Why Is My Roofing Set Rate So Low? The Real Causes and How to Fix Them

You ran the numbers at the end of the week and the math is ugly. Your crew knocked four hundred doors, walked twenty-something roofs, and you signed three contingencies. Somewhere a sales trainer told you a good rep should be setting one inspection for every ten or twelve doors and closing half of what they inspect. You're not close. The natural reaction is to blame the pitch — buy a new script, run another role-play night, fire the guy who keeps coming back empty. Most of the time that's the wrong move, because the pitch is rarely the part that's actually broken.

Set rate is one of those numbers that everybody quotes and almost nobody defines the same way. Before you can fix a low one you have to know exactly what you're measuring, where in the funnel it's leaking, and whether the leak is even something a rep controls. A rep can be saying every word right and still post a terrible set rate because he's standing in front of the wrong houses. A rep can have a beautiful pitch and a clean set rate and still produce nothing, because the inspections aren't turning into signed jobs. Those are three different problems wearing the same costume.

What follows is the way an operator who has actually run boards and ridden along on routes thinks about it: how to define the metric so the number means something, the eight reasons a set rate craters (in rough order of how often they're the real culprit), and a rebuild sequence you can run starting Monday. There are worked examples, benchmark ranges, scripts you can steal, and the edge cases that trip up even experienced managers. No magic, no fluff.

First, define what you're actually measuring

Half of all "my set rate is low" conversations are really "I'm measuring two different things and comparing them." Get this straight before anything else.

In door-to-door and storm roofing, the funnel has roughly five stages, and "set rate" can mean the conversion between almost any two of them depending on who's talking:

1. Doors knocked — a person physically raps on, or rings, a door.
2. Contacts — a human answers and you get to say more than "sorry to bother you."
3. Inspections set — the homeowner agrees to let you get on or look at the roof, now or scheduled.
4. Inspections completed — you actually get up there and document it.
5. Signed agreement — a contingency, a retail contract, or a scheduled job depending on your model.

The word "set" technically points at stage 3: you set an inspection or an appointment. But in the field people sling the term around to mean inspection-to-sign, contact-to-inspection, even knock-to-sign. So the first fix costs you nothing: pick one denominator and one numerator and write them on the whiteboard so the whole team is grading the same test.

The two ratios that matter most, kept separate:

• Set rate = inspections set ÷ doors knocked (or ÷ contacts, if you want a cleaner read on the pitch — more on that below).
• Close rate (or sign rate) = signed agreements ÷ inspections completed.

Keep them apart, because they fail for opposite reasons. A low set rate is usually a top-of-funnel problem: wrong houses, wrong timing, weak opener, or a rep who can't get past the door. A low close rate is a bottom-of-funnel problem: weak inspection process, no urgency, no clear next step, price objection, or you're inspecting roofs that were never going to sell.

Set rate off doors vs. off contacts

Here's the distinction that quietly wrecks most diagnoses. If you measure set rate off raw doors, you're blending two things together: how often someone's home and willing to talk (a function of timing and area), and how good the pitch is once they do talk. A rep working a neighborhood where nobody answers at 2 p.m. on a Tuesday looks like a bad closer when he's really just a victim of the clock.

So track both:

• Contact rate = contacts ÷ doors knocked. This is mostly about when and where you knock, not the rep's mouth.
• Set-off-contact rate = inspections set ÷ contacts. This is the cleanest measure of the actual pitch.

When your overall set rate is low, splitting it this way tells you which lever to pull in about ninety seconds. Low contact rate, healthy set-off-contact? Stop coaching the script and fix your knock times and your route. Healthy contact rate, low set-off-contact? Now you've got a real pitch or rapport problem, and a role-play night might actually help.

What "good" looks like

Benchmarks vary wildly by market, weather, season, and whether you're working a fresh storm or cold retail, so treat these as orientation, not law. Rough field ranges that experienced canvass managers will recognize:

If you're below the "weak" column on set-off-doors but your set-off-contacts is fine, your problem is contact, not closing. That single split kills a huge share of misdiagnosed "low set rate" complaints. Don't anchor too hard on the exact percentages — anchor on the shape: figure out which stage is dragging the chain.

The eight real reasons your set rate is low

Ranked, loosely, by how often each one turns out to be the actual cause when you dig in. Most shops have two or three of these stacked on top of each other.

1. You're knocking the wrong houses

This is the single most common hidden cause, and it's invisible from the porch. A rep can do everything right and post a 2% set rate because the street he's working is full of roofs that don't need him — ten-year-old architectural shingle that took a glancing storm, new construction, recent re-roofs, rentals where nobody can say yes. He's not failing to set; there's nothing to set.

The tell: your set-off-contact rate is mediocre and it's mediocre everywhere, regardless of who's knocking. When even your best rep can't move the needle on a given area, the area is the problem. Good people on bad doors produce bad numbers and quit.

The roofs worth a knock are the ones that are actually due — old enough that replacement is on the table, or worn by a real storm. The trouble is you can't see roof age from the curb, and you definitely can't see it from a list of addresses. County records and Zillow show year built, not year roofed; a re-roof is invisible to them, so a 1985 house with a five-year-old roof and a 2015 house with a beat-up builder-grade roof both look the same on paper. You end up knocking on a wall of roofs that look identical and aren't.

This is the leak that no script night will ever fix, because it's a targeting problem masquerading as a sales problem. We'll come back to how to attack it specifically.

2. Your contact rate is killing you and you blame the pitch

If only one in six doors produces a live conversation, your set-off-doors number is going to look terrible no matter how good anyone is. Contact rate is mostly mechanical:

• Time of day. Knocking 1–4 p.m. on weekdays is the graveyard shift — working households are gone, retirees nap. The money window for most residential areas is roughly 4:30–8 p.m. weekdays (until you lose daylight) and 10 a.m.–2 p.m. on Saturdays.
• Day of week. Sunday mornings, holiday weekends, and the dinner-prep crunch around 6 p.m. are weak. Saturday late morning is gold in most markets.
• Density and walkability. Reps lose half their day walking between far-apart rural doors. Tighter routes mean more knocks per hour, which means more contacts even at a flat contact rate.
• Knock approach. Standing dead-center in the peephole, sunglasses on, clipboard up like a process server, truck idling at the curb — people just don't answer. Stand slightly off to the side, hat off, hands visible, vehicle parked normally.

Fix the clock and the route before you touch the script. A team that shifts its knocking window can lift contacts thirty to fifty percent without anyone changing a word they say.

Run the arithmetic on it, because the leverage is bigger than it feels. Say a rep knocks 25 doors an hour at an 18% contact rate and sets 25% of contacts. That's about 4.5 contacts and a bit over one set per hour. Now suppose moving to the evening window pushes contact rate to 30% with no other change. Same 25 knocks an hour, but now 7.5 contacts and nearly two sets per hour — roughly a 65% jump in sets from one scheduling decision. You did not hire anyone, train anyone, or change a single word of the pitch. Reps who feel "slow" are frequently just stuck in the dead window, grinding away at empty houses and concluding they're bad at the job.

The re-knock discipline compounds with the window. Most reps treat a not-home as a permanent loss, but a not-home at 2 p.m. is often just a working homeowner who'll answer at 6. Log every not-home with the time you knocked, and route a deliberate second pass through those addresses in the money window. In a lot of shops the recovered not-homes are worth more than any pitch tweak, because the contact was never the problem — presence was.

3. Your opener triggers the reflex “no”

The first seven seconds decide most doors. Homeowners run an autopilot rejection for anything that pattern-matches to “salesperson,” and a lot of roofing openers walk right into it. “Hi, we're doing roofs in the neighborhood, do you have a minute?” is a yes/no question whose easy answer is no, and you handed it to them.

What triggers the reflex:

• Asking permission to talk (“got a sec?” / “mind if I…?”) — invites a no.
• Leading with your company instead of their roof.
• Vague “storm in the area” openers that sound like every other knocker who hit them this month.
• Over-rehearsed, fast, salesy cadence — it sounds like a script because it is one.

What works is specific, local, and about their house, delivered calm and a little under-eager. Something closer to: “Hey, I'm with [Company] — we've been on a few roofs over on [nearby street]. I noticed yours has some [granule loss / aging at the ridge / the kind of wear we've been seeing from the [month] storm]. I'm not here to sell you anything today, just wanted to flag it — want me to take a quick look while I'm here?”

The mechanics that make that land: it references a real nearby street (proof you actually work here), it names a specific observation about their roof, it lowers the temperature (“not here to sell you anything today”), and it ends on a low-commitment ask (“a quick look”, not “an inspection,” not “sign here”). You can't say “I noticed yours has aging at the ridge” unless you actually have a reason to believe it — which loops straight back to targeting.

There's a body-language layer most managers never coach. The reflex "no" fires off appearance before a word is spoken. Approach at a relaxed pace, not a march. Stop a full step back from where a salesperson would plant — give the homeowner room. Hat or sunglasses off so they can read your eyes. Hands visible and low, not clutching a clipboard up like a shield. Turn your body slightly toward the street, as if you might leave at any second; people relax around someone who clearly isn't going to camp on their porch. Let your first sentence be a little slower and quieter than feels natural — urgency in the voice reads as a pitch, and calm reads as a neighbor. Reps who fix the words but keep the process-server energy still get the door closed on them.

And handle the early objection without a fight. When the homeowner says "I'm not interested" before you've even finished — which is reflex, not a decision — the worst move is to argue. The move that keeps the door open is to agree and lower the stakes again: "Totally fair, I'm not trying to sell you a roof. I'm only flagging it because I'm already up the street — thirty seconds and I'll tell you if it's actually something or nothing." You're not overcoming the objection; you're making the next step so small that the no no longer applies to it. That reframe alone recovers a meaningful slice of doors that look dead in the first three seconds.

4. You're selling the inspection like it's the close

The ask at the door should be tiny. The job of the door is to earn a ladder, nothing more. Reps with low set rates routinely try to win the whole deal on the porch — they get into shingle brands, financing, “we can probably get this covered,” before they've even seen the roof. That's a mistake on two fronts: it raises the commitment the homeowner has to make to say yes (so fewer say yes — your set rate drops), and it makes promises you can't keep before you have any facts.

The set-rate fix is to shrink the ask. “Can I take a quick look” sets more inspections than “can I schedule a full inspection and estimate.” Same activity, smaller word, higher yes. Save the real conversation for when you're standing on the roof with photos.

5. No reason to act now, so they “think about it”

A homeowner who agrees to a look but then stalls forever is a set that never converts, and it drags your effective numbers down. The cause is almost always that you gave them no honest reason the timing matters. Note the word honest. There are urgency drivers you can use and ones that will get you in real trouble.

Legitimate, truthful urgency:

• A roof at the end of its service life keeps losing granules and protection every season; documented wear that's getting worse is a real reason to look now rather than after the next storm.
• If there was a recent windstorm or hail event, the physical evidence on the roof is freshest now and degrades — again, a fact, not a threat.
• Your schedule genuinely fills up after storms; “I've got two slots this week” is fine if it's true.

What you may not do is manufacture urgency around insurance or money. Don't promise the homeowner their claim will be approved, don't tell them the deductible will be waived or absorbed, don't advertise a “free roof,” and don't tell them you'll “handle the insurance company” for them. Those aren't just sleazy — in most states, negotiating or adjusting a homeowner's claim for them is unlicensed public adjusting, and a contractor doing it for compensation is breaking the law. We'll cover the compliant version of storm urgency in its own section because it trips up so many storm-restoration sellers.

6. The rep can't get on the roof, or won't

Sometimes “low set rate” is literally a logistics problem. The rep doesn't have a ladder on the truck, isn't comfortable on steep pitch, or the inspection requires a callback that never gets scheduled. A “set” that requires the homeowner to be home for a separate appointment converts far worse than one you do on the spot. Every handoff is a leak.

Wherever it's safe and legal, the highest-converting model is knock → look now → show photos now → next step now. The more you can collapse into one visit, the less your funnel leaks between stages. If your model forces a scheduled callback, your scheduling discipline (confirmation texts, tight windows, reminders) becomes part of your set rate whether you think of it that way or not.

7. Your reps aren't tracking, so nobody can coach

You cannot fix what you don't count. Shops with low set rates very often have no per-rep, per-stage data at all — just a vague sense that “Tuesday was slow.” Without knock counts, contact counts, and set counts by rep, by area, and by time block, every coaching conversation is a guess. You can't tell the rep with a contact problem from the rep with a pitch problem from the rep working a dead area, so you give all three the same useless advice.

The minimum viable tracking: every rep logs doors, contacts, sets, and the area/time block, every day. A clipboard tally works; a phone form works better. The point is that within a week you can see the shape of each rep's funnel and coach the actual leak.

8. Burnout, churn, and the green-rep doom loop

Set rate and rep tenure are tied at the hip. A new canvasser knocking cold doors with no specific reason to be at any given house gets rejected forty times before lunch, posts a 1% set rate, makes no money, and quits inside three weeks. Then you hire another one and run the same loop. The whole team's average set rate is dragged down by a rotating cast of people who never make it past the worst part of the learning curve.

The two things that break the loop are (a) giving green reps a specific, true reason to be at each door so they sound like a veteran without ten years of pattern recognition, and (b) putting them in front of houses that can actually say yes so they get early wins. A rep who closes something in week one stays. A rep who eats forty noes a day for three weeks does not. Targeting isn't just a numbers lever; it's a retention lever, and retention is a set-rate lever, because experienced reps set far better than green ones.

Diagnose your own funnel in 30 minutes

Stop guessing. Pull last week's activity — even rough numbers — and compute four things per rep:

1. Contact rate = contacts ÷ doors.
2. Set off contacts = sets ÷ contacts.
3. Set off doors = sets ÷ doors.
4. Close rate = signed ÷ inspections completed.

Then read the pattern:

A worked example

Two reps, same week, same nominal “set rate” of about 4% off doors. Looks like the same problem. It isn't.

Rep A: 300 doors, 90 contacts, 12 sets. Contact rate 30% (fine). Set off contacts 13% (soft). Rep A is getting conversations and losing them — his opener or porch ask needs work. Coach the pitch.

Rep B: 300 doors, 36 contacts, 12 sets. Contact rate 12% (terrible). Set off contacts 33% (excellent). Rep B is a strong closer being strangled by a contact problem — wrong times, bad route, or a dead area. Coaching his pitch would be a waste; his pitch is the best thing he's got. Fix his schedule and his streets and his raw set rate could double without him changing a thing.

If you'd only looked at the 4% set-off-doors, you'd have given both reps the same advice and helped neither. The split is the whole game.

Take it one layer deeper with a third rep to see the bottom-of-funnel trap. Rep C: 300 doors, 90 contacts, 24 sets — a 27% set off contacts, double Rep A, looks like the best canvasser on the board. But of 24 inspections completed, only 4 signed: a 17% close rate. Rep C is setting inspections on anything that holds still, including roofs that were never replaceable, and his pile of "sets" is mostly tire-kickers and good roofs. His set rate is a vanity number. The cure for Rep C is the opposite of Rep A's: not a better pitch, but better qualification at the door and on the roof, plus better doors to begin with. If you'd paid Rep C on raw sets you'd have rewarded exactly the behavior bleeding your close rate. This is why you never read set rate alone — you read it next to the close rate it feeds.

The rebuild sequence

Work these in order. Each one is cheap or free, and each compounds with the next.

Step 1 — Fix the clock and the route (week 1)

Move your primary knocking window to 4:30–8 p.m. weekdays and 10 a.m.–2 p.m. Saturdays. Tighten routes so reps spend less time walking between doors. Track contact rate before and after. This is the fastest, highest-ROI change you can make and it costs nothing. Expect contacts up, which lifts raw set rate even with no other change.

Step 2 — Rebuild the opener (week 1–2)

Kill the permission ask and the company-first opener. Standardize on a structure, not a word-for-word script:

1. Local proof: “We've been on a couple roofs over on [real nearby street].”
2. Specific observation about their roof: “Noticed yours has some [specific wear].”
3. Lower the temperature: “Not here to sell you anything today.”
4. Tiny ask: “Want me to take a quick look while I'm up here?”

Role-play it until it sounds relaxed, not recited. Measure set off contacts before and after. A real opener change moves this number within days.

Step 3 — Shrink the ask, then run a tight inspection (week 2–3)

The door earns a ladder; nothing more. On the roof, document everything (we'll detail the photo workflow below), then make the next step concrete and small: “Here's what I found, here are the photos — the next step is X.” No floating “think about it.” Give an honest, specific reason the timing matters.

Step 4 — Fix the doors themselves (week 2–4)

This is the one most shops skip because it feels like it's outside the rep's control — and it is, which is exactly why it's management's job. Stop sending crews down whole streets where most roofs don't need you. Concentrate knocks on roofs that are actually due by age or storm wear. This is where the next section lives.

Step 5 — Instrument and coach weekly (ongoing)

Per-rep, per-stage, per-area numbers every week. Coach the specific leak the data shows. Celebrate set-rate improvement alongside signed jobs, so reps optimize the behavior instead of getting lucky.

Targeting: the lever almost nobody pulls

Everything above is real and worth doing, but notice how many of the eight causes trace back to one root: your people are standing in front of roofs that were never going to convert. You can have a perfect clock, a perfect opener, a perfect inspection, and still post a low set rate if half the doors on the route are new roofs, recent re-roofs, or roofs that simply aren't worn enough to replace.

The reason this leak persists is that roof age is genuinely hard to see. From the street, a fifteen-year-old roof and a five-year-old roof can look nearly identical, especially on the same builder-grade shingle. County and Zillow data give you year built, which is worthless the moment a house has been re-roofed — and you have no way of knowing which ones have. So most teams default to brute force: knock everything, let the law of averages sort it out. That “average” is your low set rate.

The better play is to know, before anyone leaves the truck, which roofs on a street are old enough to be due and which ones recently took a real storm — and to send your crew to those doors first.

Where RoofPredict fits

This is the targeting gap RoofPredict is built for. It reads aerial imagery to estimate a roof's age as a range — not an exact date, a range like 18–22 years — and it models storm physics per individual roof, scoring the hail and wind a specific house actually took rather than just whether a storm passed through the ZIP. A hail map tells you where it hailed; modeling the storm on each roof tells you which roofs the storm likely wore out. Pair the age range with the storm score and you get a ranked view of a street: these doors are due, those aren't, skip the new ones.

What that does to a set rate is direct. If your reps stop knocking new and recently re-roofed houses and concentrate on the ones that are aging out or storm-worn, the same pitch lands on far more receptive doors. The opener gets sharper too, because now the rep has a real, specific reason to be at that house — which is exactly the “specific observation” ingredient that makes openers work and that green reps can't fake on their own. It's also a retention lever: a new canvasser pointed at due roofs gets early wins, makes money, and stays, instead of eating forty cold noes a day and quitting.

It also turns your own book into doors. If you've got an old CRM full of past estimates and customers, enriching that list with roof-age and storm signals tells you which of your old contacts are now due — money already sitting in your book.

The honest limits, because a tight trade compares notes: roof age comes back as a range, not a birthday, and a storm score is odds the roof was hit hard, not proof. It won't measure the roof for you or tell you the shingle brand — that's a different category of tool. And it won't sell the job; your rep still has to knock, look, document, and earn the signature. What it does is make sure that knock lands on a door that can say yes. That's the part of the set-rate equation no script touches. If you want to see whether the ranked-street view matches roofs you already know, you can book a demo and hand over a street you understand — you decide whether it nailed it.

The storm-restoration version of all this (and the legal line)

Storm sellers have an extra set of set-rate dynamics, and an extra set of ways to get themselves in trouble. Worth its own treatment because the rules here are not optional.

Why storm set rates swing so hard

After a real hail or wind event, set rates spike — homeowners are primed, the damage is fresh, and the door is warm. Then two things drag them back down. First, the swarm: every contractor in three states descends on the same neighborhoods, so by week two homeowners have been knocked twenty times and the reflex “no” is fully armed. Second, targeting gets sloppy in the rush — reps knock entire affected ZIPs when only some roofs in that ZIP actually took damaging impact, because hail falls in streaks and wind funnels by exposure. The houses that got grazed waste your reps' time and your homeowners' patience.

This is exactly where per-roof storm modeling earns its keep over a flat hail map. Knowing which specific roofs in the affected area likely took damaging hail or wind — versus which ones were on the edge of the streak — lets you concentrate the crew on the doors most likely to have a real, documentable problem. Higher set rate, fewer wasted knocks, and homeowners who aren't being pitched on roofs that are fine.

The compliant urgency frame

Storm urgency is real, but you have to deliver it on the right side of a bright line. Here is what you may do, and what you may absolutely not.

You MAY:

• Inspect the roof and document what you find thoroughly, with dated photos.
• Write an accurate repair estimate aligned to standard estimating (Xactimate-style line items) for the work you would do.
• Explain to the homeowner what you observed — facts about the roof's condition and the wear or storm evidence you photographed.
• Hand that documentation and estimate to the homeowner so they can decide what to do with it.
• State that storm evidence is freshest now and degrades over time — a true fact — as a reason to document sooner.

You MAY NOT (this is the do-not-say list — teach it to every rep):

• Promise the claim will be approved or covered, or that a specific dollar amount is coming.
• Tell the homeowner the deductible will be waived, absorbed, eaten, or “taken care of.” Offering to do so is illegal in most states and is the fastest way to lose your license and your reputation.
• Advertise or imply a “free roof.”
• Say you'll “handle,” “negotiate,” “fight,” or “deal with” the insurance company on the homeowner's behalf. For a fee, that's unlicensed public adjusting in most jurisdictions.
• Interpret the homeowner's policy or coverage for them.
• Represent the homeowner against their insurer.

The safe frame is clean and it still sells: document thoroughly, write an accurate estimate, hand it to the homeowner; the homeowner files the claim, and the insurer decides coverage. Your value is the quality and speed of the documentation and the estimate, not a promise about an outcome you don't control. Reps who lean on “we'll get you a free roof” post great set rates for about a month and then get the company sued or fined. Reps who lead with “let me get up there, document exactly what's going on, and put together a clear estimate you can use” build a business that survives the audit.

The inspection-and-documentation workflow

A strong set rate dies on the roof if the inspection is sloppy, so here's a tight workflow that protects both your close rate and your compliance:

1. Establish access and safety first. Ladder secured, footing checked, nobody on a roof that isn't safe to walk. Document from a drone or eaves if pitch or conditions are unsafe.
2. Wide-to-tight photos. Start with a full-roof context shot, then each slope, then the specific wear or impact points. Every close-up needs a wide shot that locates it on the roof.
3. Reference scale. Put a chalk circle or a coin next to hail bruising or a tape next to a damaged area so size reads in the photo.
4. Document all elevations and accessories. Ridge, hips, valleys, flashing, vents, gutters, soft metals — soft metals (gutters, downspouts, vents) often show impact even when shingles are ambiguous.
5. Date and geotag. Photos that prove when and where you documented are worth far more than undated ones, especially as storm evidence degrades.
6. Note, don't diagnose coverage. Record what you see (“granule loss across the south slope, soft-metal impact on the gutter run”). Do not write “covered” or “claimable.”
7. Build the estimate. Translate the documented scope into clear, standard line items. This is the deliverable you hand the homeowner.
8. Present and hand off. Show the homeowner the photos and the estimate, explain what you found in plain language, and let them decide their next step. The next step is theirs to take with their insurer.

That workflow does two jobs at once: it raises your close rate (homeowners sign when they can see the problem), and it keeps you on the right side of the public-adjusting line.

Edge cases and things pros get wrong

Chasing set rate at the expense of quality

If you reward raw sets, reps will set garbage — inspections on roofs that were never going to sell — to juice the number. Your set rate looks great and your close rate craters. The fix is to track both together and reward the paired metric, or reward signed jobs while still diagnosing with the set-rate splits. A set rate is a diagnostic, not a payroll target.

Confusing a seasoning problem with a skill problem

A rep three weeks in should have a lower set rate than a vet. If you fire green reps for posting green-rep numbers, you'll never develop anyone and you'll live in the churn doom loop forever. Set tenure-adjusted expectations and look at the trend for new reps, not the absolute level.

Reading too much into a small sample

Forty doors is noise. A rep can have a brilliant day and a brutal day with no change in skill. Don't make personnel decisions on a week of data; look at a few hundred contacts before you conclude anything about a rep's pitch.

Letting weather mask everything

A cold rainy week tanks contact rate for reasons that have nothing to do with anyone's pitch. Annotate your tracking with conditions so you don't “coach” a rep for a thunderstorm.

The DNC and licensing basics

If you're cold-calling to set appointments, the federal Do-Not-Call rules apply, and several states layer their own door-to-door solicitation and right-of-rescission rules on top. Know your local permit and licensing requirements before you knock; a citation or a stop-work order kills a lot more sets than a weak opener. This is a where-you-operate question — check your state and municipality.

Confusing “nobody home” with “nobody interested”

A huge chunk of low set rate is just absence, not rejection. Aggressive re-knock discipline — logging not-homes and circling back at a better time — recovers sets that look like losses. Many reps mentally write off a not-home as a no; it isn't.

A 30-day plan to lift a stuck set rate

Week 1 — Measure and time. Stand up per-rep, per-stage tracking (doors, contacts, sets, time block, area). Move the knocking window to the evening and Saturday-late-morning slots. Compute everyone's four ratios and read the symptom table. Don't change pitches yet — just measure and fix the clock.

Week 2 — Opener and ask. Rebuild the opener around local proof, a specific roof observation, lowering the temperature, and a tiny ask. Shrink the door ask to “a quick look.” Role-play until it's relaxed. Track set-off-contacts daily.

Week 3 — Doors and inspection. Concentrate knocks on roofs that are actually due by age or storm wear instead of whole streets. Tighten the on-roof documentation workflow so inspections close. For storm work, drill the compliant urgency frame and the do-not-say list until every rep can recite it.

Week 4 — Coach the leaks and lock it in. With three weeks of clean data, coach each rep on their specific leak — contact problem, pitch problem, or dead area. Bring the green reps along on tenure-adjusted expectations. Make the weekly funnel review a permanent ritual.

Run that and a stuck single-digit set rate routinely moves, often substantially, because most of the gain comes from the cheap, unglamorous levers — timing, opener, and which doors you knock — not from some closing secret.

The short version

Your set rate is low for one or more of eight reasons, and the order matters. Most of the time it's not the pitch. It's that you're measuring the wrong thing, knocking at the wrong times, opening with a permission ask, or — most often and most invisibly — standing in front of roofs that were never going to convert. Define the metric, split it into contact rate and set-off-contact, read the symptom table, and work the rebuild in order. Fix the clock first because it's free. Fix the doors themselves because that's the leak nobody else is touching. And if you're in storm work, sell the documentation and the estimate, never the outcome — the homeowner files, the insurer decides, and you stay in business.

The roofs are out there and a real share of them are due right now. The whole problem is knowing which ones before your crew burns a day finding out the hard way.

FAQ

What is a good set rate in roofing door-to-door sales?

It depends heavily on market, season, and whether you're in a fresh storm or cold retail, so treat ranges as orientation. As a rough field guide: contact rate (someone answers) of 25–35% is solid; setting an inspection on 15–25% of those contacts is solid; that works out to roughly 5–8% sets off raw doors. Inspection-to-sign of 35–50% is a healthy close rate. Below those, look at which stage is dragging — don't just stare at the blended number.

Why is my set rate low even though my pitch is good?

Almost always one of two things. Either your contact rate is low (you're knocking at the wrong times or in dead areas, so few people answer and your raw set rate looks bad through no fault of the pitch), or you're knocking the wrong houses (roofs that don't need you, so there's nothing to set). Split your set rate into contacts-per-door and sets-per-contact. If sets-per-contact is healthy, the pitch is fine and the problem is timing or targeting.

How do I figure out whether it's a timing problem or a pitch problem?

Calculate two numbers per rep: contact rate (contacts ÷ doors) and set off contacts (sets ÷ contacts). Low contact rate with a healthy set-off-contact means timing, route, or a dead area — fix the clock, not the script. Healthy contact rate with a low set-off-contact means the opener or porch pitch needs work. That single split resolves most misdiagnosed low-set-rate complaints in about a minute.

What are the best times to knock to improve set rate?

For most residential areas, roughly 4:30–8 p.m. on weekdays (until you lose daylight) and 10 a.m.–2 p.m. on Saturdays produce the most contacts. Weekday afternoons 1–4 p.m. are the graveyard shift because working households are out. Shifting your knocking window alone can lift contact rate 30–50% with no change to anyone's pitch, which raises raw set rate directly.

How does knocking the wrong houses lower my set rate?

If a street is full of new construction, recent re-roofs, and roofs that simply aren't worn enough to replace, there's nothing to set no matter how good the rep is. The problem is you can't see roof age from the curb, and county or Zillow records show year built, not year roofed — a re-roof is invisible to them. Concentrating knocks on roofs that are actually due by age or storm wear puts the same pitch in front of far more receptive doors.

Should I pay reps on set rate or on closed jobs?

Be careful rewarding raw sets, because reps will set low-quality inspections to juice the number, which tanks your close rate while the set rate looks great. Either reward signed jobs (and use the set-rate splits purely as a diagnostic to coach), or reward a paired metric so quality and quantity move together. A set rate is a diagnostic, not a payroll target on its own.

How does roof-age and storm data actually help my set rate?

It tells you, before your crew leaves the truck, which roofs on a street are old enough to be due and which ones took a real storm — so reps skip the new and recently re-roofed houses and concentrate on receptive doors. That sharpens the opener too, because the rep has a specific, true reason to be at that house. RoofPredict estimates roof age as a range from aerial imagery and models storm impact per individual roof rather than per ZIP. The honest limits: age is a range, not a date, and a storm score is odds the roof was hit, not proof.

What can I legally say about insurance to create urgency at the door?

You may inspect, document the roof thoroughly with dated photos, write an accurate repair estimate for your own work, and hand it to the homeowner to use. You may say storm evidence is freshest now and degrades over time. You may NOT promise the claim will be approved, say the deductible will be waived or absorbed, advertise a free roof, or offer to handle or negotiate the claim with the insurer for the homeowner — in most states that's unlicensed public adjusting and it's illegal. The safe frame: you document and estimate, the homeowner files, the insurer decides.

Why do storm set rates start high and then crash?

Right after a hail or wind event, homeowners are primed and the door is warm, so sets spike. Then the swarm hits — every contractor in three states floods the same ZIPs, and by week two homeowners have been knocked twenty times and their reflex 'no' is armed. Targeting also gets sloppy because hail falls in streaks and only some roofs in an affected ZIP actually took damaging impact. Per-roof storm modeling helps you concentrate on the houses most likely to have a real, documentable problem instead of the ones that were grazed.

How long should it take a new rep to reach a normal set rate?

Set tenure-adjusted expectations: a rep three weeks in should post a lower set rate than a veteran, and that's normal seasoning, not a skill failure. Watch the trend, not the absolute level, and don't make personnel decisions on a week of data — forty doors is noise. The fastest way to keep green reps long enough to improve is to point them at roofs that can actually say yes so they get early wins, because a rep who closes something in week one stays and one who eats forty cold noes a day quits.