How to Find Roofing Jobs Before You Pay for Advertising

Most roofing companies start spending money on advertising long before they have squeezed the free work out of the assets they already own. A new owner signs up for a pay-per-lead service in month two, then wonders why the same homeowner is fielding calls from four other crews who bought the identical lead. A growing company hands a credit card to an agency, gets a slick dashboard, and still can't tell you which of those clicks turned into a signed contract.

There is a cheaper, slower, and far more durable way to keep your crews busy: find the jobs that are already sitting in front of you. Your old customer list. The streets you've already driven. The neighbor of the roof you replaced last August. The estimate you wrote in 2022 that the homeowner never moved on. None of those cost a dollar in media spend, and most of them close at a higher rate than anything you'll buy.

What follows is the actual order of operations a disciplined roofer uses to book work before the advertising budget ever comes out — what to do first, second, and third, with the numbers, scripts, and checklists to run each play. The thread running through all of it is ownership: a job you generate yourself is a job you control, and nobody resold it to your competitor down the street.

Why advertising is the last lever, not the first

Advertising is not bad. It's just expensive, and it's the lever most owners reach for because it feels like action. You write a check, leads appear, and it looks like progress. The problem is what you're actually paying for.

When you buy a shared lead from an aggregator, you're buying a homeowner who clicked one form and got sold to a handful of contractors at once. You're now in a price race with crews who have no relationship with that person either. When you run your own paid search or social ads, you're paying for clicks — most of which are not in the market, are tire-kickers, or are the wrong roof entirely. Either way, customer acquisition cost climbs, and the work you win is the work where you had the least leverage.

Compare that to the homeowner who already has your invoice in a drawer, or whose neighbor just watched your crew tear off and dry-in a roof in a single day. That person isn't comparison-shopping five strangers. They know your truck, your name, and your work. The close rate is multiples higher and the price pressure is far lower.

So the sequence matters. Spend the free and near-free channels to exhaustion before you rent attention. A rough hierarchy, cheapest to most expensive per booked job:

The top of that table is where the cheap, high-trust work lives. Most companies skip straight to the bottom because the top requires discipline and follow-through, and the bottom just requires a card number. The companies that grow profitably do the top first, every time, and only layer paid spend on once the free channels are genuinely tapped.

Play 1: Mine the money already in your book

Your customer relationship system — even if that "system" is a shoebox of invoices or a spreadsheet — is the single most underworked asset in a roofing company. The people in it have already paid you once. They live near other roofs. They had repairs you flagged but never did. This is the first place to look, and most roofers never look at all.

The three buckets hiding in your records

Bucket 1: Past full-replacement customers. A roof you replaced has gutters, flashing, ventilation, skylights, and chimney work that ages on its own schedule. The homeowner trusts you. They are also your single best source of referrals because they live on a street full of houses the same age as theirs was when you replaced it. Subdivisions get built in waves; if one roof on the cul-de-sac was 20 years old, the rest are within a year or two of it.

Bucket 2: Repair-only and maintenance customers. Someone who had you fix a leak or replace a section is a replacement candidate who hasn't crossed the line yet. Track when you did the repair and the age of the roof at the time. A 17-year-old roof you patched two years ago is a 19-year-old roof now.

Bucket 3: Dead estimates. This is the most overlooked pile of money in the building. Every estimate you wrote that didn't close is a homeowner who, at some point, was actively shopping for a roof. They didn't go away because they stopped needing a roof — they stalled on budget, timing, an insurance question, or they just got busy. A meaningful share of stalled estimates are still live a year or two later, and you already did the measuring and the relationship work.

A worked re-engagement workflow

Here is a concrete way to run the book over four weeks, assuming a company with a few hundred records.

1. Export everything into one list. Pull every name, address, phone, email, job type, job date, and estimate amount you can find. Don't worry about it being clean yet.
2. Tag each record into the three buckets above, plus a fourth tag for "roof age unknown."
3. Sort by likely roof age. For past replacements, the roof age is years-since-you-did-it. For repairs and dead estimates, you need the age of the existing roof — which is the hard part, and the section on roof-age data below solves it.
4. Prioritize the top 50. Anything where the roof is plausibly 15+ years old, plus every dead estimate from the last 24 months, goes to the top.
5. Call, don't email, the top tier. Email is for the long tail. The top 50 get a personal phone call from a person who can speak to their specific address.

A simple script for a dead estimate that respects the homeowner's time:

"Hi Mrs. Alvarez, this is Dave from Summit Roofing — we came out and looked at your roof back in spring of 2023 and put together a number for you. I'm not calling to push anything; I was reviewing our older files and your roof was getting up there in age even then. I wanted to check whether you ever got it handled, and if not, whether it'd be worth me swinging by to take another quick look so you know where you stand. No charge for that."

That call works because it's specific, honest, and low-pressure. You're not pretending to have a deal expiring at midnight. You're a professional who remembered them and noticed something real about their roof.

For the long tail — the records that don't justify a phone call — a short, plain email or postcard that references the relationship ("We replaced your roof in 2014; here's what to watch for at the 10-year mark, and here's a referral reward if you know a neighbor who needs us") keeps you top of mind for pennies.

What a small list is actually worth

Run the math so this stops being abstract. Say you have 400 records and you work them honestly:

• 60 are dead estimates from the last two years. You reach 40, book 8 inspections, and close 3. At a $14,000 average replacement, that's $42,000 from phone calls.
• 120 are past replacement customers now 8–14 years out. You're not selling them a roof yet, but you ask each for a referral and seed a reward. Even a 5% referral rate is 6 warm introductions.
• 80 are old repair customers. You re-inspect the oldest 20, find 5 that have crossed into replacement territory, close 2. Another $28,000.

That's real revenue from a list you already owned, before any ad dollar. The constraint is never the list — it's the discipline to work it.

Cleaning a messy book fast

Most roofers avoid working the list because it's a mess: duplicate entries, dead numbers, old addresses, half-finished records. Don't let that stop you. You don't need a perfect database to start calling; you need the top tier reachable.

• Deduplicate by address, not name. A homeowner who got a repair in 2019 and a quote in 2022 is one record, not two. Address is the stable key.
• Verify phone numbers in batches. A free or cheap number-validation pass strips out the disconnected lines before your caller wastes time.
• Fill the gaps that matter, ignore the ones that don't. You need address, a working phone, and roof age. Email is nice but optional for the top tier. Don't spend a week perfecting fields you'll never use.
• Stamp every record with a 'last touched' date the moment you start. From here forward you're building a living book, not a graveyard.

A practical segmentation grid keeps the calling focused:

A three-touch re-engagement sequence for the long tail

The records that don't justify a personal call still deserve a sequence, not a single shot. A simple three-touch cadence over six weeks does the work without a marketing team:

1. Touch one (week 1) — the relationship reminder. A short note that names the prior job: "We replaced your roof in 2015. At the 10-year mark, here are three things worth checking." Useful, not salesy.
2. Touch two (week 3) — the social proof. A photo of a recent job in a nearby neighborhood with a line about being in the area. This is where a yard-sign photo or a five-star review screenshot earns its place.
3. Touch three (week 6) — the soft offer. A free roof check before the season, plus the referral reward. "If your roof's getting up there, or you know a neighbor whose is, here's how to reach us."

Keep it to three touches. The point of the long-tail sequence isn't to convert everyone — it's to surface the handful who were ready to act and just needed a nudge, while keeping your name in front of the rest for pennies.

Play 2: Build a referral engine that actually runs

Every roofer says referrals are their best source of work, and almost none of them have a system. They wait for referrals to happen instead of engineering them. A referral engine is just a repeatable set of moments where you make the ask and make it easy to say yes.

The three moments to ask

1. At the moment of peak happiness — the final walkthrough. When the job is done, the yard is clean, and the homeowner is looking at their new roof, that is the highest-emotion point of the entire relationship. That is when you ask. Not a month later by email.
2. At the one-week follow-up. A short call or text to confirm everything's good gives you a second, natural ask after they've lived with the result.
3. At the seasonal check-in. Before storm season, a quick "we're booking up, if anyone you know has been putting off their roof, now's a good time" reminds past customers you exist.

Make the ask specific and easy

"Let us know if you hear of anyone" produces nothing because it asks the customer to do all the work. Hand them a reason and a tool:

• A specific prompt: "You mentioned your sister's place over on Maple is about the same age — would it be alright if I dropped a card by, or would you rather pass my number along?"
• A reward worth caring about: A meaningful referral reward — cash, a gift card, or a credit toward future work — paid when the referred job closes. Keep it simple and pay it fast. A roofer who pays a referral reward the day the referred job signs gets a reputation among customers that produces more referrals than any clever program.
• Something physical to hand over: A few branded cards, a yard sign that stays up for two weeks (with the homeowner's permission), a magnet for the fridge.

The yard-sign and job-site multiplier

A tear-off is the loudest advertisement you'll ever run, and it's free. Dumpster in the driveway, crew on the roof, new shingles going down — every neighbor sees it. Capture that attention deliberately:

• Put a clean, professional sign in the yard the morning the job starts, not the day it ends.
• Have whoever's running the crew leave a door hanger on the 10–20 nearest houses that simply says "We're working on your neighbor's roof this week — if you'd like us to take a look at yours while we're in the area, here's how to reach us."
• The neighbor canvass is covered in detail in the next play, but the point here is that the referral engine and the job-site presence reinforce each other. The happy customer vouches for you, and the visible work proves it.

Keep a tiny ledger of where referrals come from. Within a quarter you'll know which customers are your connectors — the handful of people who send you three or four jobs a year — and those people deserve your best reward and a personal thank-you.

Play 3: Work the neighbors of your job sites

This is the most efficient knocking you will ever do, because you've already paid to be on the street. Your crew is there. The roof going on next door is your proof. The only marginal cost is a few minutes and a door hanger.

Why neighbor canvassing converts

Roofs on a street tend to be the same age. Developments go up in phases, so when one roof has aged out, its neighbors are close behind. You also have something no cold knock has: visible, in-progress proof of your work fifty feet away. "We're doing the Hendersons' roof this week" is a credible opener that a stranger knocking cold can never match.

A repeatable neighbor-canvass routine

1. Define the radius. The ten houses on either side and the ten across the street — roughly 20–30 doors — is plenty.
2. Time it for the tear-off day. The roof is open, the activity is highest, the proof is loudest.
3. Lead with the neighbor, not the pitch. "Hi, I'm with the crew working on the Hendersons' roof over there. We had a couple openings while we're in the neighborhood and I'm offering the neighbors a free roof check — no obligation. Mind if I take a quick look at yours?"
4. Inspect honestly. If their roof is fine, tell them so and leave a card. Telling someone their roof has years left is the single most trust-building thing you can do, and that's the person who calls you in three years and refers you next month.
5. Document what you find. Photos of the actual condition, on their actual roof, are worth more than any brochure.

Knocking compliance you can't skip

Door-to-door selling is regulated, and ignoring the rules turns a free channel into a liability.

• The FTC Cooling-Off Rule gives consumers the right to cancel certain door-to-door sales of $25 or more (and $130 or more at their home) within three business days, and you're required to give written notice of that right. Build the cancellation notice into your contract and you stay clean.
• Local solicitation permits are common. Many cities and counties require a peddler's or solicitor's permit and have allowed hours. Check the municipality before you put a crew on doors.
• Respect "No Soliciting" signs and registries. It's the law in places and just good sense everywhere.

Put a one-page compliance checklist in every canvasser's folder: permit on file, hours respected, cooling-off notice in the contract, no-soliciting honored, honest inspection only.

Play 4: Knock the right doors with an old-roof list

Neighbor canvassing rides on a job you already have. Sometimes you want to generate that first job on a street, or work a target area before you've got a crew on it. That's where targeted door knocking comes in — and the entire game is the quality of the list.

The brutal math of random knocking

Untargeted door knocking is a payroll furnace. Walk a rep down a random street and most of the doors are new roofs, vacant, renters who can't decide anything, or people who replaced theirs two years ago. If only one roof in eight on a block is actually old enough to need you, your rep spends seven-eighths of their day, gas, and energy on doors that were never going to buy. That's the math that burns out new reps and convinces owners that knocking "doesn't work."

Knocking works fine. Knocking randomly doesn't. The difference between a rep who quits in three weeks and one who sticks is almost never grit — it's whether the doors they knocked had any chance of being a job.

What makes a door worth knocking

Two things, mostly: roof age and what the weather has done to that specific roof. A roof that's 20 years old is a candidate on age alone. A roof that's 12 years old but took a real hail event last spring is a candidate too. A 6-year-old roof on a street that's never seen a serious storm is a waste of a knock no matter how friendly your rep is.

The trouble is you can't see age from the curb, and the public data lies to you. The "year built" on a county record or a real-estate site is the year the house went up — not the year the roof was last replaced. A 1995 house may have a 2018 roof, and you'd never know it from Zillow. Re-roofs are invisible to those sources. That single confusion sends reps to thousands of wrong doors.

Where roof-age and storm data come from

This is the part where the work has genuinely changed in the last few years. You no longer have to guess.

• Aerial and satellite imagery can be read — increasingly by software — to estimate how worn a roof is, producing a roof-age range rather than a guess from house age. A range, not a date: nobody can read the exact install year off a photo, and any tool that claims a precise date is overselling. But "this roof reads as 18–22 years old" is enough to decide whether to knock.
• Storm history per address. Hail and high-wind events are tracked, and the path of a storm is knowable. What's harder — and more useful — is knowing what a storm did to a specific roof, rather than only which ZIP it crossed. A hail map shows you where it hailed. It doesn't tell you that the steep south-facing slope on 412 Oak took the brunt while the house next door, shielded by a tree line, barely felt it.

This is exactly the gap RoofPredict is built to close. It reads aerial imagery to estimate a roof-age range house by house, and it models the physics of hail and wind on each individual roof — rather than only the storm's footprint — to score which roofs a storm actually wore out. The output isn't a lead to buy; it's a ranked picture of your own streets and lists so your crew knocks and mails the roofs that are due and skips the new ones. You can also feed it your existing customer book or a mailing list and have it enriched with roof-age and storm signals, which is what makes Play 1 above so much sharper — suddenly those "roof age unknown" records have a range attached.

Honest limits, because a tight trade compares notes: it's a range, not an install date, and the storm score is odds that a roof was affected, not proof that it was. It tells you which doors are worth your time; your rep on the ladder still confirms condition. Used that way — to rank where to spend your knocking and mailing hours — it turns the brutal one-in-eight math into something a rep can actually live on, because most of the doors they hit now have a real reason to be there.

A targeted-knock workflow

1. Pick a target area — a few subdivisions you can name, ideally ones built in the same era so the roofs aged together.
2. Get a roof-age-and-storm ranking of those streets, so you know which blocks and which houses are due.
3. Route the rep to the top of the list — the oldest roofs and the storm-worn roofs first, not the nearest doors.
4. Open with proof, inspect honestly, document with photos — same discipline as the neighbor canvass.
5. Track outcomes by list source so you learn which signals (age vs. storm vs. both) convert best in your market, and feed that back into next week's routing.

The rep economics, worked out

Targeting isn't a nice-to-have; it changes whether door knocking is profitable at all. Run two reps side by side for a week to see it.

Rep A — random street routing. Knocks 200 doors. Roughly 1 in 8 roofs is actually old enough to matter, so about 25 doors had any chance. Of those, the rep catches maybe half home and willing to talk — 12 conversations — and books 3 inspections. The other 175 doors were payroll and gas spent on roofs that were never buying. The rep ends the week tired and discouraged, which is how reps quit.

Rep B — ranked old-roof routing. Knocks the same 200 doors, but they're routed oldest-and-most-storm-worn first, so maybe 5 in 8 are real candidates — 125 doors with a reason. Same catch-and-willing rate yields 60-plus conversations and 12 inspections. Same payroll, four times the booked inspections, and a rep who's making money and stays.

The doors didn't get easier. The list did. That's the entire argument for spending your effort on the routing before you spend it on the knocking.

Pairing the list with route efficiency

A good list still gets wasted if the rep crisscrosses town. Once you've ranked the due roofs, cluster them:

• Group by block, then by age within the block. Hit the oldest roof on a street first; its neighbors are usually close in age, so a single stop often opens several.
• Set a daily door floor and a quality floor. "Knock 40 doors" alone produces busywork; "knock 40 doors, all from the ranked list" produces inspections.
• Re-rank weekly. Roofs you inspected and found fine drop off; roofs that just took a storm move up. The list is a living thing, not a one-time pull.

Play 5: Turn storms into work — the right way, on the right side of the line

Storms create a surge of legitimately needed roof work. They also create the single biggest compliance trap in the trade. You can build a real, ethical storm pipeline — but only if you stay strictly on the documentation and estimate side of the line and never cross onto the homeowner's side of their insurance claim.

The line you must not cross

Here is the framing every owner and every rep needs to have memorized. As a roofer, you may:

• Inspect a roof and document damage thoroughly with dated photos.
• Write an accurate repair estimate for the work you would do — ideally aligned to the same line-item pricing language adjusters use, so it's apples-to-apples.
• State facts about your scope of work to the homeowner and, where appropriate, to the carrier about what your repair entails.
• Hand that documentation and estimate to the homeowner so they have a clear, professional picture of the condition and cost.

As a roofer, you may not — unless you are a licensed public adjuster:

• Negotiate, adjust, or "handle" the homeowner's claim for a fee.
• Interpret their policy or tell them what is and isn't covered.
• Promise a specific payout, approval, or that the claim will go through.
• Promise the deductible will be waived, absorbed, eaten, or "taken care of." Waiving or rebating a deductible is illegal in many states and is insurance fraud.
• Advertise a "free roof."
• Represent the homeowner against their insurer. That's unlicensed public adjusting, and it's prosecuted.

The safe frame is simple: you document, you estimate, you hand it over. The homeowner files. The insurer decides coverage. You are the expert on the roof, never on the policy.

The do-not-say list — train your reps on these

Post this where every canvasser sees it. These phrases get companies sued, fined, and shut down:

• "We'll get your roof approved."
• "We'll handle the whole claim for you."
• "You won't pay anything — we'll cover your deductible."
• "This will definitely be covered."
• "Free roof."
• "We'll fight the insurance company for you."

What to say instead:

• "I'll document the damage thoroughly and write you a detailed estimate so you have a clear, professional picture of the condition."
• "You'd file the claim and your adjuster makes the coverage decision — I'll make sure you're walking in with solid documentation."
• "Your deductible is set by your policy; I can't change that, but I can make sure the scope is accurate so there are no surprises."

A clean storm-documentation workflow

1. Confirm a real event. Verify that a hail or wind event actually hit the address before you knock. Modeling what the storm did to that roof — rather than only the ZIP — is exactly where per-roof storm data earns its keep, so you're knocking roofs that plausibly took damage rather than spraying a whole region.
2. Inspect and photograph systematically. Date-stamp everything. Capture the slopes, the soft metals (gutters, vents, flashing), and any collateral (screens, AC fins) that corroborates the hail. Document the whole roof, not only the worst spot.
3. Write the estimate in adjuster-aligned line items. Tear-off, underlayment, drip edge, starter, field shingle, ridge, flashing, ventilation, disposal — itemized at fair market pricing. This is the homeowner's tool, not your claim argument.
4. Hand the homeowner a clean package. Photos, the estimate, and a plain-language note that they file the claim and the adjuster decides coverage. That's it.
5. Stay in your lane on every follow-up. If they ask whether it'll be covered, the honest answer is "that's the adjuster's call — my job is to make sure the damage and the scope are documented accurately."

Done this way, storm work is one of the strongest free channels you have: real need, visible damage, and a community of neighbors who all took the same hail. Done the wrong way, it's a fast track to a state Department of Insurance complaint.

Play 6: Make the internet find you for free

Organic discovery compounds. It's slower than a phone call but it's the closest thing to advertising that you don't keep paying for. The goal isn't to become a marketing company — it's to claim the free real estate that's sitting there.

Your Google Business Profile is the highest-leverage free asset

For a local roofer, the map pack and your Business Profile drive more free phone calls than anything else online. It's free and most roofers half-fill it out.

• Complete every field: services, service area, hours, and a real description.
• Get reviews systematically. Build the review ask into the same final-walkthrough moment you use for referrals. A simple "would you mind leaving us a quick review? Here's a link" texted from the field, at the moment of peak happiness, is how you get reviews. Respond to every one, good or bad, like a professional.
• Post job photos regularly. Before-and-after shots of real roofs in named neighborhoods tell Google — and homeowners — exactly where and what you do.

A handful of pages that answer real questions

You don't need a blog mill. You need a small number of genuinely useful pages that answer what local homeowners actually type: how to tell if a roof needs replacing, what storm damage looks like, what a re-roof costs in your area, how the insurance-claim process works (factually, on the homeowner's side — you document, they file, the adjuster decides). Write them once, write them well, and they earn calls for years.

Reviews and reputation as a lead source

Reviews are both ranking fuel and the thing a homeowner reads before they call. Treat your reputation as a channel: every clean job is a review opportunity, every review is social proof, and the compounding of both pulls in calls you never paid for.

Play 7: Partner with the people who already meet your customers

Some of the steadiest free work comes from people who are in front of homeowners with roof problems all day and aren't roofers.

• Real-estate agents need roof inspections and quick repairs to close deals, on tight timelines. Become the reliable roofer an agent can call on a Friday afternoon and you'll get a steady drip of pre-sale inspections and repairs — and pre-sale inspections turn into full replacements.
• Property managers have portfolios of roofs aging on a schedule. One relationship can be dozens of roofs.
• Insurance agents (not adjusters working a specific claim — agents) field calls from policyholders with damage and need someone competent to refer for documentation. Keep this relationship strictly on the documentation-and-repair side.
• Adjacent trades — gutter installers, solar companies, general contractors, painters — see roofs constantly and trade referrals naturally.

The move is the same in every case: be reliable, be fast, make them look good to their client, and reciprocate. A single property manager or a couple of active agents can keep a small crew busy with zero ad spend.

How to actually start a partnership

These relationships don't appear because you want them. They start because you made yourself useful before you asked for anything.

• Lead with reliability, not a pitch. Tell an agent: "Next time you've got a listing that needs a roof looked at on a tight timeline, call me — I'll be out within a day and you'll have a clear answer fast." You're solving their problem, which is closing the sale, not selling them your service.
• Make the documentation clean and fast. Agents and property managers live and die by deadlines. A roofer who turns around a clear inspection report the same day becomes the only number they call.
• Reciprocate visibly. Refer the good agents back. Send the painter the leads you can't use. Reciprocity is the engine of every steady referral relationship.
• Pick a handful, go deep. Two engaged agents and one property manager will out-produce a stack of business cards handed out at a networking event. Depth beats breadth here.

Measure cost-per-job before you ever buy an ad

The single discipline that separates roofers who scale profitably from roofers who burn cash is knowing what a booked job costs them, by channel. You can't make a smart decision about advertising until you know the number it has to beat.

Track these four numbers per channel

1. Activity: calls made, doors knocked, asks issued. No input, no output.
2. Inspections booked: the first real conversion, and the truest early signal that a channel is alive.
3. Jobs signed: the only number that pays the bills.
4. Cost to produce the job: for free channels, that's payroll and time; for paid, it's media spend plus payroll. Divide total cost by jobs signed and you have cost-per-job.

When you compute cost-per-job for your free channels — the book, referrals, neighbor canvassing — you'll usually find it's a fraction of what a bought lead costs. That number is your benchmark. If a paid channel can't beat it, you don't run the paid channel; you pour more into the free one.

A simple attribution habit

You don't need software for this. You need one question, asked every time: "How did this homeowner come to us?" Log the answer in one column of your job records — referral, neighbor canvass, old customer, ranked knock, Google, partner. Within a quarter the pattern is obvious, and you'll know exactly which play to feed and which to fix. Most roofers can't answer this question at all, which is why they keep buying the most expensive channel by default.

When advertising finally makes sense

Paid advertising isn't the enemy — running it blind is. Once your free channels are genuinely tapped, you know your cost-per-job, and you have the crew capacity to handle more volume, ads can be the right next lever to add scale. Go in with the benchmark number, measure paid against it with the same attribution column, and kill anything that can't beat what you already generate for free. That's the difference between advertising as a growth tool and advertising as a money leak.

Putting it together: a 90-day plan with no ad budget

Plays in isolation produce activity. Sequenced, they produce a schedule. Here's how a company with no advertising budget fills the next quarter.

Days 1–30: Mine what you own

• Export and tag the entire customer book into the three buckets (replacements, repairs, dead estimates).
• Get the "roof age unknown" records — and your target neighborhoods — enriched with a roof-age range and storm history, so the list stops being guesswork.
• Call the top 50 (every dead estimate under 24 months old, plus the oldest roofs). Book inspections.
• Stand up the referral program: a clear reward, a card to hand out, and the final-walkthrough ask scripted for every crew lead.

Days 31–60: Work the streets

• Run the neighbor canvass on every active job site — 20–30 doors per tear-off, honest inspections, photos.
• Start targeted knocking in one or two ranked neighborhoods, routing reps to the oldest and most storm-worn roofs first.
• Fully build out the Google Business Profile and start the systematic review ask on every completed job.

Days 61–90: Compound and partner

• Re-work the long tail of the customer list by email/postcard.
• Open two or three local partnerships — an agent, a property manager, an adjacent trade.
• Review your tiny source ledger: which play booked the most work per hour? Double down there next quarter.
• Only now, with the free channels genuinely running, decide whether paid advertising is worth layering on — and if so, you'll do it knowing your true cost to book a job, which is the only way to spend on ads without lighting money on fire.

A simple weekly scorecard

Track five numbers and you'll know whether the engine is running:

What pros get wrong

A few mistakes show up again and again, even in established companies:

• They buy ads before working the book. The customer list is full of money and it sits untouched while the owner pays for clicks. Always work what you own first.
• They knock randomly. Without a roof-age-and-storm-aware list, reps burn the day on new roofs and quit. Targeting is the difference between a rep who stays and one who's gone in a month.
• They trust "year built" as roof age. House age is not roof age. Re-roofs are invisible to county records and real-estate sites. Decisions made on that number are wrong a large share of the time.
• They ask for referrals weakly, late, or never. "Let me know if you hear of anyone" produces nothing. Ask at peak happiness, be specific, make it easy, reward it fast.
• They cross the insurance line. Promising approvals, eating deductibles, or advertising a "free roof" feels like a shortcut and is a fast way to a fine or a shutdown. Document, estimate, hand it over — the homeowner files and the insurer decides.
• They don't measure source. If you can't say which play booked which job, you can't double down on what works or know your real cost per job before you start buying ads.

The bottom line

The cheapest jobs you'll ever book are the ones already sitting in your records, on your streets, and in your customers' goodwill. Work those to exhaustion — the dead estimates, the aging past customers, the neighbors of every tear-off, the referral asks at peak happiness, the targeted doors where the roof is genuinely due — and most companies can fill a schedule without spending a dollar on advertising.

The reason to do it in this order isn't just cost. It's ownership. A job you generate from your own book or your own street is a job no aggregator resold to four competitors and no storm dictated the timing of. You control it, it repeats, and it compounds. Tools like RoofPredict make the old-roof and storm-worn doors visible so your free channels hit the right houses instead of every house — but the engine is the discipline. Run the plays in order, measure the source, and only reach for the advertising budget once you actually know what a booked job costs you. That's how you grow on jobs you own instead of leads you rent.

FAQ

How can I find roofing jobs without paying for leads or ads?

Start with the assets you already own. Work your customer book first — call past customers, re-inspect old repair jobs, and re-engage dead estimates from the last two years. Run a referral program with a real reward asked at the moment a job finishes. Canvass the neighbors of every job site while your crew is on the street. Target door knocking to roofs that are genuinely old or storm-worn. Build out your free Google Business Profile and get reviews. Only after those are running should you consider paid advertising.

Why are bought roofing leads so expensive to close?

Shared or aggregated leads are sold to several contractors at once, so you're in a price race with strangers who also have no relationship with the homeowner. You're paying cash for a contact where you have the least leverage. A past customer or a referral closes at a far higher rate with less price pressure because the trust already exists, and it costs you a phone call instead of a per-lead fee.

How do I find out how old a roof is before I knock?

Don't rely on the 'year built' from county records or real-estate sites — that's the age of the house, not the roof, and re-roofs are invisible to those sources. Modern tools read aerial imagery to estimate a roof-age range house by house. It's a range, not an exact install date, but a reading like '18 to 22 years old' is enough to decide whether a door is worth knocking. Pairing that age range with storm history per address gives you the strongest possible knocking list.

What is the most overlooked source of roofing jobs?

Dead estimates. Every quote you wrote that didn't close is a homeowner who was actively shopping for a roof and stalled on budget, timing, or an insurance question — not because they stopped needing a roof. A meaningful share are still live a year or two later, and you already did the measuring and the relationship work. Calling your stalled estimates from the last 24 months is often the single highest-return hour an owner can spend.

How do I get more roofing referrals?

Engineer them instead of waiting. Ask at three moments: the final walkthrough when the homeowner is happiest, the one-week follow-up, and a seasonal check-in. Make the ask specific ('your neighbor on Maple is about the same age — mind if I drop a card by?'), offer a reward worth caring about, and pay it fast when the referred job closes. Give the customer something physical to hand over, like cards or a yard sign. A weak, late, generic ask produces nothing.

Is door knocking for roofing jobs still worth it?

Yes, but only if it's targeted. Random knocking is a payroll furnace because most doors are new roofs, vacant, or renters. If only one roof in eight on a block is old enough to need you, your rep wastes most of the day. Knock with a list ranked by roof age and storm exposure so reps hit doors that have a real reason to convert. That targeting is usually the difference between a rep who sticks and one who quits in three weeks.

Can a roofer help with a homeowner's insurance claim?

Only on the documentation side. You may inspect, photograph damage thoroughly, write an accurate repair estimate in adjuster-aligned line items, and hand that package to the homeowner. You may not negotiate or handle the claim, interpret their policy, promise it'll be covered or approved, waive or absorb their deductible, advertise a 'free roof,' or represent the homeowner against their insurer — all of that is unlicensed public adjusting and, in the case of deductibles, often fraud. The homeowner files; the adjuster decides coverage.

What phrases should roofing sales reps never say about insurance?

Never say 'we'll get your roof approved,' 'we'll handle the whole claim,' 'we'll cover your deductible,' 'this will definitely be covered,' 'free roof,' or 'we'll fight the insurance company for you.' These get companies fined and shut down. Instead, say you'll document the damage and write a detailed estimate so the homeowner has a clear picture, that the adjuster makes the coverage decision, and that the deductible is set by their policy and can't be changed.

How does RoofPredict help find jobs without advertising?

It makes your free channels hit the right houses. RoofPredict reads aerial imagery to estimate a roof-age range house by house and models hail and wind on each individual roof — rather than only the storm's ZIP-level footprint — to show which roofs a storm actually wore out. It's not a lead service; it ranks your own streets and lists so you knock and mail roofs that are due and skip the new ones. You can also have it enrich your existing customer book or mailing list with roof-age and storm signals. Honest limit: it gives a range and odds, so your rep on the ladder still confirms condition.

How long before these free methods fill my schedule?

Working your customer book and dead estimates can book inspections within days because those people already trust you. Referrals and neighbor canvassing build over the first 30 to 60 days as you run them on every job. Organic search and reviews compound more slowly over months. A reasonable 90-day plan is to mine your book and stand up referrals in month one, work the streets and your Google profile in month two, and add local partnerships in month three — measuring cost per booked job by source the whole way.