How to Get Roofing Work Without a Referral Network

Referrals are the cheapest work you'll ever get. A past customer hands you a neighbor, you show up, you close at the kitchen table, and the cost of acquisition is roughly a thank-you card. Every seasoned roofer knows this, which is exactly why the advice "just get more referrals" is so useless to the contractor who doesn't have them yet.

Maybe you're two years in and your past-customer list is thin. Maybe you moved markets. Maybe you ran storm work in another state and now you're trying to build something local and repeatable. Maybe the referrals you do get are lumpy and unpredictable, and you need to fill the gaps between them. Whatever the reason, you need to manufacture demand instead of waiting for it to walk in.

The good news: roofing is one of the few trades where you can look at a street and have a reasonable idea of who needs you. Roofs wear out on a schedule. Storms wear them out faster. The work is sitting on rooftops you can drive past today. What follows is the operator's version of how to turn that into a booked calendar, with the numbers, scripts, and workflows that actually move it, plus the mistakes that quietly drain a small roofing company's cash.

Why referrals dry up (or never start) and what that actually means

Before building the alternative, it's worth being honest about the math, because it explains why so many new roofing companies stall at the same revenue ceiling.

A referral has three things going for it that no cold channel does: trust is pre-loaded, the timing is usually right (something failed or the neighbor just got a roof), and the cost is near zero. When you don't have referrals, you have to recreate all three with money, time, and process. You can do it. But you can't pretend a cold door is the same as a warm one, or you'll burn out your reps and your bank account.

Three situations create the no-referral problem:

• You're new. You simply haven't done enough roofs for the flywheel to spin. A roofing company that's done 40 jobs has maybe 4-8 reliable referral sources; one that's done 400 has a self-sustaining network. You're early.
• Your past work is invisible. You did the jobs but never collected reviews, never asked, never stayed in touch. The referrals exist in theory and never get spoken out loud.
• You changed the game. New market, new crew, new service line (you used to do repairs, now you want full replacements). The old network doesn't transfer cleanly.

In all three cases the fix is the same shape: you need a way to consistently put your truck in front of homeowners who have a roof problem, whether they know it yet or not. The rest of this breaks down the channels that do that, ranked by how fast they pay and how much they cost to run.

The one idea that changes everything: target old roofs, not whole streets

Here's the mistake that defines the difference between a roofing company that grows and one that grinds: working the whole street.

New contractors think more doors equals more jobs, so they knock everything, mail everything, drive everything. The problem is that most of a street doesn't need you. A roof that was replaced four years ago is not a customer this year no matter how good your pitch is. Every door you knock on a 6-year-old roof is gas, payroll, and rep morale spent on a guaranteed no.

The entire game is density of demand. If 1 in 20 homes on a street has an aging or storm-worn roof and you knock all 20, your effective rate is brutal. If you can knock only those qualified homes, the same hour of labor produces several times the conversations that matter. You haven't gotten better at sales; you've stopped wasting at-bats on people who structurally cannot buy.

So the foundational skill in a no-referral world is qualifying a roof before you spend money on the homeowner. There are three signals that matter:

1. Roof age. Asphalt shingle roofs run roughly 15-25 years depending on shingle grade, ventilation, pitch, and climate. A roof entering its late teens is a real prospect. A roof under 8 years old almost never is.
2. Storm exposure. Hail and high wind compress that timeline. A roof that took a serious hail event can be functionally worn out years before its age would suggest.
3. Visible condition. Granule loss, curling, missing tabs, sagging, moss, prior patch jobs. The stuff you can read from the curb or a ladder.

When all three line up, you have a roof that is due. That's the home worth a knock, a mailer, and a follow-up. Everything that follows is about finding those homes cheaply and then converting them.

A quick reality check on "year built"

A lot of new roofers try to shortcut this by pulling "year built" off a real-estate site and assuming an old house means an old roof. It doesn't. A 1968 house may have been re-roofed three times; the last replacement is what matters, and that re-roof is invisible to property records. Year built tells you the house is old. It tells you almost nothing about the roof. Treat it as a weak hint, never a target list.

The same caution applies to a few other shortcuts roofers reach for:

• Measurement tools tell you size, not age. Aerial measurement reports (the ones that give you squares, pitch, and ridge length for a bid) are excellent for what they do, but they answer "how big is this roof," not "which roof needs replacing." Different question, different category. You still have to qualify the roof's age and condition separately.
• Permit pulls are useful but partial. A re-roof permit, where it was filed, is a hard signal that a roof is newer. But plenty of re-roofs were done without a permit, so the absence of a permit doesn't prove the roof is old. Use permits to exclude recently re-roofed homes, not to build your list.
• Drive-by guessing scales to about three streets. You can read curling, granule loss, and patchwork from the truck, and a sharp estimator gets pretty good at it. But you cannot drive-by-qualify a whole service area in a season, and the south-facing slope (the one that ages fastest) is often the one you can't see from the road.

The practical takeaway: build your target list from the strongest age-and-condition signals you can get, exclude the homes you can prove are newly roofed, and accept that you're working in ranges and probabilities, not certainties. Nobody can tell you the exact day a roof was installed by looking at it. The goal isn't certainty; it's tilting the odds so most of your at-bats are against roofs that can actually buy.

How long roofs actually last (so you target the right age band)

Knowing the real service-life ranges keeps you from targeting too young (no demand) or assuming too old (the roof already got replaced). Rough working bands for common residential materials in a typical climate:

Climate, ventilation, sun exposure, and installation quality swing every one of these. A poorly ventilated attic can take five years off an asphalt roof; a brutal south-facing slope in a hot, sunny region ages faster than the north slope on the same house. Storms compress all of it. The point of the table isn't precision, it's to stop you wasting outreach on a 6-year-old architectural roof that has two decades left, and to remind you that an asphalt roof in its late teens is the meat of your market.

Channel 1: Your own list (the referrals you forgot you had)

Before you spend a dollar on a stranger, work the warmest list you've got, which is almost always bigger than you think.

Most roofing companies are sitting on money in their own files: old estimates that never closed, past repair customers who are candidates for a full replacement, and homeowners you bid against three years ago whose roofs are now three years closer to failing. This is the closest thing to a referral channel you can build without referrals, because there's already a relationship, however thin.

Here's the workflow:

Step 1: Pull every estimate from the last 5 years

Export your CRM, your QuickBooks, your shoebox of carbon copies, whatever you have. You're looking for three buckets:

• Unclosed estimates (lost bids). You inspected the roof, quoted it, and they went elsewhere or did nothing. A meaningful share of "did nothing" homeowners still have that aging roof.
• Repair-only customers. You did a patch, a few shingles, a flashing fix. Many of these roofs were already near end-of-life when you patched them. They're now full-replacement candidates.
• Completed replacements over 8-10 years old. Not for re-roofing, but for referrals, gutters, maintenance, and reviews.

Step 2: Re-qualify each one by current roof age

This is the step nobody does, and it's the difference between a smart re-engagement campaign and spamming your old contacts. An estimate from four years ago on a roof that was 14 years old then is a 18-year-old roof now. That's prime. An estimate on a roof you replaced two years ago is not.

Layer in storm history: if a hail or wind event hit those addresses since you last talked to them, that's your reason to reach out, and it's a legitimate, factual one.

Step 3: Reach out with a reason, not a "checking in"

Nobody responds to "just following up." They respond to specifics:

"Hi Janet, it's Mike from [company]. We bid your roof back in 2022, you weren't ready then. I was looking at the area and your roof's now getting into the range where shingles start giving up. We took a couple of storms through here since. Happy to come re-check it, no charge, so you know exactly where it stands before something starts leaking."

That message converts because it's true, it's specific, and it puts you back in front of a roof that's genuinely closer to needing work.

A worked example: turning 300 dead estimates into booked work

Walk the math, because this is where small roofers leave the most money on the table. Say you have 300 estimates and old-customer records over five years:

1. Re-qualify by current roof age. Maybe 80 of those roofs are now in the late-teens-and-up band, plus another 25 repair-only customers whose roofs were already marginal. Call it ~100 worth re-touching.
2. Layer storm history. Of those 100, suppose 35 sit in areas that took a notable hail or wind event since you last talked. Those move to the top of the call list, because you have a fresh, factual reason to reach out.
3. Run the sequence. A disciplined multi-touch sequence (text, call, voicemail, a second text a few days later) to 100 warm contacts might book 12-15 inspections. These people already know your name, so the appointment rate runs several times what cold outreach produces.
4. Convert. At a normal inspection-to-sale rate for a warm contact, you're looking at a handful of signed jobs from records that were sitting dead in a file.

The acquisition cost was your time and a few hours of texting. Nothing else in the playbook pays this fast, which is why it's always step one.

Build the follow-up sequence, don't wing it

One call is not a campaign. Most of these book on the second or third touch, when something on the roof finally bugs the homeowner or they have a free Saturday. A simple sequence that works:

• Day 1: Personal text referencing the prior estimate and the specific reason (age or storm).
• Day 2: Call. If no answer, leave a short, specific voicemail, no "just checking in."
• Day 5: Second text with the offer (free check-up, photos, no obligation).
• Day 12: Final touch, often a "we'll be working in your neighborhood next week, want me to swing by?"

Stop after four touches on a non-responder; move them to a quarterly re-qualify list. Their roof is only getting older, and in two seasons they may be ready.

Channel 2: Door knocking, done like a professional, not a panhandler

Door knocking has a bad reputation because most people do it badly, knocking randomly, leading with a discount, and sounding desperate. Done right, it's still one of the fastest ways for a roofer with no network to put work on the board, because you control the volume entirely. More doors, more conversations, more jobs, today.

The key is everything we covered above: knock qualified roofs, not whole streets. A canvasser sending two hours on 40 carefully chosen doors will outproduce one knocking 120 random doors, and won't burn out by Thursday.

The professional door approach

The opener has one job: don't get the door closed. You're not closing a roof on the porch; you're earning a look at the roof.

"Hey, I'm not selling anything door to door. I'm [name] with [company], we're a roofing crew working a few houses in the neighborhood. I noticed your roof's got some age on it from the street, the granule wear on the south face especially. I'm offering free roof check-ups while we're here. Takes ten minutes, I'll take some photos up top and show you exactly what I see, good or bad. No obligation. Worst case you find out your roof's fine and you sleep better."

Why this works:

• "I'm not selling anything door to door" disarms the reflex. (You are, of course, but you're leading with value.)
• A specific observation ("granule wear on the south face") proves you actually looked, which separates you from the script-readers.
• "Good or bad" signals honesty. You're not there to manufacture a problem.
• The ladder is the close. Once you're on the roof with a phone full of photos, you're in a fundamentally different conversation than a stranger on a porch.

What to do once you're up there

Document everything, methodically, every time. Photograph the field shingles, the valleys, the flashings, the penetrations, the ridge, the gutters (granule accumulation is a tell), and anything that looks like prior damage. You're building a visual case the homeowner can see for themselves. Show them on your phone right there. "Here's your roof. See this? That's where the granules have worn off and the mat's exposed. This is what we're worried about."

This photo-first approach does two things: it makes the sale because they can see the problem, and it builds the documentation file you'll need later if the homeowner decides to involve their insurer (more on the compliance line in the storm section).

A few discipline points that separate pros from porch-talkers on the roof:

• Shoot wide, then tight. A wide shot establishes which slope, a tight shot proves the defect. A folder of close-ups with no context is useless six weeks later.
• Get the date and address baked in. Use a camera app that stamps date and location, or photograph the house number and street first. Undated photos are worth far less if a claim is ever involved.
• Document the whole roof, not only the bad spot. The condition of the field, valleys, flashings, pipe boots, and drip edge tells the real story. A homeowner who sees you photographing everything trusts the one bad photo more.
• Don't manufacture damage. Ever. Stepping wrong to crack a tile, lifting a tab to make it look storm-torn, that's fraud, it gets crews blacklisted, and it's exactly the behavior that gives door-knocking roofers a bad name. Your edge is honesty, not theater.

Working with the gatekeeper and the not-home majority

Most doors won't answer, and many that do are a spouse, a renter, or someone screening for the homeowner. Two adjustments:

• Leave a real door hanger on no-answers with a specific note: "Stopped by, your south slope has visible granule wear, happy to do a free check-up, call [name]." A note that proves you looked beats a generic flyer.
• For renters, ask for the owner, don't pitch the tenant. "This is a rental? No problem, do you know if the owner handles the property themselves or through a company? I'll leave info for them." You're not going to sell a renter a roof, but you can get a path to the decision-maker.

A clean rebuttal library

Green reps freeze on objections. Give them three or four honest, non-pushy responses to memorize:

• "I'm not interested." — "Totally fair, most folks aren't thinking about the roof until it leaks. That's actually the point of the free check-up, so you find out before it does. Ten minutes, I'll show you photos, and you decide."
• "My roof's fine." — "Might be, and if it is I'll tell you that and you're done. I just noticed some wear on the [slope] from the street. Worth a look so you know for sure."
• "I already have a roofer." — "Good, everybody should. I'm just offering the free check-up while we're in the area. If anything turns up, you've got photos to take to whoever you trust."
• "How much is this going to cost me?" — "The check-up's free, no obligation. If you've got real wear, I'll write you a clear estimate so you know the number. What you do with it is your call."

Notice none of these argue, pressure, or promise. They lower the stakes and keep the goal narrow: earn the look at the roof.

Canvassing math and the churn problem

Let's get concrete. A disciplined canvasser working qualified doors might hit these rates:

Those are workable numbers for a focused rep on good doors. On random doors, every line collapses, and that's where the real cost shows up: rep churn. A green canvasser who knocks bad doors gets demoralized, makes no money, and quits in six weeks. You eat the recruiting and training cost and start over. A canvasser who knocks roofs that are actually due gets conversations, closes, makes money, and stays. The single biggest lever on your canvassing cost isn't the script. It's putting reps on doors that can convert.

Channel 3: Direct mail that gets opened

Direct mail still works for roofing, but most roofing mail is wasted because it goes to whole ZIP codes, including thousands of new roofs that will never call. You pay to print and mail a piece to a roof you replaced last year. The waste is the problem, not the channel.

The fix is the same idea again: mail the old roofs, skip the new ones.

What makes roofing mail convert

• Targeting beats creative. A mediocre postcard to the right 500 homes beats a beautiful one to the wrong 5,000. Get the list right first.
• Specificity beats hype. "Roofs in [neighborhood] built around 2005-2010 are reaching the age where shingles start to fail. We're offering free roof check-ups this month." That's a reason, tied to their actual situation.
• A real offer, not a fake deadline. A genuine free inspection is a real offer. "ACT NOW, 50% OFF" on a roof is a discount nobody believes. Don't manufacture urgency; the roof provides its own.
• Repetition. One mailer is a lottery ticket. The same targeted list hit 3-4 times over a season is a campaign. People call on the third piece, when something starts to bug them, not the first.

The mail math

Roofing direct mail response rates typically land in a low single-digit-percent range to a good list, and a fraction of that to a bad one. The economics live or die on list quality:

Same creative, same offer, the only variable is who you sent it to. That's why targeting is the whole game.

Pair mail with the knock

Mail and door knocking aren't competitors; they're a one-two punch. Mail the targeted list, then knock the same list a week later. Now the door isn't cold: "You may have seen our postcard, we're the crew doing the free roof check-ups in the neighborhood." Familiarity does a lot of work on a porch.

Format choices that actually matter

Don't overthink the creative, but a few format calls move response:

• Postcard over letter for the first touch. A postcard gets seen at the mailbox with zero effort to open. Save the letter format for a follow-up to people who responded or for higher-value targets.
• Real local photos beat stock. A picture of an actual aging roof in their kind of neighborhood reads as "this company works here," which stock houses never do.
• One offer, one call to action, one phone number. A postcard trying to say five things says nothing. Free roof check-up. Call this number. Done.
• Put a face and a local address on it. Homeowners are wary of out-of-area storm-chasers. A local business name, a real address, and an owner's name signal you'll still be around when the roof needs service.

Track it or you're guessing

Use a dedicated phone number or a simple "mention this card" line so you can tell which neighborhoods and which mail drop produced calls. Without tracking you'll keep funding the ZIP that doesn't convert and starve the one that does. After two or three rounds you'll know your best neighborhoods cold, and you can pour budget there.

Envelope it all in the same rule that governs every channel here: the creative is a minor variable, the list is the whole game. Mail old roofs, skip new ones, hit them more than once, and track what comes back.

Channel 4: Digital presence and inbound, the slow compounding play

Everything above is outbound: you go find the work. Inbound is the homeowner finding you, and while it's slower to start, it compounds and eventually lowers your cost per job. You can't live on it for the first 6-12 months, but you must be building it the whole time.

The non-negotiables

• Google Business Profile. This is the single highest-leverage free thing a local roofer can do. Claim it, fill it completely, add real job photos, and get reviews. When someone searches "roofer near me," the map pack is where the clicks go, and reviews drive ranking.
• Reviews, systematically. Ask every happy customer, every time, with a direct link, while you're standing in their driveway after a good job. A roofer with 40 genuine reviews looks like a real business; one with 3 looks like a risk. Reviews are also the closest digital cousin to a referral, social proof from strangers.
• A simple, fast website with your service area, real photos, and a phone number above the fold. It doesn't need to be fancy. It needs to load fast and make calling easy.
• Show the work. Before/after photos, drone shots, a short video of a crew tearing off and laying a clean new roof. Roofing is visual and most homeowners have never seen the process. Showing it builds trust before they ever call.

A review-collection system that actually runs

Most roofers "mean to" ask for reviews and never do. Make it mechanical so it happens on every job without you remembering:

1. Pick the moment. The walk-through at job completion, when the homeowner is happiest and standing right there, is the moment. Not a week later by email.
2. Ask in person, then send the link. "If you're happy with how this turned out, the single biggest help to a local company like ours is a quick Google review. Mind if I text you the link right now?" Then text it before you leave the driveway.
3. Make it one tap. The text is a direct link straight to the review form. Every extra step loses people.
4. Respond to every review, good or bad. It signals a real, attentive business and helps your ranking.

A roofer who does this on every job goes from 3 reviews to 40 in a season, and 40 genuine reviews is the digital version of a referral, strangers vouching for you before you ever speak.

Other inbound worth a little effort

• Yard signs and wrapped trucks. Cheap, local, and they put your name on the exact streets where you're already working. A neighbor watching a crew tear off a roof is a warm prospect.
• Neighborhood apps and local groups. When a homeowner posts "anyone know a good roofer," being the name that gets mentioned (because you did good work and asked) is a referral by another route.
• A handful of local pages on your site for the towns you serve, with real photos and real specifics, help you show up when people search your area. Keep them genuine; thin, copy-pasted town pages do more harm than good.

The honest timeline

Inbound from search and reviews is a 6-18 month build, not a this-month fix. That's exactly why you run outbound (channels 1-3) to pay the bills now while the inbound engine spins up. The contractors who win long-term do both: outbound for cash flow today, inbound for cheaper jobs tomorrow. Treat the inbound work as planting trees, you won't sit in the shade this quarter, but the roofer who started a year ago is already booking jobs from it.

Channel 5: Storm work, the right way (and the legal lines you cannot cross)

When a real hail or wind event moves through your area, a chunk of roofs that were merely aging just got pushed toward the end of their life early. That's legitimate work, and a roofer is the right person to document it. But storm and insurance work is also where contractors get into serious trouble, so this section is as much about staying compliant as it is about getting the job.

What you can do

A roofing contractor absolutely can:

• Inspect a roof and document damage thoroughly with dated photos and notes.
• Write an accurate repair estimate for the work, ideally aligned with the line-item pricing standards (Xactimate-style) that carriers recognize.
• State facts about your own scope of work, what it will take to repair the roof properly and what it costs.
• Hand that documentation and estimate to the homeowner so they can decide what to do with it.

That's a real, valuable service. After a storm, a homeowner often has no idea whether their roof took damage. You climb up, you document it honestly, you produce a clear estimate, and you give them the facts. That's how you earn storm jobs ethically.

What you cannot do (the do-not-say list)

Here's where roofers get fined, sued, or shut down. In most states, the following crosses into unlicensed public adjusting, which is illegal for a contractor. You may not, for a fee or as part of the deal:

• Negotiate, adjust, or "handle" the insurance claim on the homeowner's behalf. Filing and negotiating coverage is the homeowner's and the licensed adjuster's lane, not yours.
• Interpret the policy or tell the homeowner what's covered. You're not licensed to read coverage. Don't.
• Promise a specific payout, approval, or that the claim "will go through." You don't control the carrier's decision and can't guarantee it.
• Promise to waive, absorb, eat, or "make disappear" the deductible. In most states this is insurance fraud, full stop. The deductible is the homeowner's legal obligation.
• Advertise a "free roof" or "you pay nothing." Same problem, it implies the deductible gets erased.
• Represent the homeowner against their insurer. That is the definition of public adjusting, and it requires a license you don't have.

The safe frame is simple and worth memorizing: You document, you estimate, you hand it to the homeowner. The homeowner files. The insurer decides coverage. Stay on the documentation-and-estimate side of that line and storm work is a clean, repeatable revenue source. Step over it and you're risking your license and your business.

How to talk about it on the door

"We had a real hail event come through here on [date]. A lot of these roofs took impacts you can't see from the ground. I'll get up there, document exactly what I find with photos, and write you a clear estimate of what it'd take to repair it. That's yours to keep. If you want to take it to your insurance company, that's your call, you file, they decide. I'm just here to give you the facts about your roof."

Notice what's missing: no promise of approval, no deductible talk, no "free roof," no offer to fight the insurance company. You're giving facts and documentation. That's the whole job, and it's plenty.

What a clean storm-documentation packet looks like

When you hand the homeowner their packet, it should stand on its own as a factual record they can take wherever they choose. A solid packet includes:

• Cover page with the address, your company info, and the inspection date.
• A dated note of the storm event you're documenting ("hail event reported in this area on [date]"), sourced from public weather records, not your guess.
• Slope-by-slope photos of the field, with wide-then-tight shots showing impact marks, granule displacement, bruising, or wind-lifted/torn shingles.
• Photos of collateral indicators that an adjuster typically also looks at: soft metals (gutters, downspouts, vents, flashing), screens, and any obvious mechanical damage. Document what's there; don't characterize it as "covered."
• A line-item repair estimate for the scope to properly repair the roof, organized the way carriers expect (Xactimate-aligned line items, quantities, and pricing).
• A plain statement of roles, in writing: this is your documentation and estimate; the homeowner decides whether to file; the insurer determines coverage.

That last line isn't just polite, it's your protection. Putting the roles in writing keeps you visibly on the right side of the public-adjusting line.

Why storm work alone is a trap

Storm work is real and worth doing, but building a whole company on it is a feast-and-famine bet. The season is short, the events are unpredictable, and the moment a real storm hits, out-of-town crews swarm in, hammer the neighborhoods, and compete every door, often the ones cutting the legal corners you won't. When the storm season ends, those crews leave and the local roofer who has nothing but storm muscle goes quiet.

The roofer who also works aging roofs and their own list has a baseline of work storm or not. Storms become an accelerant on top of a steady business, not the whole business. That's the difference between a company that survives a quiet hail year and one that doesn't.

Closing the roofs you generated: don't waste the at-bat

Generating the conversation is half the battle; a roofer with no referral network can't afford to fumble the close once they've earned it. A few principles that turn inspections into signed jobs without slimy pressure:

• Sell the problem you can prove, not fear you invent. Show the photos. "Here's where the mat's exposed; once water gets under here, you're looking at decking." Let the roof make the case.
• Give a real range on the spot when you can. Homeowners hate the "I'll email you a quote next week" black hole. If you can ballpark or write the estimate at the table, you close more, because the moment of motivation is now, not next Thursday.
• Present options, not an ultimatum. A good-better-best on materials and scope lets the homeowner choose up instead of choose whether. It also makes the middle option feel reasonable.
• Be straight about what they don't need. If the roof has five good years left, say so. The homeowner who hears "you're fine for now" from you remembers you when they're not, and tells their neighbor. Honesty is a slower close and a faster reputation.
• Make the next step concrete. "I can have a crew here [date]; we'll need a deposit to lock materials." Vague endings die in the follow-up pile.

For a company building from zero, your close rate on self-generated inspections is one of the few numbers fully in your control. Track it, and if it's low, the fix is usually clarity and proof, not more pressure.

Putting it together: a 90-day plan to build work from zero

Channels in isolation are interesting. A sequenced plan is what actually fills a calendar. Here's how to stack them for a roofing company starting cold.

Days 1-14: Harvest what you already have

• Pull every estimate and customer record from the last 5 years.
• Re-qualify by current roof age and any storm history since you last talked.
• Launch the re-engagement campaign (calls and texts) to the aging/storm-hit subset. This is your fastest cash.
• Simultaneously: claim and fully build out your Google Business Profile, and set up a dead-simple system to ask every past customer for a review.

Days 15-45: Build the outbound engine

• Identify the neighborhoods in your service area with the highest density of aging roofs.
• Mail a targeted, specific postcard to the old-roof homes in those neighborhoods.
• Start knocking those same neighborhoods, qualified doors only, a week after mail drops.
• Document every inspection thoroughly. Show homeowners their own roofs on your phone.
• Track your numbers daily: doors, conversations, inspections, opportunities, signs.

Days 46-90: Tighten and compound

• Re-mail the same targeted list (hit two, hit three). Repetition is where mail pays.
• Knock the neighborhoods that produced best; drop the ones that didn't.
• Bank every review you can; you're building the inbound asset.
• Build a referral ask into your job-completion process so the flywheel finally starts: a satisfied customer, asked directly, with a card to hand a neighbor.

The goal by day 90 isn't a finished machine; it's three things turning at once: warm re-engagement producing fast jobs, targeted outbound producing steady jobs, and the first reviews and referrals starting to compound for the long game.

The numbers to watch (and the ones to ignore)

A roofer building from zero drowns in vanity metrics if they're not careful. Track the few that actually predict revenue:

Don't obsess over doors knocked or pieces mailed in isolation, those are inputs, and a high input with a low inspection rate just means you're working the wrong roofs faster. The metric that exposes targeting quality is inspection-to-opportunity. If it's low, you're knocking and mailing roofs that don't need you, and no amount of extra effort fixes a list problem.

The targeting problem, and where data does the heavy lifting

Every channel above hinges on one thing: knowing which roofs are actually due before you spend money on the homeowner. Re-qualifying your old list, picking which neighborhoods to mail, deciding which doors to knock, all of it is a targeting decision. Get it right and every channel gets cheaper and converts better. Get it wrong and you're back to working the whole street.

Doing this by hand is possible but slow. You can eyeball roofs from the truck, pull permits, cross-reference age hints, and check storm reports. It works, but it doesn't scale past a few streets, and "year built" leads you astray on every re-roofed house.

This is the gap RoofPredict is built to close. It reads aerial imagery to estimate each roof's age as a range (not an exact date, no one can promise that from the sky), and it models storm physics per roof, scoring how hail and wind actually hit a specific house rather than just whether a storm passed through the ZIP. A hail map tells you where it hailed; modeling each roof tells you which ones the storm likely wore out. The output is a ranked picture of your area, which homes are old enough or storm-worn enough to be worth your time, so you knock and mail the right doors and skip the new ones.

It's worth being clear about what that is and isn't. It's a targeting tool, not a lead-buying service, it doesn't sell you the same homeowner it sold to five competitors. It doesn't measure the roof for you (that's a different category of tool) and it doesn't claim to know the exact day your shingles were installed. What it does is enrich your own outreach, your own list, your own streets, with roof-age and storm signals so your reps, your mail, and your follow-up land on the homes that can actually convert. For a roofer building work without a referral network, that's the difference between a channel that pays and one that bleeds: not working harder, but stopping the spend on roofs that were never going to buy.

Use it as the qualification layer underneath the whole 90-day plan: to re-rank your old estimate list by current roof age, to pick which neighborhoods get the mail, and to hand canvassers a per-home reason to knock so a green rep sounds like a veteran without climbing a ladder first.

Mistakes that quietly kill a roofing company's growth

A few patterns show up over and over in roofing companies that struggle to build work without referrals. Avoid these and you're ahead of most of your competition.

• Working the whole street. Covered at length above, and it's the master mistake. Every dollar spent on a new roof is a dollar gone.
• Confusing year built with roof age. Re-roofs are invisible to property records. Old house, new roof, wrong target.
• Never asking for reviews or referrals. You do good work and stay silent about it. The flywheel never starts.
• Quitting a channel after two weeks. Mail needs repetition; door knocking needs reps to climb a learning curve; inbound takes months. Most roofers quit right before the channel would have paid.
• Burning out green reps on bad doors. The fastest way to lose money on canvassing is to put new hires on random streets where they can't win. They quit, you re-hire, the cost compounds.
• Crossing the insurance line. Promising payouts, eating deductibles, advertising free roofs, handling claims. It feels like it helps you close. It can end your business.
• No documentation discipline. Showing up without a phone full of photos. The photos make the sale, and they're your record if a claim is ever involved.
• Chasing only storm work. Storms are feast and famine, and out-of-town crews swarm in and compete every nail. A roofer who can also work their own aging roofs and own list has steady work storm or not, and isn't at the mercy of the weather.

The bottom line

You don't need a referral network to build steady roofing work. You need a way to consistently get your truck in front of homeowners whose roofs are actually due, and you need the discipline to stop spending on the ones that aren't.

Start with the list you already own, the old estimates and past customers, re-qualified by current roof age. Layer in targeted door knocking and targeted mail on the neighborhoods thick with aging roofs. Document every inspection so the photos do the selling. Build your Google presence and reviews quietly in the background so that in a year, the homeowners start finding you. And when a storm hits, work it honestly, document and estimate, hand it to the homeowner, and stay nowhere near the insurance-handling line.

Referrals will come. Do enough good roofs, ask every time, and the flywheel starts to spin on its own. But you don't have to wait for it. The work is sitting on rooftops you can drive past today. The only question is whether you're spending your money knocking on the right ones.

If the targeting piece is what's slowing you down, deciding which roofs are due before you spend a dollar on the homeowner, that's exactly the problem RoofPredict was built to solve: which roofs are old enough or storm-worn enough to be worth your time, scored house by house, so the streets you can see today turn into the jobs you book next month.

FAQ

What's the fastest way to get roofing jobs if I have no referrals yet?

Work your own records first. Pull every estimate and past customer from the last five years, re-qualify them by how old their roof is now and whether any storms have hit since, and reach out with a specific reason ('your roof's in the range where shingles start failing, and we took a storm through here'). These people already know your name, so the cost is just a phone call and the appointment rate is far higher than cold outreach. It's the highest-ROI work you can do this week.

Is door knocking still worth it for roofers?

Yes, when it's done on qualified doors instead of whole streets. The fatal mistake is random knocking, which burns out reps and produces almost nothing. Knock only roofs that are aging or storm-worn, lead with a specific observation about that roof, offer a genuine free check-up, and use the ladder as the close: once you're up top with photos, you're in a real conversation. A focused rep on good doors easily outproduces one knocking three times as many random doors.

Why shouldn't I just mail or knock the whole neighborhood?

Because most of any neighborhood doesn't need a roof. A home re-roofed four years ago is a guaranteed no, and every dollar you spend mailing or knocking it is wasted gas, postage, and payroll. The entire economics of outbound roofing marketing live in list quality: a mediocre postcard to the right 500 old-roof homes beats a beautiful one to the wrong 5,000. Target the roofs that are actually due and every channel gets cheaper and converts better.

Can't I just use 'year built' to find old roofs?

No, and it's a common trap. Year built tells you the house is old, not the roof. A 1968 house may have been re-roofed three times, and that last re-roof is invisible to property records and real-estate sites. Treat year built as a weak hint at best. To actually target, you need a read on current roof age and condition, from aerial imagery, visible wear, or a tool that estimates roof age as a range, not the year the house went up.

How much does direct mail cost and does it work for roofing?

All-in cost runs roughly $0.50-$0.70 per piece depending on volume and format. It works, but only with two things: a targeted list of aging roofs (not a whole ZIP) and repetition. One mailer is a lottery ticket; the same targeted list hit three to four times over a season is a campaign. Pair it with door knocking on the same list a week later so the door isn't cold, and your cost per qualified conversation drops sharply versus blanket mail.

How long until inbound leads (Google, reviews, website) start working?

Plan on 6-18 months for search and reviews to produce steady inbound, which is exactly why you run outbound to pay the bills in the meantime. Start now anyway: claim and fully build your Google Business Profile, ask every happy customer for a review with a direct link while you're in their driveway, and post real before/after job photos. Those compound quietly and eventually lower your cost per job, but they won't fill next week's calendar.

Can I handle a homeowner's insurance claim to win storm jobs?

No. Negotiating, adjusting, or 'handling' a homeowner's claim is unlicensed public adjusting in most states and is illegal for a contractor. You also can't interpret their policy, promise a specific payout or approval, waive or absorb the deductible, or advertise a 'free roof.' What you CAN do is inspect, document the damage thoroughly with dated photos, write an accurate repair estimate, and hand it to the homeowner. They file the claim; the insurer decides coverage. Stay on the documentation-and-estimate side of that line.

Is it ever okay to cover a homeowner's deductible to close a roof?

No. Waiving, absorbing, or 'eating' a deductible is considered insurance fraud in most states and can end your business and license. The deductible is the homeowner's legal obligation under their policy. Advertising a 'free roof' or 'you pay nothing' implies the same thing and carries the same risk. Compete on quality, documentation, and honest service, not on illegally erasing the homeowner's financial responsibility.

How does RoofPredict help if it isn't a lead service?

It solves the targeting problem that sits underneath every channel. It reads aerial imagery to estimate each roof's age as a range and models storm physics per roof, then ranks your area so you can see which homes are old enough or storm-worn enough to be worth your time. It doesn't sell you leads (it never resells the same homeowner to your competitors), doesn't measure the roof, and doesn't claim an exact install date. It enriches your own list and your own streets so your mail, knocks, and follow-up land on roofs that can actually convert.

How do I stop my new canvassers from quitting?

Put them on doors that can convert. The number one cause of canvasser churn isn't a weak script, it's green reps knocking random streets where almost nobody needs a roof, getting demoralized, making no money, and quitting in six weeks. When a new rep knocks roofs that are genuinely due, they get conversations, they close, they make money, and they stay. Hand them a per-home reason to knock and they'll sound like a veteran without first climbing a ladder.