two or three established crews working it will give you a high cost to acquire each customer because you're the third sign on the lawn. An under-served pocket, often an aging neighborhood that's awkward to reach from where the big competitors are based, frequently converts better and costs less to win even if it scores slightly lower on paper. Use the data to find ripe-but-under-worked seams between competitors' cores.

How much does drive time really affect roofing profitability?

More than almost anything else you control. A crew that loses an extra hour a day to travel loses roughly 250 production hours a year. If a productive crew-hour is worth around 350 dollars in installed revenue, that's about 87,500 dollars of lost capacity per crew per year, and you paid wages and fuel for it. Across multiple crews it compounds fast. That's why you draw service-area boundaries in minutes of real travel time, concentrate density inside your core band, and treat far markets as opportunistic or as a separate branch.

How often should I re-evaluate my service area?

At least once a year, and again after any significant storm season. Roofs age into and out of the ripe band, neighborhoods get re-roofed after storms (which removes inventory), and competitors enter and leave. Track close rate and damage-found rate by block, feed those results back into your territory scoring, and re-rank. A service area is an operating decision you maintain, not a boundary you set once and forget.

How is choosing a commercial roofing service area different from residential?

Commercial flips the density logic. Low-slope and commercial jobs are larger, rarer, and worth a far longer drive, so the qualifying-density-per-square-mile math matters much less than relationships, building age, and the property-manager network. One TPO or modified-bitumen project can justify windshield time that no single residential re-roof would. If you run both lines, draw two separate maps and don't let the wide commercial radius inflate your residential ambitions or vice versa.