Why Is My Roofing Canvassing Conversion Rate Low? A Field Diagnosis

You sent four reps out for eight hours. Between them they knocked maybe 400 doors, had 60 real conversations, set 9 inspections, and walked away with 2 signed contracts. By Friday you're staring at the payroll and the gas receipts and asking the same question every roofing owner eventually asks: why is my canvassing conversion rate so low?

The honest answer is that "canvassing conversion rate" is not one number. It's a chain of four or five separate conversions multiplied together, and when the final number looks bad, the real failure is almost always hiding in one specific link. A crew can have a beautiful pitch and still book nothing because they're knocking 12-year-old roofs in a subdivision that was re-roofed last spring. Another crew can knock perfect doors and still book nothing because the opener sounds like every solar and pest-control rep who came before them. The skill is figuring out which link is broken before you blow another month of payroll guessing.

What follows is the way experienced sales managers actually run this diagnosis. We'll define every number in the funnel, give you benchmark ranges that aren't fantasy, walk door-knock math you can run on a napkin, and go link by link through the five reasons crews stall. There are scripts, a contact-rate table, a rep-ramp timeline, and a section on the one input almost nobody measures: door quality. By the end you'll be able to point at the exact stage that's leaking and fix that stage instead of throwing motivation at the whole machine.

First, define the rate you're actually trying to fix

Most owners say "conversion rate" and mean completely different things. One means contracts per door. One means contracts per conversation. One means contracts per inspection. Until everyone uses the same denominators, you can't compare reps, can't compare neighborhoods, and can't tell whether a change helped.

Here is the canvassing funnel broken into the stages that matter. Track every one of them. If you only track the first and last, you are flying blind in the middle, and the middle is where almost all the money leaks.

Stage	Definition	What it tells you
Doors knocked	Physical doors approached	Effort / activity volume
Contacts	Doors where a person actually answered and engaged	Are people home? Right time of day?
Conversations	Contacts that got past the opener into a real exchange	Is the opener working?
Inspections set	Homeowner agrees to a roof inspection / appointment	Is the offer compelling at the door?
Inspections completed	You actually got on or looked at the roof	Are appointments holding?
Contracts signed	Signed agreement to do work	Is the inspection-to-close process working?

Now the math that matters. Your door-to-contract rate is the headline, but it's the product of every stage:

Contract rate = (Contacts/Doors) x (Conversations/Contacts) x (Inspections/Conversations) x (Completed/Set) x (Contracts/Completed)

If each of those five fractions is mediocre, the product is catastrophic. Say each stage runs at a middling level: 35% answer the door, 50% of those engage, 25% of conversations set an inspection, 70% of set inspections actually happen, and 35% of inspections close. Multiply: 0.35 x 0.50 x 0.25 x 0.70 x 0.35 = about 1.07%. That means roughly one contract per 93 doors. Knock 400 doors, get four contracts. That's the trap: every individual number looks "fine," and the product is dismal, because multiplying five fractions punishes you brutally.

Now nudge two stages. Get answer rate to 45% by knocking the right hours, and set rate to 35% by fixing the opener and walking up with a reason to be there. Same other numbers: 0.45 x 0.50 x 0.35 x 0.70 x 0.35 = about 1.93%. You nearly doubled output by improving two links, not by working harder. That is the entire game.

The single most useful number to start with

If you're going to track one ratio first, track inspections set per real conversation. It strips out "nobody was home" (a door-quality and timing problem) and "didn't close" (an estimating and follow-up problem) and isolates the part canvassing actually controls: can your rep, in a sixty-second curbside exchange, earn a look at the roof? Good residential crews running warm, non-storm neighborhoods land somewhere around 15-30% of genuine conversations into a set inspection. In a fresh storm zone where homeowners are already worried, that can run much higher. If you're under ~10%, your problem is at the door, not on the roof.

Benchmark ranges (and why you should distrust anyone quoting one number)

There is no universal "good" roofing canvassing conversion rate, and anyone who gives you a single figure is selling something. The number swings massively based on three things: whether there's been a recent storm, whether it's retail/age-driven replacement, and how warm the door is before you knock it.

Use these as honest planning ranges, not promises. Your market will differ.

Scenario	Door-to-inspection-set	Inspection-to-contract	Rough doors per contract
Cold retail, random subdivision	3-8% of contacts	25-40%	120-250+
Cold retail, pre-qualified by roof age	8-15% of contacts	30-45%	60-120
Fresh storm zone, visible damage	20-40% of contacts	35-55%	25-70
Referral / past-customer follow-up	30-60% of contacts	45-70%	varies, very low

Two things to notice. First, the storm and the past-customer columns aren't "better salespeople" — they're warmer doors. The homeowner already has a reason to say yes before anyone speaks. Second, the gap between "random subdivision" and "pre-qualified by roof age" is enormous, and it's purely a targeting input. You can double your effective rate without touching the pitch by simply not knocking new roofs.

Keep your own rolling baseline. The benchmark that matters is your crew's number last month versus this month on comparable doors. National ranges are for sanity-checking, not for performance reviews.

The five reasons your conversion rate is low

Almost every stalled canvassing operation breaks at one (or two) of these five links. Diagnose in this order, because earlier links poison everything downstream — a perfect pitch on a wrong door still fails.

Reason 1: You're knocking the wrong roofs (the targeting problem)

This is the most common cause and the most invisible, because it doesn't feel like a problem. The rep is knocking, the homeowner is polite, nothing seems broken — but the house simply doesn't need a roof, so there was never a sale to win. You can't out-pitch a roof that's eight years old.

Here's why this hides so well. A subdivision built in 2006 looks uniform from the street, but the roofs are not uniform. Some were replaced after a 2017 hailstorm. Some were replaced at resale. Some are still original and well past their service window. From the curb, a green canvasser cannot tell a 9-year-old roof from a 22-year-old roof — they're often the same color and the granule loss isn't visible from the sidewalk. So the crew sprays the whole street evenly, and most of those doors had no job behind them no matter what was said.

What the data says about timing: asphalt shingle roofs, which dominate U.S. residential, generally carry a service life in the rough range of 15-25 years depending on product, ventilation, and climate exposure, per industry technical guidance. That means in any given neighborhood, a meaningful slice of homes are inside the replacement window and a meaningful slice aren't — and the slices are interleaved house by house. Knocking them as if they're identical is the targeting error.

How to know this is your problem:

• Reps report lots of "nice conversations" but few real needs surface.
• Your inspection-to-contract rate is healthy (30%+) but door-to-inspection is in the gutter. That pattern means the people who do let you up the ladder convert fine — you just aren't finding enough roofs that warrant it.
• You're working neighborhoods chosen by drive-time or gut, not by any roof signal.

The fix is to stop knocking doors at random and start knocking doors with a reason. There are a few ways to do that, in rough order of cost:

1. Permit data. Many counties publish re-roof permits. Pulling them tells you which homes were recently re-roofed so you can skip them — a roof permitted in 2021 is dead weight on your route. This is tedious and incomplete (lots of work happens without permits) but free.
2. Age inference from build year — with a heavy caveat. Property records give you year built. That is a weak proxy and a common trap: year built is not roof age. A 1998 house may be on its third roof. Tax and listing records show the structure's age, not the covering. Use build year only as a faint first filter, never as truth.
3. Visual aging signals on the route. Train reps to read curb-visible cues — heavy granule loss in gutters, dark streaking, curling or cupping at the edges, patched sections, moss on north faces. These are real but only catch advanced wear, and a rep can't see most roof planes from the street.
4. Roof-age and storm modeling per address. This is where modern targeting data earns its keep, covered in its own section below — because the entire point is to rank a street by which roofs are actually due before anyone leaves the truck.

The payoff math is the cleanest in the whole operation. If half the doors on a random street were never going to buy because the roof is too young, then cutting those doors doesn't just save time — it roughly doubles your effective conversion rate on the doors you do knock, because the denominator drops without the numerator changing. Targeting is the highest-leverage fix because it multiplies through every downstream stage.

Reason 2: Your opener sounds like every other knocker

Assume the roof is right and someone's home. The next ten seconds decide everything, and most crews fail them. The homeowner's brain is running one program when a stranger appears on the porch: "What are you selling and how do I make this end?" If your opener confirms "I'm a salesperson here to sell you a roof," you've activated every defense they have.

The classic weak opener: "Hi, we're doing roofs in the neighborhood, are you interested in a free inspection?" It's weak because it's generic (every trade says "in the neighborhood"), it leads with "free" (which reads as a hook, not a reason), and it asks a yes/no question that the homeowner can end with one word. "No thanks." Door closed.

What strong openers share:

• A specific, true reason for being at this house. Specificity signals you're not spraying the street. "I'm not knocking the whole block — I'm at the houses where the roof's getting up there in age" is dramatically more disarming than "the neighborhood."
• They lower the stakes of the ask. You're not asking them to buy a roof. You're asking for thirty seconds and, ideally, offering to leave them something useful whether they hire you or not.
• They give the homeowner an easy, low-commitment yes. People say yes to "Can I show you something quick?" far more than to "Are you interested in a roof inspection?"
• They sound like a person, not a script. Memorized cadence reads as fake. Reps should internalize the beats, not parrot the words.

Here's a structure that works, broken into beats so reps adapt the wording:

1. Pattern interrupt + disarm (5 sec): "Hey, I'll be quick, I'm not trying to sell you anything at the door." (Names the fear and removes it.)
2. Specific reason you're here (10 sec): "I work with [Company], we're a local roofing crew, and I'm specifically stopping at the houses on this street where the roof looks like it's getting toward the end of its life."
3. The low-stakes offer (10 sec): "I'm not asking you to decide anything. I can take a quick look and leave you a simple write-up on where your roof actually stands — age, any wear, what to watch — so you've got it whether you ever do anything or not."
4. Easy yes (5 sec): "Worst case you learn your roof's fine and I'm out of your hair. Want me to take a look while I'm here?"

Notice what's missing: no "free," no "limited time," no claim about insurance or payouts. Notice what's present: a reason this house, a deliverable they keep, and a one-step yes. The deliverable is doing real work — "I'll leave you a write-up on your roof" converts far better than "free inspection" because it reframes the interaction from sales call to getting useful information.

A note on storm-zone openers, which need extra care for compliance reasons covered in Reason 5: leading with "there was a storm and you may have damage" is fine as a fact, but the moment a rep starts promising the insurance company will pay, that the deductible disappears, or that they'll "handle the claim," they've crossed a legal line and also blown their credibility. Keep the door opener about looking at the roof and documenting what's there, not about money the homeowner hasn't been promised by anyone.

Reason 3: Nobody's home (the contact-rate and timing problem)

If your reps are knocking hard but the answer rate is low, you can have a perfect pitch and a perfect roof and still book nothing — because there's no one to say yes. Contact rate is a timing and route problem, and it's wildly underappreciated.

Answer rates swing enormously by hour and day. Knock a residential street at 1 p.m. on a Tuesday and most working households are empty; you're talking to retirees and remote workers. The same street from late afternoon into early evening, and from late morning on weekends, fills up. Roughly:

Window	Typical residential answer rate	Notes
Weekday 9 a.m.-noon	Low	Mostly empty; retirees, shift workers, remote
Weekday noon-4 p.m.	Low-moderate	Dead zone for most working households
Weekday 4-7 p.m.	High	Prime knocking window
Saturday 10 a.m.-5 p.m.	High	Best volume window of the week
Sunday afternoon	Moderate-high	Good, but respect the door faster
After ~8 p.m. / dark	Avoid	Safety, legality, and it annoys people

If you're sending crews out at 9 a.m. and pulling them at 4 p.m., you're spending your reps' freshest energy on the emptiest streets and quitting right as the doors fill up. Re-shape the day around contact rate: lighter logistics/admin in the dead midday hours, and stack your real knocking into the 4-7 p.m. weekday window and the weekend daytime block.

Also watch local rules. Many municipalities require a door-to-door solicitation permit and set lawful hours; some streets post no-soliciting. Knowing and following local ordinances isn't just legal hygiene, it's reputation — and a rep who respects a "no soliciting" sign and moves on protects the brand for the whole crew. Check your city or county requirements before you deploy.

A second contact-rate lever: route density. A rep who walks 40 doors in a tight cul-de-sac loop gets far more conversations per hour than one driving between scattered addresses. The tighter your target list clusters geographically, the more doors per hour, the more contacts, the more at-bats. This is another reason random targeting hurts twice — scattered "maybe" doors waste windshield time that should be knock time.

Reason 4: Your reps aren't ramped (the training and consistency problem)

When one rep on the crew is closing and three aren't, the system isn't the problem — the ramp is. Door-to-door is a learned skill with a brutal early failure curve, and most roofing companies under-train, under-script, and under-support new canvassers, then wonder why they churn out at three weeks having booked nothing.

The uncomfortable truth: a brand-new canvasser will be bad for a while, and that's normal. The job is to shorten "a while" and keep them alive through it. A realistic ramp:

Phase	Weeks	What "good" looks like
Survival	1-2	Can deliver the opener without freezing; not taking rejection personally; learning the routes
Competence	3-6	Conversations feel natural; consistently setting some inspections; reading roofs at the curb
Productivity	7-12	Hitting team-average set rates; handling common objections without a manager
Veteran	12+	Above-average sets, mentoring new reps, trusted on tougher doors

Most canvasser churn happens in the Survival phase, and it's almost always because the rep got zero wins and zero feedback, just rejection. The fix is structural, not motivational:

• Shadow then be shadowed. New reps spend day one watching a veteran, days two-three being watched and corrected after each door, not at end of day. Real-time feedback compresses the learning curve.
• Give them a real reason at the door. This loops back to targeting: a green rep who knocks a door already flagged as "likely old roof" and can say "your roof looks like it's getting up there" sounds competent on day one. A green rep knocking random doors with nothing to anchor on sounds like a kid reading a script. Good door selection is the cheapest training aid you have — it lets a new hire sound like a vet without ten years of pattern recognition.
• Script the objections, not only the opener. Most reps freeze on the third objection, not the first. Drill the common ones.
• Set activity goals, not only outcome goals, for new reps. Week one's goal is "deliver 50 clean openers," not "book a job." You can't control outcomes through a brand-new rep; you can control reps and conversations.

Objection drills that matter (have a real answer for each):

• "I just had my roof done." → "Perfect, then you've got nothing to worry about — mind if I ask roughly when? I keep notes so we don't bug you again." (Confirms the door is dead, leaves a good impression, and feeds your list.)
• "I'm not interested." → "Totally fair, most people aren't until something leaks. I'm only leaving a quick note on where your roof stands so it's not a surprise later — takes me two minutes."
• "How much does a roof cost?" → "Depends entirely on the roof, which is why I don't quote at the door — that'd be guessing. Let me actually look and I'll give you a real number, no pressure to use it."
• "Are you trying to get my insurance to pay?" → "I document what's actually on your roof and give you the write-up. Whether anything goes to insurance is your call and theirs, not mine. I'm just here to tell you the condition." (This answer is also your compliance shield — see Reason 5.)

Consistency is the other half. A rep who knocks hard Monday, coasts Wednesday, and disappears Friday will never build the rep-density and rhythm that produces sets. Track activity per rep daily — doors, contacts, conversations, sets — and the inconsistent performers reveal themselves instantly. You manage what you measure.

Reason 5: The handoff and follow-up leak (sets that never close)

The last link: you set the inspection and it still doesn't become a contract. Two distinct leaks live here, and they need different fixes.

Leak A: the inspection never happens. A rep sets an appointment for "Thursday" with no confirmed time, no confirmation call, no calendar invite — and the homeowner forgets, leaves, or cools off. No-show rates on loosely-set canvassing appointments are punishing. Tighten it: set a specific time, confirm contact info on the spot, send a same-day text confirmation, and call the morning of. If your set-to-completed rate is below ~60%, this is almost certainly your leak, and it's pure process, not sales skill.

Leak B: the inspection happens but doesn't close. Now you're past canvassing and into your estimating and documentation game, which is its own discipline. The common failures:

• No documentation the homeowner can hold. If your rep climbs down and says "yeah it's worn, we can do it for $14,000" with nothing in writing and no photos, the homeowner has no reason to trust the number or act now. Hand them something real: dated photos of the actual conditions, a clear written assessment of what you found, and a line-item estimate.
• A vague estimate. "Roof replacement: $14,000" loses to a line-item scope every time. Itemize tear-off, decking inspection/replacement allowance, underlayment, the shingle product and color, flashing, ventilation, pipe boots, and cleanup. A detailed scope reads as competence and justifies the price.
• Storm-claim overreach. This is where roofers torch deals and invite legal trouble. The compliant, durable position: you can inspect the roof, document damage thoroughly with dated photos, and write an accurate, itemized repair estimate aligned to standard estimating practice, then hand that documentation to the homeowner. The homeowner files their own claim, and the insurer decides coverage. You do not, for a fee, negotiate or "handle" the claim, interpret the homeowner's policy, promise a specific payout or approval, promise the deductible will be waived or absorbed, advertise a "free roof," or represent the homeowner against their insurer — those acts stray into unlicensed public adjusting and deceptive-practice territory in many states. Stay on the documentation and estimate side and you keep both your license posture and your credibility intact.

The do-not-say list every storm canvasser should have memorized, because one of these sentences can blow a deal and draw a regulator:

• "We'll get your insurance to pay for the whole thing."
• "Your deductible will be waived / we'll eat it / it's basically free."
• "We handle the whole claim for you."
• "You're definitely covered / this is definitely approved."
• "Free roof."

Replace all of them with the same honest move: "I document exactly what's on your roof and give you a detailed estimate. You decide whether to file; your insurer decides what they cover." It's legal, it's true, and homeowners trust it more than the magic-payout pitch precisely because it doesn't sound too good to be true.

Where roof-age and storm data fits the targeting fix

Reason 1 — knocking the wrong roofs — is usually the biggest single leak, and it's the one you can fix before a rep ever opens their mouth. The whole point is to walk a street already knowing which roofs are due and which to skip, so every door has a job behind it.

This is the gap RoofPredict is built for. Instead of spraying a subdivision evenly, you get a per-address read on which roofs are likely due, house by house, by combining two signals: an estimated roof-age range read from aerial imagery, and storm exposure modeled on each individual roof — not merely "a storm passed through this ZIP," but a physics-based estimate of the hail and wind each specific roof actually took. That lets you rank a street so your crew knocks the worn-out and aging-out roofs first and skips the ones that were re-roofed last spring. It also enriches your own list — your CRM, your old estimates, a mailing list you already own — with those age and storm signals, so you're not buying anyone's leads, you're sharpening the outbound you already do.

Honest limits, because anyone who oversells this is lying to you:

• Roof age comes back as a range, not a date. Aerial inference gives you something like "this roof reads 18-22 years" — a tight, useful window for prioritizing a route, not a birth certificate. You still confirm on the roof.
• Storm modeling gives odds, not proof. "This roof likely took significant hail" is a probability that ranks your route and justifies a look. It is not evidence of damage and is never a promise of an insurance outcome. The ladder still settles the question.
• It targets; it doesn't sell. The data tells you which doors are worth a knock. The opener, the documentation, and the estimate still have to be good — every other reason on this list still applies.

Used honestly, the effect on the funnel is exactly the multiplier from the math section: when a far higher share of the doors you knock actually have a roof behind them, your door-to-inspection rate climbs without changing the pitch, your reps get more real conversations per hour because the list clusters, and your green hires sound competent because they're knocking with a reason. You can hand RoofPredict a roof you already know the answer on and judge whether the read holds up before you trust it on a street you don't know.

How to actually track the funnel (without buying anything fancy)

Everything above depends on one habit most crews skip: writing down the middle stages. Owners track doors knocked because it feels like effort, and they track contracts because that's the money — but the diagnostic gold is in contacts, conversations, and sets, and those are exactly the numbers nobody captures. You don't need a six-figure platform to fix this. You need a discipline.

The minimum viable tracker is a shared sheet with one row per rep per day and these columns: date, rep, neighborhood, doors knocked, contacts (someone answered and engaged), conversations (got past the opener), inspections set, inspections completed, contracts signed, and a notes field. That's it. At end of day each rep enters their tallies; the manager scans for the stage where the funnel collapses. If a rep posts 100 doors, 40 contacts, 30 conversations, and 1 set, the collapse is conversation-to-set — a pitch-and-offer problem. If they post 100 doors, 12 contacts, the collapse is contact rate — a timing or door-quality problem. The numbers point at the link.

A few rules to keep the data honest:

• Count contacts and conversations separately and define them tightly. A contact is a human who answered and didn't immediately shut the door. A conversation is one that engaged past the opener — they asked a question, let the rep talk, or pushed back with something other than "no thanks." Reps will blur these unless you draw the line sharply.
• Count doors as physical doors, not addresses you drove past. Inflated door counts make every downstream ratio look worse than it is and hide real progress.
• Tag the neighborhood and the targeting source. When you compare "random street" rows against "age-ranked street" rows after a couple of weeks, the targeting argument proves itself in your own data, not in a benchmark table.
• Review weekly, not only monthly. The whole point of leading indicators (doors, contacts, conversations) is that they tell you something is wrong before the lagging indicator (contracts) confirms it a month late.

When you outgrow the sheet, a CRM with a mobile canvassing app and territory mapping pays for itself by capturing this automatically and letting you see set rates per rep and per neighborhood at a glance. But buy the tool to scale a habit you already have — buying software to create the habit usually fails, because the reps never enter the data and the dashboard sits empty. Get the manual version working first.

Leading vs lagging indicators

The reason this matters is timing. Contracts are a lagging indicator — by the time they're down, the damage is two to four weeks old. Doors, contacts, conversations, and sets are leading indicators that move first. A manager watching only contracts is always reacting late. A manager watching conversation-to-set daily catches a slumping rep or a bad neighborhood on day two, not at the next pay period. Manage the leading numbers and the lagging number takes care of itself.

Reading a roof from the curb: the skill that anchors the opener

Reason 1 is about choosing the right streets before you deploy. But once a rep is standing on a sidewalk, a second skill multiplies everything: reading visible roof age and wear at a glance. This is the difference between a rep who says "we're doing roofs in the neighborhood" and one who says, with specificity the homeowner can feel, "I noticed your shingles are losing a lot of granules along the south side." Specific observation is the single most disarming thing a canvasser can do, and it only comes from knowing what to look for.

Train every rep on these curb-visible signals, roughly in order of how strongly they suggest a roof is near the end of its service life:

• Granule loss in the gutters and at downspout splash blocks. Asphalt shingles shed their protective granules as they age and after hail. A gutter line full of black grit, or a pile of granules at the bottom of a downspout, is one of the most reliable visible age signals. Reps can often spot it from the driveway.
• Surface streaking and bald spots. Dark streaks are often algae (cosmetic), but broad areas where the shingle looks shiny, smooth, or a different shade than the rest signal granule loss and exposed asphalt — genuine wear.
• Curling and cupping at shingle edges. As shingles age and dry out, edges curl up or the centers cup. Visible curling from the ground usually means an advanced-age roof.
• Patched or mismatched sections. A rectangle of newer-looking shingles means a prior repair, which is a clue the roof has a history worth asking about — and sometimes a clue of an unresolved leak.
• Sagging rooflines or visible decking deflection. A roof plane that dips or waves can signal decking or structural moisture issues. This is a strong reason to look closer and a strong talking point.
• Moss and heavy organic growth, especially on north-facing planes. Moss holds moisture against the roof and accelerates deterioration; thick growth suggests an older, shaded, moisture-retaining surface.
• Exposed or deteriorated flashing, rusted vents, cracked pipe boots. These accessories fail before the field of the roof and are common leak origins. A rep who notices a cracked pipe boot has a precise, credible reason to ask for a closer look.

The honest limit, and reps must own it: from the curb you can see at most one or two roof planes, and the worst wear is often on the slope you can't see. Curb reading catches advanced wear and gives the rep a credible opener, but it misses plenty of due roofs whose visible side still looks fine — which is exactly why pre-deployment targeting by roof-age range matters. The two skills stack: data picks the street and ranks the doors, curb reading sharpens the specific opener at each one. A rep armed with both walks up to a flagged-old roof, points at the granule-filled gutter, and sounds like someone who's been doing this for twenty years, even if it's week three.

A short curbside checklist reps can run in ten seconds before they knock:

1. Glance at the gutters and downspout base — granules?
2. Scan the visible slope — streaking, bald patches, color unevenness?
3. Check the edges — any curling or cupping?
4. Look for patches, sagging, moss, and rusted/​cracked accessories.
5. Pick the single most specific true thing you see and lead the opener with it.

That last step is the payoff. "I noticed [specific thing]" beats any generic opener because it proves the rep is looking at this roof, not reciting a pitch — and it earns the look up the ladder that the whole funnel depends on.

Storm canvassing: the same funnel, higher stakes

Storm-driven canvassing runs on the same five links, but two of them change character enough to deserve their own treatment, because storm work is where crews make their best months and also where they get themselves in legal trouble.

Targeting changes from age to exposure. In retail canvassing you're hunting roofs that aged out. After a storm you're hunting roofs that took a beating — and the trap is assuming an entire ZIP code got hit evenly. It didn't. Hail falls in swaths and varies street to street and even roof to roof depending on slope orientation, exposure, and the storm's exact track. A crew that knocks a whole ZIP because "there was a storm" wastes most of its day on roofs that took little. The targeting win is knowing which roofs in the affected area actually saw significant hail or wind, which is precisely what per-roof storm modeling estimates — not "the storm passed here" but "this specific roof likely took meaningful impact." Pair that with roof age and you rank the route by genuine probability of a justified inspection.

The opener and the close get a compliance overlay. Storm homeowners are warmer — they already suspect they have damage — so set rates run higher. That warmth is also where reps overreach and torch the deal. The durable storm posture is the same one from Reason 5, stated plainly to the homeowner: you'll get on the roof, document exactly what's there with dated photos, and write an accurate, itemized estimate they keep. You will not promise their insurer pays, will not touch their deductible, will not "handle" the claim, and will not call it a free roof. Beyond the legal exposure, the magic-payout pitch actually lowers trust with the more careful homeowners — the ones most likely to sign a real contract — because it sounds too good to be true. "I document what's on your roof; you and your insurer decide the rest" is both the compliant line and the higher-converting one.

One more storm-specific operational note: speed of documentation matters, but speed of promises destroys you. Get the inspection and the written, photo-backed estimate done quickly and thoroughly so the homeowner has real information in hand. Resist the urge to convert that speed into guarantees about coverage you don't control and can't legally make.

A 30-day plan to raise your canvassing conversion rate

Don't fix all five links at once; you won't know what worked. Run this sequence.

Week 1 — Measure. Put the full funnel on a simple tracker: doors, contacts, conversations, sets, completed, contracts, per rep, per day. You almost certainly aren't tracking the middle stages today. You can't fix what you can't see. By Friday you'll know which stage is actually leaking.

Week 2 — Fix targeting and timing (the cheap, high-leverage links). Stop knocking random streets. Build target routes from a roof signal — permit pulls to exclude recent re-roofs at minimum, age-and-storm data to rank doors at best. Re-shape the schedule so real knocking lands in the 4-7 p.m. weekday window and weekend daytime. These two changes move the numbers fastest because they multiply through everything downstream.

Week 3 — Fix the opener and objections. Rewrite the opener around the four beats: disarm, specific reason for this house, low-stakes deliverable, easy yes. Drill the top five objections until reps don't freeze. Have every rep deliver it to you cold before they hit doors. Kill every "free," "limited time," and any insurance-payout promise.

Week 4 — Fix the handoff and the close. Tighten appointment setting: specific time, confirmation text same day, call morning-of. Standardize what the homeowner walks away with — dated photos, a written condition assessment, a line-item estimate. Audit every storm interaction against the do-not-say list.

Then recompute the funnel and compare to Week 1 on comparable doors. You'll be able to point at the stage that moved.

A worked example: turning 1% into 2%

Let's make the multiplier concrete with a four-rep crew. Starting numbers, measured in Week 1:

• 400 doors/day across the crew
• 35% answer (140 contacts)
• 50% engage past the opener (70 conversations)
• 14% of conversations set (about 10 sets)
• 65% of sets complete (about 6 inspections)
• 33% close (about 2 contracts/day)

That's ~2 contracts per 400 doors — roughly 0.5% door-to-contract, and it feels like effort isn't paying off, because it isn't.

Now apply the plan over a month, realistically — no fantasy numbers, just fixing two or three links:

• Targeting: cut the third of doors that were re-roofed in the last few years and re-rank around due roofs. Same 400 knocks, but now far more have a job behind them, so conversation-to-set climbs from 14% to 22%.
• Timing: shift to evening/weekend windows, answer rate rises from 35% to 44%.
• Handoff: confirmation texts and morning-of calls lift set-to-completed from 65% to 78%.
• Leave the opener and close roughly where they were, so the gains are conservative.

Recompute: 400 x 0.44 = 176 contacts; x 0.50 = 88 conversations; x 0.22 = ~19 sets; x 0.78 = ~15 inspections; x 0.33 = ~5 contracts/day. You went from ~2 to ~5 contracts a day on the same 400 knocks and the same reps — by fixing where you knock, when you knock, and how appointments hold, not by knocking more. That's the whole argument: canvassing output is a product of fractions, and you win by fixing the weak fractions, not by adding effort to a leaky machine.

Common mistakes that keep the rate low

A quick checklist of the errors that quietly cap conversion, even in crews that work hard:

• Tracking only doors and contracts. Without the middle stages you can't find the leak. This is the number-one operational failure.
• Treating build year as roof age. A 1998 house can be on its third roof. Year built is a faint hint, not a target.
• Knocking the dead midday hours. Burning your reps' best energy on empty streets.
• Leading with "free." It reads as a hook and triggers sales defenses; lead with a reason and a deliverable instead.
• No documentation handed to the homeowner. A verbal price with no photos and no written scope gives them nothing to trust or act on.
• Storm-claim overreach. Promising payouts, waived deductibles, "free roofs," or "handling the claim" — kills trust and invites regulators.
• Under-training new reps, then blaming them. Survival-phase churn is a coaching failure, not a hiring failure.
• Random routes with low door density. Scattered targets waste windshield time that should be knock time.
• No confirmation process on set inspections. Loose appointments no-show at brutal rates.
• Chasing one magic script. No script overcomes a wrong roof, an empty house, or a no-show; the script is one of five links.

The bottom line

A low canvassing conversion rate is almost never one problem and almost never a motivation problem. It's a chain of five conversions multiplied together, and the product looks terrible when even two links are weak. Measure all five stages, find the one or two that are leaking, and fix those specifically. In most stalled operations the biggest leak is targeting — knocking roofs that were never going to buy — followed closely by timing and the appointment handoff. Fix where and when you knock, give reps a real reason at the door and something useful to leave behind, keep storm conversations strictly on the documentation side, and watch the same crew on the same streets produce two or three times the contracts. The effort was never the issue. The aim was.

FAQ

What is a good roofing canvassing conversion rate?

There's no single right number, because it depends on whether there's been a recent storm, whether you've pre-qualified doors by roof age, and how warm the door is. Cold random subdivisions might yield one contract per 120-250 doors; pre-qualified or fresh storm zones can be several times better. Track your own crew's rolling baseline on comparable doors and improve against that rather than chasing a national figure.

How many doors does it take to get one roofing sale?

On cold, randomly-targeted residential streets, plan for roughly 120-250+ doors per signed contract. With doors pre-qualified by roof age, that can tighten to 60-120. In a fresh storm zone with visible damage it can be 25-70. The biggest lever isn't knocking more doors, it's making sure a higher share of the doors you knock actually have a roof old enough or worn enough to need work.

Why do my reps have lots of friendly conversations but book almost nothing?

That pattern usually means a targeting problem, not a pitch problem. You're knocking roofs that don't need replacing, so the homeowners are pleasant but there was never a job to win. Check whether your inspection-to-contract rate is healthy while door-to-inspection is low; if so, the people who let you on the roof close fine and you simply aren't finding enough due roofs. Fix which doors you knock first.

What's the best opener for roofing door knocking?

Use four beats: disarm the sales fear ('I'm not selling anything at the door'), give a specific true reason you're at this house ('I'm stopping at the homes where the roof looks like it's getting up there in age'), make a low-stakes offer with a deliverable they keep ('I'll leave you a quick write-up on where your roof stands'), and ask for an easy one-step yes. Avoid leading with 'free' or a yes/no question they can end with one word.

What hours should I send canvassers out?

Residential answer rates are highest weekday evenings from about 4-7 p.m. and during weekend daytime hours, roughly Saturday 10 a.m.-5 p.m. The weekday midday block is mostly empty for working households. Reshape the day so your real knocking lands in the high-contact windows and the dead midday hours go to admin, routing, and breaks. Always follow local solicitation hours and permit rules.

Can year built tell me a roof's age?

No, and treating it that way is a common targeting trap. Property and tax records show the structure's build year, not when the roof covering was last replaced. A house built in 1998 may already be on its third roof. Use build year only as a faint first filter, then confirm with permit data, curb-visible aging signals, aerial roof-age estimates, or an actual inspection.

How long before a new canvasser is productive?

Plan for a real ramp: weeks 1-2 are survival (just delivering the opener without freezing), weeks 3-6 build competence, and weeks 7-12 reach team-average productivity. Most new-rep churn happens in the first two weeks because they get rejection with no wins and no feedback. Shadow them, give real-time coaching after each door, set activity goals instead of outcome goals early, and send them to pre-qualified doors so they sound competent on day one.

What should a canvasser leave with a homeowner who isn't ready to buy?

Something useful they keep regardless of whether they hire you: dated photos of the actual roof conditions, a short written assessment of the roof's age and any wear you found, and your contact info. This reframes the visit from a sales pitch to free, useful information, which converts far better than a verbal 'it's worn, call us.' It also keeps you on the roof's condition rather than overpromising anything about cost or insurance.

What can a roofer legally say about insurance at the door?

You can state facts: there was a storm, you can inspect and document the roof's condition with dated photos, and you can write an accurate, itemized repair estimate and hand it to the homeowner. What you must not do for a fee is negotiate or 'handle' the claim, interpret the homeowner's policy, promise a specific payout or approval, promise the deductible will be waived or absorbed, advertise a 'free roof,' or represent the homeowner against their insurer, which strays into unlicensed public adjusting in many states. The homeowner files; the insurer decides coverage.

How does roof-age data improve canvassing conversion?

It attacks the biggest leak before anyone knocks: which doors you knock. A per-address roof-age range plus storm exposure modeled on each individual roof lets you rank a street so reps hit the worn-out and aging-out roofs and skip the recently re-roofed ones. That raises door-to-inspection rate without changing the pitch, packs more real conversations into each hour because the list clusters, and lets green reps knock with a reason. The age comes back as a range and the storm read as odds, so you still confirm on the roof.