Roofing Canvassing: The Doors-to-Appointment Ratio Benchmark (And How to Beat It)

Every roofing owner who runs a door team eventually asks the same question, usually after a rep comes back from a four-hour shift with one appointment and a story about a dog. "Is that normal? How many doors should it take to book one?" The honest answer is that the number ranges more than anyone selling you a course wants to admit, and the version floating around the internet — "knock 100, set 10" — is a fantasy on most streets in most markets on most days.

What follows is the real math, pulled from how a working door operation actually keeps score: what counts as a door, what counts as an appointment, what a healthy ratio looks like in a non-storm market versus a fresh hail event, why two reps on the same street post wildly different numbers, and the specific levers that move the ratio without burning out your team. There's also a section on the single biggest hidden multiplier most owners ignore — which doors get knocked — because a rep working 60 worn-out roofs and a rep working 60 random houses are not running the same experiment, even if their pitch is word-for-word identical.

First, define your terms or your benchmark is garbage

The reason published "benchmarks" are useless is that nobody agrees on what they're counting. Before you compare your team to anything, lock down four definitions and write them on the whiteboard. If your reps count differently from each other, you don't have a metric, you have noise.

A door (a knock). A door is a house your rep physically approached with intent to make contact. Walking past a house because the truck was in the driveway and they didn't feel like it does not count as a door, and it doesn't count as a no — it never happened. Be strict here. The most common way reps inflate their conversion rate is by under-counting the doors they skipped or where nobody answered. If 60 percent of your "doors" are actually no-contacts, your apparent ratio looks great and tells you nothing.

A contact (a conversation). A contact is a live human who opened the door (or was in the yard) and exchanged at least a few words with your rep. This is the number that actually matters, and almost nobody tracks it. The gap between doors and contacts is enormous — in most residential neighborhoods on a weekday afternoon, somewhere between 1-in-3 and 1-in-5 doors produces a live human. Track contacts separately and a lot of mysteries solve themselves.

An appointment (a set). An appointment is a specific, scheduled time for an inspection or estimate that the homeowner agreed to, with a name, address, and time you'd actually put on a calendar. "They said come back sometime" is not an appointment. "Maybe my husband will call you" is not an appointment. A real set has a date attached. The fastest way to make your numbers look good and your business fall apart is to let reps log soft maybes as appointments — then your sit rate craters and you can't figure out why.

A sit (a kept appointment / inspection). A sit is an appointment that actually happened — the homeowner was home, let you up or let you walk the property, and you got to do your thing. The drop between sets and sits is where a lot of canvassing operations quietly bleed. We'll come back to it.

With those four locked, your real funnel reads:

Doors → Contacts → Appointments → Sits → Inspections that turn into work.

The phrase "doors-to-appointment ratio" technically skips two of those stages, which is exactly why it's so abused. A 20-to-1 doors-to-appointment ratio means nothing if half those appointments never sit. Track the whole chain.

One more definition that saves arguments later: decide upfront whether a single rep's two visits to the same house count as one door or two. The cleanest rule is that a door is a house-visit event, so a return knock is a second door — but you tag it as a follow-up rather than a fresh cold door. That way your cold-door ratio and your follow-up ratio stay separate, and you don't accidentally credit your follow-up wins to your cold-knock pitch. Teams that blur this almost always overrate their opener and underrate their persistence, then they coach the wrong thing.

And write down what a "contact" is not. A kid answering the door is not a contact. A renter who can't make a roofing decision is a contact for tracking purposes but a disqualify for the funnel — log the conversation, mark it dead, and move on. A spouse who says "talk to my husband, he's not home" is a half-contact: real human, no decision-maker, prime follow-up. Getting granular here feels fussy for about a week, then it becomes the difference between a team that can explain its numbers and a team that just shrugs.

The actual benchmark ranges (no fairy tales)

Here is the part you came for, with the honesty most articles skip. These are working ranges for residential retail and storm-restoration door knocking, expressed as doors knocked per appointment set, and then the realistic conversion downstream. They are bands, not promises, and where you land inside the band depends heavily on market type, list quality, season, and rep skill.

Read that table slowly, because it kills three myths at once.

First, in a cold, untargeted, non-storm neighborhood, 40 to 100 doors per appointment is normal and not a sign your reps are bad. If someone tells you their team books one in ten cold doors year-round in a calm market, they are either counting no-contacts as wins, calling soft maybes appointments, or selling you something.

Second, a fresh storm event compresses the entire funnel. Homeowners are primed, neighbors are talking, and tarps are visible from the street. Twelve to thirty doors per appointment is achievable when the event is recent and real. This is also why storm markets attract swarms of out-of-town crews — the ratios are temporarily fantastic and everyone knows it.

Third, and most important for the rest: targeting collapses the middle of the funnel even without a storm. When the doors a rep knocks are pre-filtered to roofs that are genuinely old or genuinely storm-worn, contacts-per-appointment drops because more of those conversations are with people who have a real, current problem. That's the lever almost nobody pulls correctly, and we'll spend real time on it below.

A worked example so the math is concrete

Say a rep knocks for a solid four-hour shift and gets through 80 doors (a brisk but realistic pace once you subtract walking, no-answers, and a couple of long conversations). In a cold non-storm market at the middle of the band:

• 80 doors × (1 contact per 5 doors) = 16 contacts
• 16 contacts × (1 appointment per 12 contacts) ≈ 1.3 appointments set
• 1.3 sets × 60% sit rate ≈ 0.8 inspections
• 0.8 inspections × 30% close ≈ 0.24 jobs

So roughly one signed job per four shifts in a cold market. That sounds brutal until you price it out: if your average job nets a few thousand dollars in gross profit, one job per four shifts can still pencil. But it's fragile, and it explains the churn — a green rep who lives that math with no feedback quits inside a month.

Now run the same rep, same pitch, same hours, on a targeted list of worn and aging roofs:

• 80 doors × (1 contact per 5 doors) = 16 contacts
• 16 contacts × (1 appointment per 7 contacts) ≈ 2.3 appointments set
• 2.3 sets × 68% sit rate ≈ 1.6 inspections
• 1.6 inspections × 38% close ≈ 0.6 jobs

Same human, same words, roughly 2.5x the output, purely from knocking better doors. The pitch didn't change. The list did. Hold that thought.

Why the band is so wide

Owners hate hearing "40 to 100" because they want one number. But the width of that band is the truth, and the things that swing you from one end to the other are mostly knowable. Here's what pushes a cold-market team toward 40 doors per appointment (the good end) versus 100+ (the grind):

• Season. Late spring through early fall, roofs are top-of-mind and people are outside; ratios tighten. Deep winter in a cold climate, people don't answer and don't want a ladder on their house; ratios stretch.
• Recency of any weather. Even a minor wind event a few weeks back primes a neighborhood. "Did your area get that wind last month?" is a real, verifiable opener that lifts contacts-per-appointment.
• Neighborhood age mix. A subdivision where most roofs are 18-plus years old converts far better than a five-year-old development where everything is new. This is the variable targeting controls directly.
• Median home value and roof complexity. Higher-value homes with steep, cut-up roofs skew toward insurance-and-replacement conversations; starter neighborhoods skew toward repair-and-defer. Neither is wrong; they just run different funnels.
• Competitive saturation. If three other crews knocked the street last week, your ratio suffers regardless of skill. Knock first or knock different doors.

When you log doors, contacts, and appointments by neighborhood, these stop being theories and become a map of where your reps should be standing.

Why two reps on the same street post different numbers

Before you blame the list or the market, understand that rep-to-rep variance on identical streets is often larger than market-to-market variance. The same 60 houses can produce one appointment for one rep and four for another. The drivers, in rough order of impact:

Hours of contact, not hours clocked. A rep who knocks from 4:30 to 7:30 on a weekday hits people coming home from work. A rep who knocks from 1:00 to 4:00 hits empty houses and retirees. Same effort, completely different contact rate. Track contacts-per-hour by time-of-day and you'll find your golden window. In most residential markets it's late afternoon into early evening on weekdays and mid-morning to early afternoon on Saturday.

The first ten seconds. Roofing reps lose more doors in the opening line than anywhere else. "Hi, I'm with ABC Roofing, do you need a new roof?" gets a reflexive no. The reps who set appointments lead with a specific, true, local reason they're standing there — something the homeowner can verify with their own eyes — and they ask permission to share it rather than pitching.

Disqualifying fast. Good reps spend almost no time on doors that won't convert and pour their energy into the ones that will. A rep who tries to win every door books fewer appointments than a rep who reads the room, leaves the wrong doors quickly, and is fully present on the right ones.

Follow-up discipline. A huge fraction of appointments come from the second or third touch — the callback note, the door hanger left when nobody answered, the "I told you I'd swing back" return visit. Reps who only count first-knock conversions undersell the whole motion.

Belief. Unquantifiable but real. A rep knocking a list they trust knocks differently than a rep who thinks they're wasting their time. This is the quiet reason targeting changes close rates beyond the raw math: reps work a curated list harder because they believe in it.

The opener, broken down

Since the first ten seconds carry so much of the ratio, it's worth dissecting what a high-setting opener actually does. It is not a pitch. It is a permission request anchored to a true, specific, verifiable reason. Compare two openers on the same worn roof:

"Hi, I'm with ABC Roofing — do you need a new roof?"

This hands the homeowner a yes/no, and the reflexive answer at a stranger's door is no. Dead.

"Hey, sorry to bug you — I'm Mike with ABC Roofing, I'm working a couple of the older roofs on this street. I noticed yours has some granule loss running into the gutters and the south slope's taken a lot of sun. I'm not selling anything right now, I just wanted to point it out — mind if I show you what I mean from the driveway?"

That opener names the rep, gives a true local reason, references something the homeowner can verify with their own eyes, defuses the sales fear ("not selling anything right now"), and asks a small yes ("from the driveway") instead of a big one. It converts to a conversation far more often, and a conversation is where appointments live.

Notice the opener leaned on a real observation about an old, worn roof. A rep working a targeted list of aged roofs can use this opener honestly on most doors. A rep on a random street is bluffing half the time, the homeowner senses it, and the opener dies. The list and the opener reinforce each other.

The ask, broken down

The second place reps leak appointments is the transition from conversation to set. Weak reps end with "so, would you maybe want an estimate sometime?" — a soft, open question that invites a soft, open answer. Strong reps assume the inspection and offer a choice of times:

"Here's what I'd do — let me get up there, take some photos, and show you exactly what I'm seeing so you actually know what you're working with. I'm back in this area Thursday afternoon and Saturday morning. Which works better for you?"

That's an assumptive close with a two-option time choice and a clear value ("so you actually know what you're working with"). It produces real, dated appointments instead of "come back sometime" — which protects your sit rate downstream. The ask is where doors-to-appointment and appointment-to-sit are both won or lost in the same sentence.

What "good" looks like as a daily and weekly target

Owners want a number to manage to. Here are defensible daily and weekly targets for a full-time residential canvasser, stated as ranges by market type. Use these to set expectations and spot problems, not to whip people.

If a rep is way below the doors-per-shift band, you have an activity problem — they're not actually knocking. If doors are fine but contacts are low, you have a timing problem — they're knocking when nobody's home. If contacts are fine but appointments are low, you have a pitch problem — fix the opener and the ask. If appointments are fine but sits are low, you have a qualification problem — they're logging soft maybes. Each symptom points at a different fix. That diagnostic chain is worth more than any single benchmark number.

The math owners actually need: cost per appointment and cost per job

The ratio is a means, not an end. What you really manage is the cost of an appointment and the cost of a job, because that's what tells you whether door knocking beats your other acquisition channels. Build this simple model and update it monthly.

Start with fully loaded cost per knocking hour. If you pay a rep a base plus commission, take the base for the period, add payroll taxes and any guarantee, add gas and vehicle wear if you provide it, add the loaded cost of the canvassing manager's time spread across the team, and add any list or data cost. Divide by the rep's actual knocking hours (not clocked hours — subtract drive time, lunch, and standdowns).

A worked example, numbers chosen to be illustrative and easy to follow:

• Rep base for the month: 2,800
• Payroll taxes and burden (~12%): 336
• Gas and vehicle: 250
• Share of canvass manager and list/data: 600
• Total monthly cost to field this rep: 3,986
• Knocking hours that month: 100
• Fully loaded cost per knocking hour: ~40

Now layer the funnel. Say this rep, on a targeted list, sets 10 appointments that month. That's:

• Cost per appointment set: ~399
• At a 68% sit rate, ~6.8 sits → cost per sit: ~586
• At a 38% close on sits, ~2.6 jobs → cost per signed job: ~1,533

Now you can actually decide things. If your average residential job nets several thousand in gross profit, a 1,500-ish acquisition cost per signed job is healthy and door knocking earns its place in the mix. Run the same model on the cold-market funnel and the cost per job roughly doubles or worse — which is the entire financial argument for targeting.

Notice what this model does that a raw ratio can't: it lets you compare door knocking against paid leads, mailers, and digital on equal footing, in dollars per signed job. That's the comparison your bank account cares about.

Tying compensation to the right number

How you pay reps quietly shapes the ratio, because people optimize what you measure and reward. Three common structures, and what each does to behavior:

• Pay per appointment set. Fast to build pipeline, but it rewards quantity over quality and floods your inspectors with soft maybes that don't sit. If you use it, gate it on sat appointments, not set ones, or you'll pay for air.
• Pay per sit (kept inspection). Better. The rep only earns when a real homeowner kept a real appointment, which forces them to qualify hard at the door and confirm the set. Sit rates climb noticeably under this structure.
• Pay per signed job (commission). Aligns the rep with the business but has the longest feedback loop, which is brutal for green reps who quit before their first check. Most healthy operations blend a per-sit spiff with a back-end commission so reps eat something this week and chase the bigger number too.

The meta-point: if your comp rewards raw appointments, don't be surprised when your doors-to-appointment ratio looks fantastic and your sit rate is garbage. Pay for the stage that's closest to money you can actually keep.

A quick benchmark on rep ramp time

One number owners forget to model is how long a new rep takes to reach the targeted-list band. Realistic ramp for a coachable rep with daily field support is roughly two to four weeks to stop being a net drain and eight to twelve weeks to hit steady-state numbers. If you're cycling reps faster than that — and many shops churn green knockers inside 30 days — you are paying full freight to train people and firing them right before they'd become productive. The fix is rarely "hire better"; it's a tighter feedback loop and a list good enough that early wins come fast enough to keep the rep believing.

The hidden multiplier: which doors, not how you knock

Most canvassing advice obsesses over the pitch and ignores the list, which is backwards. You can squeeze maybe 20 to 40 percent more out of a rep with better scripting and coaching. You can multiply their output by improving which houses they stand in front of, because list quality moves the two hardest stages of the funnel — contacts-per-appointment and close rate — at the same time.

Think about what makes a door convert. The homeowner has to have a roof that's actually near the end of its life or visibly compromised, the budget or insurance pathway to do something about it, and the openness to talk. You can't control budget or openness from the curb. But you can control the first one — if you know which roofs are old or storm-worn before you knock.

This is where most teams leave enormous money on the table. They knock by geography ("do this subdivision") instead of by condition ("do the worn-out roofs in this subdivision"). A subdivision built in waves over fifteen years has roofs ranging from nearly new to overdue, scattered house by house. Knock it blind and most of your conversations are with people whose roofs are fine — dead doors that eat the shift. Knock only the aged and worn roofs and every conversation starts closer to a real need.

What you can and can't see from the street

Reps are taught to read roofs from the curb, and good ones get decent at it — spotting missing or lifted shingles, granule loss streaking the gutters, patched valleys, a sun-baked south face that's clearly older than the north. But curb-reading has hard limits. You can't see the back slope from the street. You can't tell a 12-year-old roof from a 19-year-old roof of the same color from 40 feet away. And you certainly can't pre-sort a whole neighborhood by age before you decide where to spend the shift.

This is the gap RoofPredict was built to close. It takes aerial imagery plus weather history and gives a roofing contractor, house by house, a roof-age range (not an exact install date — nobody can give you that honestly from the air) along with the storm history that roof has actually taken: the hail and wind events modeled per roof, rather than a flat "it hailed somewhere in this ZIP." The output is a ranked view of a neighborhood — which roofs are old enough to be due, which ones a storm likely wore out — so your reps spend the shift on the doors most likely to have a real problem and skip the obviously-new roofs that waste a knock.

Be clear-eyed about what that does and doesn't do. It does not promise a roof is bad — a roof-age range is a range and a storm forecast is odds, not proof; your rep still has to inspect and your estimator still has to verify. It does not replace the pitch, the relationship, or the close. What it changes is the denominator: instead of 80 random doors with maybe 15 worn roofs hiding in them, your rep knocks 80 doors that are already weighted toward old and storm-worn. That's the move that turned one job per four shifts into the targeted-list math above. The pitch stayed the same; the list got smart. You can also feed the same age-and-storm signal into your existing CRM or mailing list, so the homes you mail and the old estimates you re-work are sorted the same way — it sharpens the outbound you already do rather than handing you yet another list to buy.

The honest limit, stated plainly: targeting raises your odds, it doesn't remove the work. You'll still knock roofs that turn out fine and miss roofs that turn out shot. But across a season, weighting every shift toward worn-out roofs is the single biggest lever on the doors-to-appointment ratio that doesn't depend on your reps getting better — and your reps get better anyway, because they trust a list that keeps paying off.

How to fold targeting into a route without slowing reps down

The practical worry owners raise is that a targeted route means reps zig-zagging across a neighborhood, skipping houses, and wasting walk time. It's a fair concern and it's solvable. A few field rules:

• Cluster, don't scatter. Work targeted blocks, not single targeted houses spread across a square mile. A street that's mostly aged roofs gets worked door to door; a street that's mostly new gets skipped entirely. You're choosing streets and pockets weighted toward old roofs, then knocking those pockets thoroughly.
• Knock the new ones if you're already there. If a worn roof sits between two new ones on the same walk, knock all three — the walk cost is already paid. Targeting decides where you start and how you allocate scarce evening hours, not a rigid skip-list that adds steps.
• Use the data to rank, then let the rep run. The point of a ranked, age-and-storm-scored view is to hand the rep a route that's already weighted before they leave the truck. Decision fatigue at the curb is a real time-sink; pre-deciding the route removes it.

Done right, targeting reduces wasted walking, because reps stop pacing through five-year-old developments hunting for the rare old roof and instead spend their hours where old roofs are dense.

What targeting does to canvasser retention

There's a second-order effect worth naming because it shows up in your hiring budget, not your ratio sheet. Green canvassers quit when the math feels hopeless — when they knock all afternoon, talk to almost no one with a real problem, and go home with nothing. Put that same green rep on a list weighted toward worn roofs and their early conversations actually go somewhere. They set an appointment in week one, it sits, it closes, they get a check, and they stay. Retention isn't a soft benefit here; replacing a canvasser costs you the recruiting time, the training shifts, and the productive weeks you'll never get because they quit at week three. A list that produces early wins is a retention tool disguised as a targeting tool.

Storm markets: real ratios and the legal line you don't cross

Fresh storm events are the best ratios in the business and the easiest place to get yourself in trouble. The ratios are real — 12 to 30 doors per appointment when an event is recent because the whole neighborhood is primed. But the way many crews chase those ratios crosses a bright legal line, and it's worth being precise so you capture the demand without becoming the cautionary tale at the next licensing-board meeting.

Here's the clean version of what a roofing contractor may do at a storm door. You may knock, you may inspect the roof, you may document damage thoroughly with dated photos and notes, and you may prepare an accurate, line-item repair estimate — ideally aligned to the same estimating standards adjusters use — for the work you would do. You hand that documentation and estimate to the homeowner. The homeowner files their own claim. The insurer decides coverage. You can state facts about your own scope to the carrier if asked.

Here's what crosses the line into unlicensed public adjusting in most states, and you do not do it no matter how good it would be for your ratio. You do not, for a fee, negotiate or "handle" the homeowner's claim. You do not interpret their policy or tell them what's covered. You do not promise a specific payout or that the claim will be approved. You do not promise the deductible will be waived, absorbed, or "taken care of" — in many states that's outright illegal, and it's a fast way to lose your license. And you do not advertise a "free roof." The homeowner files and the insurer decides; your job is to document and estimate.

This isn't just compliance theater — the safe frame actually converts better and protects your sit rate. A rep who shows up with thorough photos and a clean, defensible estimate, and who tells the homeowner "here's the documentation, you file, the insurer decides," builds more trust than the cowboy promising a free roof and a waived deductible. The cowboy sets more soft appointments and far fewer real ones, and a chunk of his "wins" evaporate at the sit when reality lands. Document well, estimate honestly, hand it over, and let the ratio come from competence instead of promises you can't legally keep.

On the targeting side, storm markets are also where per-roof storm modeling earns its keep. A blanket hail map tells you a ZIP got hit; it doesn't tell you which roofs on which streets actually caught the worst of it. Knowing, house by house, which roofs likely caught the damaging hail or wind — paired with which of those roofs were already old — lets your reps work the doors most likely to have real, documentable damage, instead of canvassing the whole impact zone and burning the shift on roofs the storm barely grazed.

A repeatable weekly system that actually moves the ratio

Benchmarks are useless without a loop that improves them. Here's a concrete weekly operating rhythm that a small-to-midsize roofing team can run without a full-time analyst. The goal is a tight feedback cycle: knock, measure, diagnose, adjust, repeat.

1. Set the week's list before Monday. Don't let reps wander. Each rep starts the week with a defined set of streets or, better, a defined set of worn-and-aged roofs pulled from your targeting data and seeded into their route. Pre-deciding the doors removes the single biggest time leak in canvassing: reps standing on a corner deciding where to go.

2. Track the four numbers per rep per shift. Doors, contacts, appointments set, and (next day) sits. A simple shared sheet or your CRM does this. If you can only track one extra thing beyond doors and appointments, track contacts — it's the most diagnostic number you're probably missing.

3. Hold a 15-minute daily debrief. Not a lecture — a number review and one coaching point. "You knocked 70, talked to 11, set 1. Your contact rate's low — you were out at 1 p.m., let's push you to 4 p.m. tomorrow." Specific, fast, repeatable.

4. Field-ride the outliers weekly. Spend one shift a week shoulder-to-shoulder with your weakest rep and one with your strongest. You'll learn more about your real ratio in two ride-alongs than in a month of spreadsheets. Steal the strong rep's opener and teach it to everyone.

5. Run a follow-up sweep. Every no-answer door and every "come back later" gets a second touch within the week — a door hanger, a callback, a return visit. A meaningful share of your appointments live in the second touch. Reps who only knock once leave them on the table.

6. Review cost per appointment monthly. Roll up the funnel and the dollars (the model above), compare door knocking to your other channels, and reallocate. If targeted doors are beating mailers on cost per job, shift budget toward more knocking hours on better lists.

This loop is boring and it works. Most teams that "can't get canvassing to work" simply never close the measurement loop — they knock, they don't track contacts, they can't diagnose, so they tweak the pitch randomly and conclude door knocking is dead. Door knocking isn't dead. Untracked, untargeted door knocking is dead.

Reading the numbers like a diagnostician

Once the four numbers are flowing, the skill that separates a sharp sales manager from a busy one is reading them as a diagnosis instead of a scoreboard. Walk the funnel top to bottom and the failing stage tells you the fix:

That table is the whole point of measuring contacts and sits separately. Without them, every problem looks like "the pitch," and you'll spend months rewriting scripts when the real issue was that a rep was knocking empty houses at 1 p.m. or working a subdivision full of five-year-old roofs.

Don't confuse a slump with a trend

A caution on small samples. A single rep's single shift is almost pure noise — one chatty homeowner or one rainy afternoon swings it wildly. Don't coach hard off one shift. Look at rolling weekly numbers per rep and rolling monthly numbers per neighborhood. The signal you can trust shows up across roughly 200-plus doors or a couple of weeks, not across one bad Tuesday. Owners who react to daily noise create whiplash; reps stop trusting the feedback and start gaming the count. Steady weekly review beats frantic daily reaction.

A canvasser pre-shift and post-shift checklist

Hand this to every rep. It bakes the system into the daily motion.

Before the shift:

• Pull today's route — worn-and-aged roofs first, skip the obviously-new ones.
• Confirm you're knocking the contact window (late afternoon/early evening weekdays).
• Phone charged, door hangers stocked, pitch and one local-proof line ready.
• Know your one specific, true reason for being on the street.
• Set a target: doors, contacts, appointments — write them down.

During the shift:

• Count every real door; don't inflate by counting skipped houses.
• Log contacts separately from doors.
• Ask permission before pitching; lead with the verifiable local reason.
• Disqualify dead doors fast; pour energy into live ones.
• Every no-answer gets a hanger; every "later" gets a callback note.
• Only log a real appointment — name, address, date, time. No soft maybes.

After the shift:

• Enter doors, contacts, appointments before you go home.
• Note your best and worst door and why — one line each.
• Flag any roof that needs a documented inspection.
• Bring one question to tomorrow's debrief.

Common mistakes that wreck the benchmark

A short field guide to the ways teams fool themselves.

Counting no-contacts as doors-without-conversions. This makes your contact rate invisible and your pitch look broken when the real issue is timing. Separate doors from contacts or you're flying blind.

Logging soft maybes as appointments. Inflates the set number, craters the sit rate, and hides the real problem. A non-scheduled "maybe" is a follow-up, not a set.

Knocking by geography instead of condition. The most expensive habit in canvassing. Random subdivisions are mostly fine roofs. Target the worn and aged ones and the whole funnel tightens.

Coaching the pitch while ignoring the list. You can wring 20–40% from scripting. You can multiply output by fixing which doors get knocked. Owners spend 90% of their attention on the smaller lever.

Chasing storm ratios with illegal promises. Waived-deductible and free-roof pitches set soft appointments that die at the sit and risk your license. Document, estimate, hand it over — let the homeowner file and the insurer decide.

Knocking once and quitting the door. A large share of appointments live in the second and third touch. No follow-up sweep means leaving real money in the no-answer pile.

Measuring nothing past doors. If the only numbers you have are doors knocked and jobs closed, you can't diagnose anything in between. Track the whole chain: doors, contacts, appointments, sits, jobs.

Putting a real number on your own team

Stop benchmarking against the internet and benchmark against yourself. For the next two weeks, have every rep track the four numbers honestly. At the end, you'll have your doors-per-appointment ratio, your contact rate by time of day, your sit rate, and your cost per signed job. That's the only benchmark that can run your business. The published ranges in the table above are a sanity check — if you're wildly outside them, you now know exactly which stage to investigate.

Then pull the biggest lever. For most teams, it isn't the pitch — it's the list. Weight every shift toward the roofs that are actually old or storm-worn, and the same reps with the same script post a meaningfully tighter ratio because more of their conversations start with a real problem. That's the role RoofPredict plays: a house-by-house read on which roofs are due — a roof-age range plus the storms each roof has actually taken — so your team knocks the doors most likely to convert and skips the ones that waste a knock, and so the homes you mail and the old estimates you re-work get sorted the same smart way. It won't knock the doors for you, and it won't promise a roof is bad. It just makes sure your reps are standing in front of the right houses. If you want to see what a ranked, age-and-storm-scored view of your own area looks like, that's what a demo walks through.

Knock the right doors, count honestly, close the measurement loop, and the benchmark takes care of itself.

FAQ

What is a good doors-to-appointment ratio for roofing canvassing?

It depends heavily on market type. In a cold, untargeted, non-storm neighborhood, 40 to 100 doors per appointment set is normal. On a targeted list of worn and aging roofs, it tightens to roughly 25 to 55 doors per appointment. In a fresh storm market, 12 to 30 doors per appointment is achievable while the event is recent. Anyone quoting one number for all situations is oversimplifying.

Why does my ratio look so much worse than the '10 percent' figure I see online?

Because that figure usually counts no-contacts loosely and logs soft maybes as appointments. A 10 percent set rate off live conversations is plausible; a 10 percent set rate off raw doors in a calm market almost never is. Track doors, contacts, and appointments separately and your real numbers will line up with the honest ranges, not the marketing ones.

What's the difference between a door, a contact, and an appointment?

A door is a house your rep physically approached. A contact is a live human who actually talked with them. An appointment is a scheduled inspection with a name, address, date, and time. Most teams only track doors and appointments and miss contacts entirely, which is exactly the number that diagnoses whether a low set rate is a timing problem or a pitch problem.

How many doors should a rep knock in a shift?

A full-time canvasser typically gets through 60 to 100 doors in a focused four-hour shift once you subtract walking, no-answers, and long conversations. If a rep is far below that, you have an activity problem. If doors are fine but contacts are low, they're knocking when nobody's home — push them into the late-afternoon contact window.

Does targeting old roofs really change the ratio that much?

Yes, because it moves two of the hardest funnel stages at once — contacts-per-appointment and close rate — without changing the pitch. The same rep knocking a list weighted toward worn and aging roofs has more conversations that start with a real problem, so more convert. In worked examples it commonly produces roughly two to two-and-a-half times the output of blind, geography-based knocking.

Can I see roof age from the street to target better?

Partly. Good reps read missing or lifted shingles, granule loss in the gutters, and a sun-baked older slope. But you can't see the back slope from the curb, you can't reliably tell a 12-year roof from a 19-year roof of the same color, and you can't pre-sort a whole neighborhood before the shift. Aerial-plus-weather data gives a per-house roof-age range and storm history so you can weight the route before anyone knocks.

What's a realistic sit rate, and why do my appointments not show up?

A healthy sit rate runs 50 to 70 percent in cold markets and higher when appointments are real and recent. The most common reason sits crater is reps logging soft maybes as appointments. A real set has a name, address, date, and time. If it doesn't, it's a follow-up, not an appointment, and counting it as one hides the actual problem.

How do I calculate cost per appointment and cost per job?

Take the rep's fully loaded monthly cost (base, payroll burden, gas/vehicle, a share of the canvass manager and any list cost) and divide by actual knocking hours for cost per hour. Then divide total cost by appointments set, by sits, and by signed jobs. Cost per job is what lets you compare door knocking against mailers and paid leads in dollars on equal footing.

What can I legally say at the door in a storm market?

You can knock, inspect, document damage with dated photos, and prepare an accurate line-item repair estimate for your own scope, then hand it to the homeowner who files their own claim. You cannot negotiate or handle the claim for a fee, interpret their policy, promise a payout or approval, promise the deductible is waived, or advertise a free roof — those cross into unlicensed public adjusting in most states.

How often should I review canvassing numbers?

Review the four core numbers daily in a 15-minute debrief so you can adjust timing and pitch fast, and roll up cost per appointment and cost per job monthly to compare channels and reallocate budget. The teams that say canvassing doesn't work almost always never closed this measurement loop — they knocked, didn't track contacts, and tweaked randomly.