How to Pick a Roofing Farm Area That Actually Pays Off

Most roofers pick a farm area the same way they pick a lunch spot: whatever is close, whatever they drove past, whatever a buddy mentioned closed well last spring. Then they pour six months of gas, payroll, and postage into it and wonder why the close rate is soft. The neighborhood you farm is the single biggest lever on your cost per job, and almost nobody treats it like the decision it is.

A farm area is a defined set of streets you work on purpose, over and over, until you own the mindshare there. You knock it, you mail it, you put yard signs up, your trucks get seen, and eventually homeowners start saying your name when a neighbor asks who did their roof. Done right, a good farm compounds: every job you close makes the next one easier because the proof is sitting on the roof three doors down. Done wrong, you are renting attention in a place that will never pay you back.

This is the operational version of that decision. Not motivational fluff about "working hard" and "hustle" — the actual variables, the math behind them, how to score a neighborhood before you commit a season to it, and the edge cases that quietly kill territories that looked great on paper. By the end you should be able to drive into any zip code, pull a few numbers, and know within an hour whether it is worth your crew's time.

What a farm area actually is (and what it is not)

A farm area is a geographic block you commit to. The word comes from the idea that you are cultivating ground — you plant (mail, knocks, signs, sponsorships), you tend it for months, and you harvest jobs that come from accumulated familiarity rather than a single cold touch. The opposite of farming is chasing: driving forty minutes to a one-off lead, working a storm two states away for three weeks, buying a name off a lead site that five other companies also bought.

Farming is not the same as a sales territory on a map a software drew for you, and it is not a marketing radius. A real farm has three properties:

• It is small enough to saturate. If you cannot realistically touch every door in it four to six times a year across knocks, mail, and signage, it is a region, not a farm. Most healthy residential farms are 800 to 2,500 homes. A green canvasser can cover roughly 80 to 120 doors a day; a two-person team can blanket 1,500 homes in a couple of weeks and then circle back.
• It has a reason to convert. Age, storm exposure, home value, ownership patterns — something about the homes there makes a roof sale likely and the money real. We will spend most of our time here.
• It is defensible. You can be the company people there think of first. That means you can physically be present often enough, and the competition is not already three feet deep on every lawn.

What a farm is not: it is not "the whole city." It is not wherever the last lead happened to come from. And it is emphatically not a place you picked because it is wealthy and you assumed wealthy means easy. We will get to why that assumption burns people.

The seven variables that decide a farm area

Every neighborhood can be scored on the same handful of factors. None of them alone tells you to go; the combination does. Here they are in roughly the order they move the needle.

1. Roof age — the variable everyone gets wrong

A roof that does not need work will not buy work, no matter how good your pitch is or how nice the lawn looks. Age is the engine of the whole decision. A 20-year-old asphalt shingle roof in a wind-and-hail region is a candidate; a 6-year-old roof on the same street is a waste of a knock and a piece of mail you paid for.

The problem is that age is the one variable almost nobody can see from the curb or from a desk. Here is the trap that catches even experienced roofers:

The county records and Zillow tell you year built, not year roofed. A house built in 1994 may have been re-roofed in 2016. A 2005 build may still wear its original roof. Public property data, tax assessor files, and the listing sites all carry the construction date and quietly let you assume it is the roof date. In a neighborhood built out over three or four years, a quarter to a third of the homes have already been re-roofed at least once by the 20-year mark, and those re-roofs are invisible in the records. If you mail the whole tract because "it was built in 2003," you are paying to talk to a pile of homeowners who replaced their roof two years ago.

How pros actually estimate roof age:

• Tract build-out gives you a baseline. A subdivision built in 2001-2003 with original architectural shingles (typical rated life 20-30 years, but real-world failures cluster earlier in harsh climates) is hitting the replacement window now. That is a starting hypothesis, not a fact about any single house.
• Aerial imagery refines it. Granule loss, streaking, patched sections, mismatched planes, and obvious tarp/patch jobs show up from above. A roof that has been redone reads differently than one that has not — cleaner, more uniform, sometimes a different shingle line entirely.
• Permit records confirm it where they exist. Many jurisdictions require a permit for a re-roof. Pulling re-roof permit history for a tract tells you which addresses already bought — i.e., which to skip. (Permit coverage is wildly inconsistent by jurisdiction; treat absence of a permit as weak evidence, not proof the roof is original.)

The honest version of roof age is a range, not a date. "This roof is 18 to 22 years old" is a true, useful statement. "This roof was installed on March 4, 2004" is a claim nobody can back up from imagery, and pretending otherwise is how people lose trust. When you build a farm, you are looking for streets where a large share of homes fall into the late-life range, with as few recent re-roofs mixed in as possible.

2. Storm exposure — but model it per roof, not per map

If you do restoration work, hail and wind history is the second engine. A neighborhood that took a verified hail event in the last year is full of roofs that may now qualify for an insurance-funded replacement — and full of homeowners who do not yet know their roof took a beating, because hail damage is famously hard to see from the ground.

The mistake here is treating a hail map as the answer. A hail map shows you the general area a storm passed over. It does not tell you which specific roofs got worn out, and that distinction is the whole game. Within a single "hit" polygon, you will find roofs that caught direct impacts at a bad angle and roofs two blocks over that were shadowed by terrain, tree cover, or the storm's actual track and took almost nothing. Hail size, fall angle, wind direction, and roof slope all change what a given roof actually absorbed. Two houses on the same street can have completely different damage from the same storm.

The operational implications:

• Confirm the event before you build a storm farm. NOAA's Storm Prediction Center and the National Weather Service publish storm reports and local storm event data. IBHS and insurance-industry hail data corroborate severity. Do not commit a crew based on a homeowner's say-so or a single radar screenshot.
• Rank within the hit, do not blanket it. Inside a confirmed storm area, you still want the roofs that took the most punishment and were already old enough to fail. A 4-year-old roof that caught marginal hail is a hard sell and a marginal claim; a 17-year-old roof that took 1.5-inch stones is the door you knock first.
• Keep a non-storm farm too. Storm work is feast-or-famine and brings out-of-town swarms of competitors. The roofers who survive lean seasons have an age-based farm that produces work whether or not the sky cooperates.

A hard line that matters legally: your job on the storm side is to document damage and write an accurate repair estimate, then hand it to the homeowner. You inspect, you photograph, you measure your scope, and you prepare an estimate aligned to standard pricing. You do not file, negotiate, or "handle" the claim, you do not interpret the homeowner's policy or tell them what is covered, you do not promise an approval or a specific payout, and you never tell a homeowner the deductible will be waived, absorbed, or erased — and never advertise a "free roof." Doing those things for a fee is unlicensed public adjusting in most states, and it is exactly the behavior that gets roofers fined and homeowners hurt. The homeowner files; the insurer decides coverage. Pick farm areas where you can win on documentation and craftsmanship, not on promises you are not allowed to make.

3. Home value and equity — the deductible math

Roofing money has to be real for the homeowner, not only for you. Three sub-factors:

• Median home value vs. roof cost. A full residential tear-off and replacement is a five-figure decision. In a neighborhood where homes sell for $180,000, that is a meaningful chunk of the asset and homeowners feel it; in a $700,000 neighborhood it is a smaller share and an easier yes — if the roof actually needs work. Use Census American Community Survey median value and median household income for the tract to gauge whether the spend is in reach.
• Owner-occupancy. Renters do not buy roofs; landlords are slower and more price-driven. ACS gives you owner-occupied rate by tract. A farm that is 80% owner-occupied behaves completely differently than one that is 45%. Higher owner-occupancy generally means more people who can say yes at the door and more pride-of-ownership decisions.
• For storm work, the deductible relationship. A homeowner with a $1,000 deductible and a damaged 18-year-old roof has straightforward economics. A homeowner with a $4,000 wind/hail deductible (common on coastal and hail-belt policies) faces a real out-of-pocket decision even with an approved claim. You are not allowed to make the deductible disappear, so pick areas where the math works honestly — where the value of the work plus normal claim funding leaves a deductible the homeowner can and will pay.

4. Density — your cost per door

Density is the quiet profit-killer. Two farms can have identical roof age and value but wildly different economics because one is packed and one is spread out.

Think of your costs per touch. Knocking 100 doors that are 50 feet apart on a grid of straight streets is a productive day. Knocking 100 doors on acre-plus lots with long driveways and cul-de-sacs is half the doors in twice the time, with a tired rep. The same logic hits mail (route density discounts), yard-sign visibility (more eyeballs per sign on a busy grid), and drive time between appointments.

A rough density screen: pull the housing-unit count for the tract from the Census and divide by land area. Suburban grid neighborhoods that farm well often run 1,500 to 4,000 housing units per square mile. Below a few hundred units per square mile, you are in rural/estate territory where the per-door cost of canvassing climbs fast and farming logic starts to break down — those areas can still produce jobs, but you work them by appointment and referral, not by saturation.

5. Competition and saturation

Walk or drive the neighborhood and read the lawns. Yard signs, dumpster brands, magnetic truck signs, and door hangers tell you who is already farming here. Some competition is healthy — it means the area produces. But if every other lawn has a competitor's sign and you are showing up cold, you are the fourth date for every homeowner.

Signs to read:

• Fresh re-roofs everywhere = either a recent storm already worked (late to the party) or a strong farmer already owns the mindshare. Check which.
• One dominant brand = an entrenched farmer. You can still win, but you need a sharper angle (better documentation, faster response, a clearer reason to switch).
• Mixed, scattered signs = a fragmented market with room to become the name people remember.
• No signs at all in an aging tract = possible blue ocean, or possible reason nobody works it (HOA restrictions, low conversion, a quirk you need to find).

6. Accessibility and HOA friction

Gated communities, strict HOAs, and no-soliciting enforcement change your playbook. A neighborhood can score perfectly on age and value and still be a poor farm if you physically cannot knock it or if signage is banned. Some HOAs prohibit yard signs entirely, which kills one of your compounding mechanisms. Gated communities require referral and appointment motions instead of cold knocks. None of these are dealbreakers, but they change the cost and the method, and you want to know before you commit.

7. Roof type and complexity

The roofs themselves shape your margin and your crew fit. A tract of simple gable asphalt roofs is fast, repeatable work your crews can turn quickly. A neighborhood of steep, cut-up roofs with lots of valleys, dormers, skylights, and multiple stories is more profitable per job but slower, more dangerous, and harder for a green crew. Tile, metal, and slate are different businesses with different crews and different sales cycles. Pick farms whose dominant roof type matches the crew and the work you actually want to be doing.

Retail farms vs. storm farms — two different animals

Before you score anything, decide which kind of farm you are building, because the same neighborhood can be a great pick for one motion and a terrible pick for the other. Roofers blur these together and then wonder why their numbers do not make sense.

A retail farm runs on age. You are selling replacements to homeowners whose roofs have simply worn out — no storm required, no insurance funding assumed, the homeowner pays out of pocket or finances. The demand here is steady and predictable because roofs age on a schedule. A retail farm wants a high concentration of late-life roofs, solid owner-occupancy, and a value-to-cost relationship where homeowners can write the check. It produces all year, it does not bring out-of-town competition, and it builds slowly and durably. The downside is the sales cycle is longer and the ticket usually smaller, because there is no claim covering the bulk of the cost.

A storm farm runs on a confirmed weather event plus age. You are documenting damage so the homeowner can pursue insurance funding for a roof that genuinely took a beating. The demand is a spike — it appears after a storm and fades — and it brings every storm-chasing crew in the region into your backyard for a few weeks. The economics per job are usually better because a claim funds the work, but the window is short, the competition is fierce, and the legal tightrope is real. A storm farm wants confirmed recent events, old roofs inside the impact zone, and deductible math that works honestly.

The roofers who run a stable business carry both: an age-based retail farm that pays the bills in quiet seasons, and the readiness to stand up a storm farm fast when a verified event hits inside or near their footprint. If you only chase storms, you starve between them and you are always fighting the swarm. If you only farm retail, you leave the highest-ticket, claim-funded work on the table when the sky hands it to you. Pick your primary motion deliberately, and know that a single neighborhood's score changes depending on which one you are running.

There is also a hybrid that the best operators exploit: a retail farm you have already saturated becomes a storm farm with a head start the day a verified event hits it. You already have the relationships, the yard signs, the referral flow, and the trust. When the storm comes, you are not a stranger knocking on doors next to six out-of-town crews — you are the company that did three roofs on the street last year, already documenting damage before anyone else parks a truck. That is the strongest position in the trade, and it only exists because you farmed the area on age first.

A scoring model you can actually run

Gut feel is fine for the first drive-through. Before you commit a season, score it. Here is a weighted model that has held up across residential markets. Score each factor 1 to 5, multiply by the weight, and total it.

Multiply, total, and convert to a 100 scale. A practical read:

• 80-100: Commit. Build the full saturation plan.
• 60-79: Workable. Fix the weak factor or accept the tradeoff (e.g., great age, tougher competition — sharpen your angle).
• Below 60: Walk away or treat it as appointment-only territory, not a farm.

Worked example

Two neighborhoods, same metro, both look fine from the truck.

Neighborhood A — "Maple Crossing." Built 2001-2004, architectural shingles, tight suburban grid, 82% owner-occupied per ACS, median value comfortably above the cost of a replacement, confirmed 1.25-inch hail event eleven months ago, scattered competitor signs but no dominant brand, no HOA sign ban.

• Roof age: 5 (right in the 20-22 year window, original-build tract)
• Storm exposure: 4 (confirmed event, but you will still need to rank per-roof because terrain varies)
• Value & owner-occupancy: 5
• Density: 5 (clean grid)
• Competition: 4 (fragmented)
• Accessibility: 5
• Roof complexity: 4 (simple gables, fast work)

Weighted: (5×.25)+(4×.20)+(5×.15)+(5×.15)+(4×.10)+(5×.10)+(4×.05) = 1.25+.80+.75+.75+.40+.50+.20 = 4.65 → 93/100. Commit.

Neighborhood B — "Stonebridge Estates." Gorgeous, $650k+ homes, but built 2014-2017, acre-plus lots, gated, strict HOA with a yard-sign ban, no storm history, one luxury-roofing brand already on the few visible jobs.

• Roof age: 2 (mostly 8-11 year old roofs, not due)
• Storm exposure: 1 (no events, young roofs)
• Value & owner-occupancy: 4
• Density: 2 (spread out)
• Competition: 2 (entrenched premium brand)
• Accessibility: 1 (gated, sign ban)
• Roof complexity: 3

Weighted: (2×.25)+(1×.20)+(4×.15)+(2×.15)+(2×.10)+(1×.10)+(3×.05) = .50+.20+.60+.30+.20+.10+.15 = 2.05 → 41/100. Walk away — or work it strictly by referral in five years when those roofs age in.

The lesson the example is built to teach: the prettier, richer neighborhood scored less than half of the plain suburban grid, because wealth is not the variable — due roofs are. This is the single most common farm-selection mistake, and the math makes it obvious.

The data you pull, and where it comes from

You do not need expensive software to start. Here is the free and low-cost stack, what each source is actually good for, and its limits.

The gap in that stack is obvious once you have run it a few times: every source either tells you year built (wrong for roof age) or a storm footprint (wrong for per-roof impact), and none of them hands you a ranked list of which specific houses are actually due. You can assemble it by hand — pull the tract, cross-reference permits, eyeball imagery street by street — and plenty of good roofers do exactly that. It is slow, and it is the part most people skip, which is why most farms get picked on vibes.

Where RoofPredict fits

The whole point of a farm is to spend your gas, payroll, and postage on the doors most likely to convert and skip the rest. The bottleneck is that the two variables that decide "likely to convert" — roof age and per-roof storm impact — are precisely the two you cannot see from a desk. That is the specific gap RoofPredict is built to close.

RoofPredict scans an area and scores every roof on it by age (as a range, from aerial imagery) and the storms it has actually taken (modeled per roof, not pulled from a map), then ranks the houses so you can see which ones are worn out and which ones are not. Instead of mailing a whole tract because it "was built in 2003," you mail the homes whose roofs read as late-life and skip the ones that already got re-roofed. Instead of blanketing a hail polygon, you knock the roofs the storm actually wore out first. It models hail trajectory and wind house by house — a hail map shows where it hailed; this shows which roofs it likely aged out — and pairs that with the roof-age range so the two signals reinforce each other. You can also feed it your own existing list or CRM book and have it enriched with the same age-plus-storm signals, so old estimates and past customers whose roofs have now aged in surface back to the top.

Honest limits, because a farm decision deserves them: roof age comes back as a range, not an install date, and the storm model gives you odds, not proof — it tells you which roofs are most likely worn out, not a guarantee any single roof will pass an inspection or qualify for a claim. It does not measure the roof (that is a different category of tool) and it does not file or decide anything on the insurance side. What it does is turn "which streets should I even farm, and which doors inside them are due" from a multi-day manual cross-reference into a ranked list you can hand a crew. Used well, it is the targeting layer underneath everything else here — the scoring model above, run automatically across every roof in the area instead of by hand on a clipboard.

The rep economics nobody runs the numbers on

Picking the right farm is not only a marketing decision — it is a hiring and retention decision, and most owners never connect the two. The fastest way to burn through canvassers and reps is to point them at the wrong doors. Walk a green hire down a street of 6-year-old roofs and 50% renters and they get door after door of "we just did it" and "I rent," they make no money, and they quit inside three weeks. Point that same green hire at a street where one in three roofs is genuinely worn out and the homeowners are owners who can say yes, and they catch a few inspections in the first week, make some money, and stay.

Rep churn is one of the most expensive line items in a roofing sales operation, and it is largely downstream of farm selection. Run the math on your own shop: count what it costs to recruit, onboard, train, and ramp a canvasser, then count how many wash out before they ever pay that back. A huge share of that churn is not a people problem — it is a targeting problem. Reps quit doors that do not convert. The single highest-leverage thing you can do for retention is hand them a farm where the doors are actually due, so effort turns into money fast enough to keep them in the game.

There is a knock-on effect on what a green rep can credibly say. A new hire who knows nothing about roofs is dangerous at the door and the homeowner can smell it. But a green rep armed with a specific, true reason to be at this house — "the roofs on this street are around twenty years old and this is the window they start failing" — sounds like a vet without having climbed a single ladder. The reason has to be real; homeowners punish fake urgency hard. A farm picked on genuine roof age gives every rep, including the brand-new one, an honest opener that lands.

Targeting the mail, not the whole route

Direct mail in roofing lives or dies on who gets the piece. Every-door route mail is cheap per piece and tempting, and it is exactly how roofers light money on fire — you are paying to land in the mailbox of every recently re-roofed home, every renter, and every brand-new build in the route. The cost looks low per piece and is brutal per qualified impression.

The discipline is the same as the rest of the farm decision: mail the homes that are due, skip the ones that are not. A practical hierarchy of mail targeting, worst to best:

1. Blanket route mail — cheapest per piece, worst per qualified impression. Use only when your data is so thin you genuinely cannot tell the due homes from the rest, and even then only in a tract you have confirmed skews old.
2. Build-year-filtered mail — better, but inherits the build-year-equals-roof-age error. You will still pay to reach every re-roofed home in the tract.
3. Permit-and-imagery-filtered mail — you suppress the homes that show a recent re-roof permit or read as freshly re-roofed in imagery, and mail the rest. Meaningfully tighter.
4. Age-and-storm-ranked mail — you mail the homes whose roofs read as late-life, weighted up where a confirmed storm added impact, and you suppress the rest entirely. Highest cost per piece, lowest cost per qualified impression and per job.

The creative matters too, but it matters less than the list. A mediocre piece to a tightly targeted list of due homes beats a beautiful piece to a blanket route every time. Reference the neighborhood by name, lead with the honest reason the homeowner is getting it (the age of roofs on their street), and make the call to action a free inspection, not a hard pitch. Track response by farm and by list tier so you can see, in dollars, what tighter targeting is worth.

Building the farm once you have picked it

Picking is half the job. A good area still fails if you touch it once and move on. Farming works because of repetition and accumulated proof. Here is the saturation cadence that turns a chosen area into a producing farm.

The touch sequence

Farming is a multi-touch motion, not a one-and-done. A workable annual cadence for a 1,500-home farm:

1. Aerial/age pass (week 0). Rank the homes by likely due-ness before anyone burns a tank of gas. This is where you decide which 500 of the 1,500 get the heavy treatment and which 1,000 get light touches.
2. First knock wave (weeks 1-3). Cover the ranked-due homes first. The goal of the first knock is rarely the sale — it is the inspection, the relationship, and the door hanger that gets left when nobody is home.
3. Mail drop (week 4). Hit the same homes with a piece that references the neighborhood by name. Route-dense mail is cheap; targeted mail to ranked-due homes is cheaper per qualified impression.
4. Job + yard signs (ongoing). The first close in the farm is worth more than the revenue — the sign and the crew on the street are proof for the next twelve doors. Get permission, put the sign up, keep it up.
5. Re-knock and referral wave (weeks 8-12). Circle back. Knock the neighbors of every job. "We just did the Hendersons' roof on Oak — wanted to take a look at yours while we are on the street" is the single highest-converting opener in farming, and it only exists because you saturated.
6. Seasonal re-touch. Twice a year minimum on the due homes. Roofs that were borderline last spring are a year closer to failure now.

Saturation math

Why small-and-saturated beats big-and-thin. Say you have budget to make 6,000 touches this season.

• Spread thin: 6,000 touches across 6,000 homes = one touch each. Marketing research is blunt about single cold touches — recall and response are low, and one piece of mail or one knock to a stranger converts at a fraction of a percent. You will close a handful and have nothing compounding.
• Saturated: 6,000 touches across 1,500 homes = four touches each, layered (knock + hanger + mail + neighbor-referral). Familiarity builds, the signs go up, the referrals start, and by the second pass you are the company people there recognize. The same budget produces a multiple of the jobs because each touch lands on a warmer, better-targeted door.

The whole argument for farming is in that contrast. Saturation only works if the area is small enough to saturate and the doors inside it are due enough to convert — which loops back to picking right in the first place.

Tracking whether the farm is working

Farm one area long enough and you can measure it like an asset. Track per farm:

• Cost per knock and cost per qualified inspection (not merely cost per lead).
• Inspection-to-sale rate — if it is low in a high-age farm, the problem is your pitch or your documentation, not the area.
• Referral rate — the truest signal a farm is maturing. When 20%+ of new jobs in an area come from neighbor referrals, the farm is compounding and you should not abandon it.
• Sign-to-call attribution — ask every inbound how they heard of you and tag the farm.

If a farm's numbers are climbing season over season, double down. If they are flat after two full cycles of honest saturation, re-score it — usually the roof-age profile was thinner than you thought, or a competitor got entrenched faster.

The documentation workflow that wins a storm farm legally

If your primary motion is storm restoration, the farm you pick is only as good as the documentation you produce inside it — and the documentation is the part you are fully allowed to control, unlike the claim itself. Here is the workflow that keeps you on the right side of the line while still winning the work.

1. Confirm the event first. Before you knock a single door, verify the storm through NOAA's Storm Prediction Center, the local National Weather Service office, and the NCEI storm events database. Note the date, the reported hail size, and the wind speeds. This is your factual basis for being on the street, and it protects you from chasing a storm that barely happened.
2. Inspect and photograph thoroughly. On the roof, document hail bruising, fractured mats, granule displacement, soft spots, damaged vents and flashing, and collateral damage on soft metals (gutters, downspouts, screens, AC fins) that corroborates the event. Photograph everything with scale references and locations. Test squares are standard practice; mark and shoot them.
3. Measure your scope. Capture accurate measurements for the work you would perform — squares, pitch, penetrations, accessories. This is your scope, stated as facts about the work you would do.
4. Write an accurate, standard-aligned estimate. Prepare a repair estimate using the standard pricing conventions the industry and carriers recognize. It reflects what it actually costs to restore the roof to its prior condition. You are stating facts about your scope and your price, which you are entitled to do.
5. Hand it to the homeowner. Give the homeowner the documentation and the estimate. Explain plainly that they file with their insurer and the insurer decides coverage. You are providing thorough documentation and an honest estimate, not adjusting the claim.

What keeps you legal is staying on the documentation-and-estimate side of every interaction. The do-not-say list, taught plainly to every rep, is part of farm discipline:

• Do not say the claim is "approved" or promise it will be — you cannot know and cannot decide it.
• Do not interpret the homeowner's policy or tell them what is or is not covered — that is the insurer's and a licensed adjuster's role.
• Do not promise a specific payout amount.
• Do not say the deductible will be waived, absorbed, eaten, or made to disappear — that is illegal in most states and a fast way to lose your license and the homeowner's money.
• Do not advertise or imply a "free roof."
• Do not offer to negotiate or "handle" the claim with the carrier on the homeowner's behalf for a fee — that is unlicensed public adjusting.

State insurance departments are explicit about this. Public adjusting is a licensed activity, and a contractor who negotiates or adjusts a claim for compensation without that license is breaking the law. The Texas Department of Insurance and the National Association of Insurance Commissioners both publish plain-language guidance on what separates a contractor's legitimate role from public adjusting. Pick storm farms where you can win on the quality of your documentation and your craftsmanship — because those are the things you are allowed to compete on, and they are also the things that build the referral flow that outlasts any single storm.

What pros get wrong

The mistakes are consistent across markets, and they are all selection mistakes that no amount of sales skill fixes after the fact.

• Chasing wealth instead of due roofs. The Stonebridge Estates trap. New construction in a rich zip is the worst farm in the world: every roof is years from needing you. Pick due, not pretty.
• Trusting build year as roof age. The number one data error. A tract "built in 2002" that already got worked after a 2019 storm is full of 5-year-old roofs now. Always check for the re-roof signal — permits, imagery, fresh shingles on the drive-through.
• Confusing a hail map with damage. Blanketing a storm polygon wastes effort on shadowed, low-impact, and too-new roofs inside the hit. Rank per roof.
• Farming too big. "My territory is the whole north side" is not a farm, it is a fantasy. You cannot saturate it, so you never compound, so you are always cold-touching strangers.
• One touch and gone. Treating mail or knocks as a single campaign instead of a multi-touch season. The jobs are in touches three through six.
• Ignoring HOA and access reality. Committing to a gorgeous tract you later discover bans signs and gates the entrance, killing two of your compounding mechanisms.
• No measurement. Running three farms and not knowing which one actually produces, so budget gets spread by feel instead of by result.
• Crossing the legal line on storm work. Promising approvals, waiving deductibles, advertising free roofs, or "handling" the claim. It is unlicensed public adjusting, it gets you fined, and it picks the wrong fight. Win on documentation and craftsmanship in areas where the honest math works.

A 60-minute field checklist

Drive into a candidate neighborhood and run this. An hour gets you a defensible go/no-go.

1. Pull the tract before you leave. ACS median value, median income, owner-occupancy, housing units, land area. Note density (units ÷ square miles).
2. Pull build-out year from the assessor/GIS. Hypothesize the roof-age window — but flag it as build year, not roof age.
3. Check storm history in NOAA/NCEI for confirmed events in the last 1-3 years. Note size and date.
4. Pull re-roof permits if the jurisdiction posts them. These are your skip list.
5. Scan imagery for re-roof tells and condition signals across a sample of streets.
6. Drive it. Read the lawns for competitor signs. Note driveway length, lot spacing, gates, no-solicit signage, HOA sign rules.
7. Spot-check a dozen roofs in person. Granule loss in gutters, streaking, curling, patches — confirm the imagery read.
8. Score it on the seven-factor model. Total to 100.
9. Decide: commit (80+), fix-the-weak-factor (60-79), or walk/appointment-only (below 60).
10. If commit, draw the boundary at a size you can saturate four-to-six times this year — typically 800-2,500 homes — and write the touch cadence.

Putting it together

The roofers who win farming are not the ones who knock the hardest or mail the most. They are the ones who pick the right ground, then saturate it patiently until they own it. The right ground is defined by due roofs — late-life age, real per-roof storm impact where applicable — sitting in a dense, accessible, owner-occupied area you can be present in often enough to become the name people remember. Wealth, looks, and proximity are distractions; age and impact are the signal.

The hard part has always been seeing the signal, because the two variables that decide it are the two you cannot read from the curb or the county record. Pulling the tract, cross-referencing permits, and eyeballing imagery street by street gets you there by hand. A tool that scans the area and ranks every roof by age range and per-roof storm impact gets you there faster — and either way, the discipline is the same: spend your season on the doors that are actually due, and skip the ones that are not. Pick the ground on purpose, prove it with the math, and farm it like you mean to own it.

FAQ

How big should a roofing farm area be?

Small enough to saturate four to six times a year across knocks, mail, and signage. For most residential roofers that lands between 800 and 2,500 homes. A two-person team can blanket roughly 1,500 homes in a couple of weeks and then circle back, which is the repetition that makes farming compound. Anything you can only touch once is a region, not a farm, and it will never build the familiarity and referral flow that farming relies on.

Why can't I just use Zillow or county records for roof age?

Because those sources carry the year the house was built, not the year the roof was installed. Re-roofs do not show up in property records or listing sites, so in any tract that is 18-plus years old, a meaningful share of homes have already been re-roofed and look just as old in the data as the ones that have not. Mailing or knocking a whole tract on build year means paying to talk to homeowners who replaced their roof recently. Use build year as a starting hypothesis, then refine it with re-roof permits, aerial imagery, and an in-person spot check.

Is a hail map enough to pick a storm restoration farm?

No. A hail map shows the general area a storm passed over, not which specific roofs got worn out. Inside a single confirmed hit, hail size, fall angle, wind direction, terrain, tree cover, and roof slope mean two houses on the same street can have very different damage. Confirm the event through NOAA or the NCEI storm events database, then rank roofs within the hit by both likely impact and existing age, so you knock the old, hard-hit roofs first instead of blanketing every door in the polygon.

Should I farm wealthy neighborhoods?

Only if the roofs there are actually due. Wealth is not the variable that drives roof sales; due roofs are. A high-value new-construction tract is one of the worst farms because every roof is years from needing work, while a plain suburban grid built 20 years ago can outscore it by more than double on a proper model. Home value matters as a secondary factor for whether the spend is in reach and the deductible math works, but it never overrides roof age.

How many times should I touch a farm area in a season?

Plan for four to six layered touches per home on your due list across the season: a knock wave, door hangers where nobody answers, a targeted mail drop, yard signs from every job, and a re-knock plus neighbor-referral wave. The jobs are rarely in touch one; they cluster in touches three through six as familiarity builds and proof accumulates on the street. Single cold touches convert at a fraction of a percent, which is why small-and-saturated beats big-and-thin for the same budget.

What data do I need to score a neighborhood, and is it free?

Most of it is free. The U.S. Census ACS gives median value, income, owner-occupancy, and housing-unit density by tract. The county assessor or GIS portal gives build year and lot data. NOAA's Storm Prediction Center and the NCEI storm events database confirm hail and wind events. City and county permit portals show re-roof permits where they exist. The one thing none of these hand you directly is a ranked list of which specific homes are due by roof age and per-roof storm impact, which you assemble by cross-referencing or with a targeting tool.

How do I tell if a competitor already owns a neighborhood?

Drive it and read the lawns. Scattered, mixed yard signs mean a fragmented market with room to become the name people remember. One brand on every visible job means an entrenched farmer; you can still win, but you need a sharper angle such as faster response or better documentation. A wave of fresh re-roofs everywhere usually means a recent storm already got worked, so you may be late. No signs at all in an aging tract is either open ground or a clue something there suppresses conversion, like an HOA sign ban.

Can roof-age targeting tell me the exact age of a roof?

No, and you should be skeptical of anyone who claims it can from imagery alone. Honest roof-age estimation gives you a range, such as 18 to 22 years, based on tract build-out, aerial condition signals, and permit history. A range is genuinely useful for deciding which homes to farm and which to skip; a precise install date is a claim nobody can back up without the actual permit or homeowner record. Treat age as a strong probability signal, not a fact about any single roof.

What can I legally say when I farm a storm-damaged area?

You can inspect the roof, document and photograph damage, measure your scope, and prepare an accurate repair estimate aligned to standard pricing, then hand that estimate to the homeowner. What you cannot do, in most states, is file or negotiate the claim, interpret the homeowner's policy or tell them what is covered, promise an approval or a specific payout, or tell them the deductible will be waived, absorbed, or erased — and you cannot advertise a free roof. Those acts are unlicensed public adjusting. The homeowner files and the insurer decides coverage; you win on documentation and craftsmanship.

How do I know when to abandon a farm area?

Measure it like an asset over at least two full saturation cycles. Track cost per qualified inspection, inspection-to-sale rate, and especially neighbor-referral rate. If referrals are climbing and cost per job is falling season over season, double down. If the numbers stay flat after two honest cycles, re-score the area; the usual culprits are a thinner roof-age profile than you assumed or a competitor who got entrenched faster than you did. Reserve abandonment for areas that fail the math, not for ones you simply got impatient with.