Roofing Canvassing KPIs to Track Per Rep (The Numbers That Actually Predict Signed Jobs)

Most roofing companies measure their canvassers the way a slot machine measures a gambler: they only count the jackpot. A rep either signed a deal this week or they didn't, and that single number decides who gets praised on the Monday call and who gets the cold shoulder. The problem is that a signed contract is the slowest, noisiest, most lagging signal you have. By the time a deal closes, the work that produced it happened three to six weeks earlier. If you wait for contracts to tell you whether a rep is good, you find out far too late to fix anything.

The fix is to track the activity and conversion numbers that sit upstream of the contract, per rep, every week. Done right, a per-rep KPI sheet tells you within seven to ten days whether someone is going to make their number, exactly where they're leaking, and what one coaching conversation will move the needle. Done wrong, it becomes a wall of vanity metrics that nobody reads and everybody games.

What follows is the full operating system: which metrics matter, the ratios that connect them, realistic benchmarks pulled from how the work actually behaves, a weekly scorecard you can copy, how to build a quota off the funnel instead of off a wish, and the edge cases that break naive tracking. The goal is that a sales manager reading this can rebuild their entire measurement system on Monday and have a sharper crew by the end of the month.

Why per-rep, and why weekly

Team-level numbers hide everything that matters. If your crew of six closed nine jobs last week, that average looks fine. It also disguises the rep who closed five, the three who closed one apiece, and the two who closed zero and are quietly deciding whether to quit. You can't coach an average. You coach a person, and you coach them on the specific step where their personal funnel narrows.

Weekly is the right cadence for canvassing because the activity is high-frequency and the feedback loop is short. A rep knocks hundreds of doors a week. That's enough volume for the ratios to stabilize and mean something. Monthly is too slow — you waste three or four weeks letting a fixable problem compound. Daily is too noisy — one rainy afternoon or one neighborhood full of renters will swing a single day's numbers so hard that you'll chase ghosts. Weekly smooths the noise without letting problems fester.

There's also a trust dimension. Reps who only ever hear about closed deals feel like the only thing the company values is money out of their pocket. Reps who get measured on the parts of the job they actually control — how many doors they hit, how many real conversations they had, how many roofs they got onto — feel like the company sees the work. That distinction quietly drives retention, and retention is the hidden KPI behind every other number, because a canvasser doesn't reach full productivity until month three or four.

The canvassing funnel, stage by stage

Every canvassing KPI is really a measurement of one stage of a funnel or the conversion between two stages. Before listing metrics, get the funnel itself straight, because the whole system hangs off it. A typical residential roofing canvassing funnel looks like this:

1. Doors knocked — every physical door (or every dialed number / mailed address, if you canvass lists) the rep attempts.
2. Contacts (answers) — doors where a human actually opens up and talks. The rest are not-homes, no-answers, and hard refusals at the threshold.
3. Conversations — contacts that go past "no thanks" into an actual exchange about the roof, the storm, the neighborhood, anything.
4. Inspections set or performed — the rep gets permission to look at the roof, whether that's a ladder-up right now or a scheduled return.
5. Sits / presentations — the rep sits with the decision-maker(s) and presents findings and a path forward (often after the inspection).
6. Signed agreements — contract, inspection agreement, or contingency document signed, depending on your model.
7. Funded / installed jobs — the deal that actually produces revenue, net of cancellations.

Not every company uses every stage, and storm-restoration funnels differ from retail-replacement funnels (more on that below). But the structure holds: bodies enter at the top, and each step is a filter. Your KPIs measure the width of each filter (volume) and the conversion through each filter (efficiency). A rep can fail by being too narrow at the top — not enough doors — or by leaking at one specific filter while everyone else holds. The whole point of per-rep tracking is to tell those two failure modes apart, because they call for completely different coaching.

The activity KPIs (what the rep controls outright)

Activity metrics measure effort and discipline. They're the least glamorous numbers and the most important, because they're the only ones a rep controls 100 percent regardless of territory, weather, or luck. A rep who under-indexes on activity has no path to results, full stop.

Doors knocked per day and per week

This is the foundation. It's the denominator under almost every other ratio, and it's the first thing to check when someone's results crater. A rep can't close what they never knocked.

What counts as a knock has to be defined precisely or reps will inflate it. A knock is an attempt at an occupied residence during canvassing hours. Driving past doesn't count. A locked gate you couldn't reach doesn't count as a productive knock (track it separately as an access fail). Define it, write it down, and audit it.

Realistic ranges, full-time canvasser, residential:

The spread is enormous, which is exactly why a raw doors number means nothing without territory context. Forty doors in a rural territory is a monster day; forty doors in a tight subdivision is a slow one. This is the single biggest reason naive scorecards mislead — they compare a rep working quarter-acre lots against a rep working five-acre parcels and conclude the second one is lazy.

Normalize by either logging productive canvassing hours alongside doors (so you can compute doors per hour) or by segmenting reps by territory type before you compare. Doors per hour is the cleaner metric because it strips out how long someone actually spent on the doors versus in the truck.

Productive field hours

Hours on the doors, not hours on the clock. The gap between the two is where most underperformance hides. A rep who's "out canvassing" from noon to six but spends ninety minutes at lunch, forty minutes regrouping with a buddy, and an hour driving to a territory they decided looked better is putting in maybe three productive hours. Their doors-per-day looks weak and they swear they're working hard — and from their point of view they are, because they were out all afternoon.

Tracking productive hours, even self-reported with spot checks, reframes the entire conversation from "you're lazy" (which they'll deny and resent) to "your door rate during productive hours is actually strong — we just need more productive hours." That's a problem reps will fix, because it doesn't insult them.

Contact rate (answer rate)

Contacts divided by doors knocked. The share of doors where a human actually opens. This is partly territory and time-of-day, partly the rep's persistence (do they wait the extra few seconds, do they try the side door, do they come back).

Typical residential contact rates run 20 to 40 percent depending heavily on time of day and day of week. Weekday mornings are brutal — everyone's at work. Late afternoon into early evening (roughly 4:00 to 7:30, respecting local solicitation ordinances and daylight) and weekend mid-mornings are when answer rates climb. If a rep's contact rate is low, the first question isn't about their pitch — it's about when they're knocking. A rep grinding 200 doors on Tuesday at 11 a.m. is mostly talking to doorbells.

The conversion KPIs (skill plus territory)

Conversion metrics measure how well a rep moves people from one funnel stage to the next. These are where coaching pays off, because conversion is mostly skill — pitch, rapport, objection handling, the ability to get a homeowner comfortable enough to let a stranger climb on their roof.

Contact-to-conversation rate

Of the people who answered, how many did the rep actually engage past the reflexive brush-off? A homeowner opening the door and saying "not interested" before it's fully open is a contact but not a conversation. Getting them to stay at the door another thirty seconds is the first real skill gate in the whole funnel, and it's the most coachable single thing in canvassing.

If a rep has plenty of contacts but few conversations, their opening line is failing. That's a five-minute fix in a role-play. Make them deliver their first fifteen seconds to you cold, ten times in a row, until it stops sounding like a pitch and starts sounding like a neighbor.

Conversation-to-inspection rate

Of real conversations, how many converted into permission to inspect the roof? This is the hinge of the residential roofing funnel. Everything before it is access; everything after it is sales. A rep who can reliably turn a doorstep chat into a ladder going up against the house has the single most valuable skill in canvassing.

Benchmarks vary by model, but in storm territory with a genuine recent event, strong reps convert a meaningful share of real conversations into same-day or scheduled inspections. In cold retail (no storm, just door-knocking for replacement business) the rate is much lower because there's no urgency manufactured by weather — you're fighting the homeowner's "my roof seems fine" default.

Inspection-to-sit (or inspection-to-presentation) rate

Of roofs inspected, how many turned into a real sit-down with the decision-maker(s)? Two failure modes hide here. First, the rep inspects but never gets both spouses / decision-makers together, so they present to half the household and stall. Second, the rep inspects, finds little, and has nothing compelling to present — which points back to which roofs they're knocking (a targeting problem, not a skill problem).

Sit-to-close rate

Of presentations, how many signed? This is the classic "close rate," and it's the most over-weighted number in the business. It matters, but on its own it's wildly misleading because it ignores how many sits the rep even generated. A rep who closes 60 percent of two sits a week (one job) is losing badly to a rep who closes 30 percent of ten sits (three jobs). Sit-to-close is a quality signal; it has to be read next to volume signals or it lies.

Residential roofing sit-to-close rates commonly land somewhere in the 25 to 50 percent range for skilled reps on qualified, properly-set sits, with storm-restoration contingency sign-ups running higher (because the homeowner is agreeing to an inspection-and-estimate process, not yet to a paid job) and cold retail running lower. Don't anchor too hard on any single number — your own historical data, segmented by lead type, beats any external benchmark.

The compounding view

The reason you track every stage instead of just the close rate is that conversion compounds. Watch what a small lift at one stage does across a realistic funnel:

Rep B is identical to Rep A in every way except a ten-point lift on one mid-funnel conversion — getting more conversations onto the roof. That single difference is 50 percent more signed jobs from the same number of doors. That is why you measure the whole funnel: it tells you the one stage where coaching produces the biggest downstream multiplier, instead of nagging everyone about "closing harder."

The two summary ratios every manager should know cold

Underneath all the stage metrics, two compound ratios summarize a rep's whole funnel and are worth computing every week.

Doors per inspection

Total doors knocked divided by inspections generated. This rolls contact rate, conversation rate, and inspection conversion into one number that answers: how hard does this rep have to work to get one roof to look at? If a rep needs 80 doors per inspection and your team average is 35, something specific is broken in their early funnel — and you now know to look at contact rate and opening pitch, not at their close.

Doors-per-inspection is also the cleanest number for forecasting, because inspections are the first stage in the funnel with real predictive power for revenue and they happen early enough to act on.

Doors per signed job (cost-of-a-deal in effort)

Total doors divided by signed jobs. This is the rep's all-in efficiency, and multiplied by your loaded cost per door (rep pay, gas, materials, management overhead) it gives you a real cost-per-acquisition by rep. A rep with a great close rate but a terrible doors-per-job ratio is expensive; their skill at the bottom doesn't make up for leakage at the top. This number is what you actually pay for, so it deserves a permanent slot on the scorecard.

Quality and integrity KPIs (the ones reps try to dodge)

Activity and conversion describe a healthy rep. These next metrics catch an unhealthy pipeline — the deals that look great on Friday and evaporate by month-end.

Cancellation / rescission rate

Signed jobs that cancel before install, as a share of signed. Every state with a home-solicitation sale gives the homeowner a right to cancel within a set window (commonly three business days under the federal Cooling-Off Rule and analogous state statutes), and some reps quietly rely on emotional, high-pressure closes that sign on Friday and unwind by Tuesday. A rep with a gaudy close rate and a quietly high cancellation rate isn't your best closer — they're your best presser, and pressing produces chargebacks, bad reviews, and complaints.

Track cancellations per rep relentlessly. A close rate is only real after the rescission window passes and the deal funds. Until then it's a promise.

Average job value / revenue per signed job

Not every signed deal is equal. A rep who signs four small repairs is not outperforming a rep who signs two full replacements. Track revenue per signed job per rep so you see who's chasing easy small wins versus building real backlog. This also surfaces a rep who's discounting their way to a high close rate — the close looks great, the margin doesn't.

Lead / list disposition discipline

If you feed reps a list (storm-affected addresses, aged-roof targets, prior estimates, a purchased or enriched mailing list), track how thoroughly they work it: percentage of assigned addresses actually attempted, percentage dispositioned (every record marked as contacted / not-home / not-interested / inspected / etc.), and how fast they hit fresh storm leads. A rep sitting on 200 fresh post-storm addresses for a week is burning the single most perishable asset in the business — urgency decays fast after a storm, and so does your standing against the competitor who knocked first.

Disposition discipline is also what makes every other number trustworthy. If reps don't mark their records honestly, your contact rates and conversion rates are fiction. Audit a sample of each rep's dispositions every week — re-knock a handful of doors they marked "not home" or "not interested" and see if the reality matches. Reps tracking will be more honest precisely because they know spot-checks happen.

Realistic weekly benchmark ranges

Benchmarks are a starting point, not gospel. Your own segmented history is the real standard. But for a sales manager building a first scorecard, here are defensible ranges for a full-time residential canvasser. Treat them as the boundaries of "normal," and investigate anyone well outside them in either direction (the high outlier might be inflating numbers; the low outlier might be drowning).

A few honest caveats. Storm-restoration funnels in the immediate wake of a verified hail or wind event will blow past the "strong" column on conversion because urgency does the selling — don't let those weeks reset your baseline for normal retail conditions. Cold retail canvassing with no storm will sit toward the "weak" column on conversion even for great reps, because they're manufacturing demand from scratch. Segment your benchmarks by lead type or the comparison is meaningless.

A weekly per-rep scorecard you can copy

Here's a concrete scorecard. Keep it to a single screen. The instant it sprawls onto a second page, managers stop reading it and reps stop trusting it. Every column is either an activity number, a conversion number, a quality number, or a forecast number — nothing decorative.

Review it the same way every week. Read the funnel top to bottom for each rep and stop at the first stage that's red. That's the constraint. Coaching the close rate of a rep whose real problem is a 12 percent contact rate is malpractice — fix the top first, because a leak at the top starves every stage below it, and a leak at the bottom only wastes what little reaches it.

Reading the scorecard: three worked diagnoses

Rep with high doors, low inspections. 680 doors, 31 percent contact rate, but only 6 inspections — 113 doors per inspection against a team average of 42. The volume and timing are fine (contact rate is healthy), so the leak is conversation-to-inspection. Their doorstep pitch isn't earning the roof. Coaching: ride along, listen to the transition from chat to "let me grab my ladder and take a quick look," and drill that specific ask. This is a skill fix, not an effort fix.

Rep with great close rate, weak output. 48 percent sit-to-close, which looks elite, but only 3 sits all week and 1 signed job. The close is genuinely good; the problem is they're not feeding the top of their own funnel — 290 doors, well under target. Coaching: this is an activity and discipline conversation, and you can deliver it as a compliment. "Your closing is the best on the team. If you put up 550 doors instead of 290 at the same conversion, you'd lead the board." Reps move fast on that framing because it respects the skill they have.

Rep with strong everything, high cancellations. Great activity, great conversion, 7 signed jobs — and 3 of them cancelled inside the rescission window. Net, they're middle of the pack and generating complaints. The close is high-pressure, not high-trust. Coaching: slow the close, set the deal up to want to stay signed, confirm both decision-makers were present and bought in. A signed deal that cancels cost you everything a no-sit costs plus the chargeback and the reputation hit.

The scorecard didn't just rank these three reps — it told you three completely different coaching conversations, and not one of them was "close harder." That is the entire payoff of per-rep, stage-by-stage tracking.

Building a fair quota off the funnel

Most roofing quotas are pulled out of thin air: "everybody needs four jobs a month." Then half the team misses, morale tanks, and nobody knows whether four was even achievable. Build the quota off the funnel instead, working backward from the revenue you need, and it becomes both fair and defensible.

Work backward through your own historical conversion rates:

1. Start with the monthly revenue target for the rep — say $80,000 in funded work.
2. Divide by average funded job value — say $16,000 — to get jobs needed: 5.
3. Adjust for cancellations. At a 12 percent cancellation rate, you need 5 ÷ 0.88 ≈ 6 signed jobs to net 5 funded.
4. Divide signed by sit-to-close. At 33 percent: 6 ÷ 0.33 ≈ 18 sits.
5. Divide sits by inspection-to-sit. At 60 percent: 18 ÷ 0.60 = 30 inspections.
6. Multiply inspections by doors-per-inspection. At 45: 30 × 45 = 1,350 doors.
7. Divide by working days. Over ~21 canvassing days: about 64 doors per day, which at a 20-door-per-hour suburban pace is roughly 3.2 productive hours of actual door time per day.

Now you have a quota expressed as an activity target the rep controls — roughly 64 doors a day or 3-plus solid hours on the doors — that you can prove leads to the revenue number, given your real conversion rates. When a rep pushes back, you walk them through the same seven steps with their own numbers. The argument stops being "the boss wants more" and becomes "here's the math, and here's the one ratio we could improve to lower the door target." That's a conversation reps respect, because it's honest.

It also instantly shows when a quota is impossible. If the math demands 1,350 doors and your territory only physically contains 900 unworked doors that month, the problem isn't the rep — it's territory supply, and no amount of pep talk fixes a supply problem.

Where the leads come from changes every number

Everything above assumes the rep is choosing which doors to knock more or less at random within a territory. The fastest way to move every downstream KPI isn't to make reps knock harder — it's to make them knock smarter doors. Two roofs on the same street can have wildly different odds of becoming a job: one is a fifteen-year-old roof that just took golf-ball hail on its weather-facing slope, the other is a four-year-old roof in the storm's shadow. Same effort, completely different outcome. If your reps are knocking both with equal weight, half their doors are nearly dead on arrival, and their doors-per-inspection number suffers for reasons that have nothing to do with their skill.

This is where targeting data does real work, and where it's worth being precise about what it can and can't do. Tools like RoofPredict score a contractor's territory or existing list house-by-house on two signals that drive canvassing odds: an estimated roof-age range per address derived from aerial imagery, and storm exposure modeled per individual roof rather than per zip code. The output isn't a guarantee about any single house — roof age is a range, not a birth certificate, and storm modeling is odds, not proof a given roof is damaged. What it does is rank the doors, so a rep spends the day on roofs that are statistically more likely to be aging out or storm-worn, and skips the four-year-old roofs that will eat time for nothing. It can also enrich a contractor's own CRM or mailing list with those roof-age and storm signals, so the addresses you already own get prioritized instead of worked blind.

What this means for your KPIs is concrete: when the door list itself is pre-ranked, doors-per-inspection drops (fewer wasted knocks), conversation-to-inspection climbs (more roofs that genuinely have something to find), and your benchmark ranges shift. That's a feature, not a problem — but it's why you re-baseline whenever you change lead sources. A rep working a ranked, storm-modeled list should not be held to the same doors-per-inspection number as a rep cold-knocking a random subdivision, and if you compare them on the same yardstick you'll reward the lucky list and punish the harder territory. Honest limits matter here: better targeting raises your odds and tightens your routes; it does not sign the deal, it does not replace a real inspection, and it does not tell you a roof is damaged. The rep still has to knock, build trust, get on the roof, document what's actually there, and present well. Targeting just makes sure the hours they spend land on doors worth knocking.

Storm-restoration vs. retail: same KPIs, different physics

The metric names are identical across storm and retail canvassing, but the numbers behave so differently that you must track them as separate segments or your averages turn to mush.

Storm restoration (canvassing after a verified hail or wind event) runs on urgency and a tight clock. Contact rates and conversion rates spike because homeowners know something happened and neighbors are already getting roofs done. But the window is short — every day after the event, more competitors knock and the urgency fades. Speed-to-door becomes a KPI in its own right: how fast does a rep hit fresh storm addresses? A separate storm scorecard should add metrics like hours from event to first knock and percentage of fresh storm list worked within 72 hours. Conversion benchmarks should be set higher because the conditions are richer; holding storm weeks to retail benchmarks makes everyone look like a genius and resets your baseline wrong.

Retail replacement (cold canvassing, no storm) is a grind of manufactured demand. Conversion is lower at every stage, the sales cycle is longer, and a single rep might nurture a homeowner across several visits before a sit. Here, pipeline metrics matter more than same-week conversion: follow-ups scheduled, callbacks completed, and the age of open opportunities. A retail rep with zero closes this week but twelve warm follow-ups booked is building something; a storm rep with twelve open follow-ups and no closes is letting perishable urgency rot.

A word on what stays the same regardless of segment: the compliance line on insurance work. Plenty of canvassing in storm markets bumps into claims. A rep can knock a storm-hit street, inspect a roof, photograph and document the damage thoroughly, and prepare an accurate, Xactimate-aligned repair estimate to hand the homeowner. The homeowner files their own claim, and the insurer decides coverage. What a rep must never do — and what you should explicitly train and audit against — is negotiate or "handle" the claim for the homeowner for a fee, interpret the policy or what's covered, promise a specific payout or approval, promise the deductible will be waived or absorbed, or advertise a "free roof." Those cross into unlicensed public adjusting and deceptive-advertising territory and create real legal exposure. The safe, durable frame is: document thoroughly, write an honest estimate, hand it over, and let the homeowner file and the carrier decide. Bake that do-not-say list into your canvassing script and spot-check it the same way you spot-check dispositions. A rep who promises a free roof at the door is a liability dressed up as a closer.

The ten mistakes that wreck per-rep tracking

Even managers who track diligently sabotage themselves in predictable ways. Watch for these.

1. Measuring only the close. The most common and most expensive error. You learn about problems weeks late and can't tell why anyone is winning or losing.
2. Comparing reps across different territories on raw counts. Five-acre lots and quarter-acre subdivisions are different jobs. Normalize by doors-per-hour and segment by territory type, or you'll fire your best rural rep.
3. Letting reps define their own "knock." Without a written definition and audits, the activity numbers inflate until they're meaningless.
4. Ignoring cancellations. A close rate before the rescission window is a rumor. Track funded, net of cancels, per rep.
5. Too many metrics. A 30-column scorecard gets read by nobody. Pick the dozen that drive decisions and cut the rest. Vanity metrics crowd out signal.
6. No territory-supply check. Quotas that demand more doors than the territory physically holds aren't quotas, they're traps. Count the unworked doors before you set the number.
7. Same benchmarks for storm and retail. They're different sports. Segment or your averages are noise.
8. Coaching the wrong funnel stage. Always fix the highest red stage first. Coaching closing on a rep with a top-of-funnel leak wastes both your time.
9. Punishing honesty. If reps get hammered the moment they log a bad day, they stop logging honestly and your whole dataset rots. Reward accurate dispositions even when the numbers are ugly.
10. Set-and-forget benchmarks. Re-baseline when lead sources, seasons, territories, or comp plans change. A benchmark that was right in last year's storm season can be wrong this spring.

Tying KPIs to comp without breaking the system

The moment a metric drives pay, reps optimize for the metric instead of the outcome — so the way you connect KPIs to compensation determines whether they help you or corrupt your data. Pay purely on signed contracts and reps press hard, cancellations climb, and your activity numbers become guesswork because nobody bothers logging the steps that don't pay. Pay purely on activity (doors knocked) and reps inflate knocks and stop caring whether anyone signs. Neither extreme works.

The durable structure is a base or activity floor plus commission on funded, net-of-cancellation revenue. The activity floor (a minimum doors-per-week or productive-hours target derived from the funnel math) protects the top of the funnel and gives newer reps a survivable income while they ramp. Commission on funded revenue — not signed, funded — aligns the rep's payout with the only number that pays the company, and it quietly kills the incentive to press a fragile close, because a deal that cancels inside the rescission window never funds and never pays. Some companies add a small spiff for clean disposition discipline or for speed-to-door on fresh storm leads, which works as long as it's small enough that reps don't game it and large enough that they notice it.

Whatever you choose, write down the rule that a metric used for pay must be one you can audit, and audit it. The instant reps believe a number affects their check and isn't being checked, that number inflates. This is why doors-knocked makes a weak primary pay driver and a fine secondary floor: it's easy to inflate, so you verify it with disposition re-knocks and pay the real money on funded revenue, which is verified by your accounting whether anyone likes it or not.

Data hygiene: the boring layer that makes every number real

None of these KPIs are worth tracking if the underlying data is dirty, and roofing canvassing data gets dirty fast. A few disciplines keep it honest.

Log at the door, not at the truck. Reps who wait until the end of the shift to record a day's dispositions are reconstructing from memory, and memory rounds in the rep's favor. A simple mobile form or canvassing app that captures each disposition in the moment produces dramatically cleaner data than an end-of-day tally. The friction of logging on the spot is the price of trustworthy numbers.

Standardize disposition codes and forbid free-text where a code belongs. "Not home," "no answer / heard inside," "not interested," "call back," "inspection set," "inspected," "sit set," "signed," "do not knock" — a short, fixed list that every rep uses the same way. The instant two reps mean different things by "not interested," your contact and conversation rates stop being comparable across the team.

Reconcile the funnel weekly so the stages add up. The number of inspections this week should be traceable to the conversations that produced them; signed jobs should trace to sits. When the stages don't reconcile — more sits than conversations, signed jobs with no logged inspection — you've found either a logging gap or a rep skipping steps in the record, and both need fixing before you trust the ratios.

Keep a "do not knock" and prior-contact memory across the team. Few things damage a brand faster than three different reps from the same company knocking the same irritated homeowner in a month, or a rep knocking an address a teammate already sold. Shared territory and contact history isn't only courtesy — it protects your contact-rate and conversion numbers from being polluted by repeat-knocking the same exhausted doors. This is also where list-enrichment and CRM data earn their keep: an address record that already carries roof-age range, storm exposure, and prior-contact history lets a rep walk up knowing whether this door is worth the knock and whether a colleague has been here before.

A 30-day rollout for a sales manager starting from zero

If you currently track nothing but closes, here's a realistic month to stand up a real system without a revolt.

Week 1 — Define and baseline. Write the precise definition of a knock, a contact, a conversation, an inspection, a sit. Pick the dozen scorecard columns. Have reps log activity for one week with zero consequences attached — you're collecting a baseline, not judging. Tell them exactly that, or they'll game it.

Week 2 — Compute your real ratios. Using week one's data plus whatever close history you have, compute your actual conversion rates at each stage, segmented by lead type. These, not internet benchmarks, become your standard. Identify your team's median for each KPI.

Week 3 — Build quotas off the funnel. Run the seven-step backward math per rep, using your week-two ratios, to convert each rep's revenue target into a daily door / productive-hour target they control. Walk each rep through their own math one-on-one. Let them see it's arithmetic, not a whim.

Week 4 — Run the weekly rhythm. Hold the first real scorecard review. For each rep, read top to bottom, stop at the first red stage, and assign one specific coaching action. Re-knock a sample of dispositions to validate honesty. Then repeat every week, adjusting benchmarks as seasons and lead sources shift.

By the end of the month you'll know, for every rep, exactly where their funnel narrows, whether their quota is mathematically fair, and which single coaching conversation moves their number most. That's a different company than one staring at a closed-deal count and guessing.

The one-sentence version

If you remember nothing else: track the whole funnel per rep every week, fix the highest red stage first, build quotas backward from your own real conversion rates, never trust a close until the rescission window passes, and feed reps the smartest doors you can — because the rep who knocks aging-out and storm-worn roofs converts more from the same effort than the rep knocking blind. Targeting tools that score roofs by age range and per-roof storm exposure, like RoofPredict, move those upstream numbers by pointing the work at doors worth knocking; the rep still has to earn the roof, document it honestly, and close clean. Measure the work that produces the deal, and the deals follow.

FAQ

What are the most important canvassing KPIs to track per rep?

Track at least one metric per funnel stage: productive field hours and doors knocked (activity), contact rate and conversation-to-inspection rate (early conversion), inspections and sits (mid-funnel volume), sit-to-close rate (closing skill), and cancellation rate plus net funded revenue (pipeline integrity). The two summary ratios worth computing weekly are doors-per-inspection and doors-per-signed-job, because they roll the whole funnel into single numbers you can forecast and compare against.

How many doors should a roofing canvasser knock per day?

It depends entirely on territory density. In tight, walkable subdivisions a full-time rep can hit 150-240 doors a shift; in standard suburban it's more like 90-150; in rural areas with large lots and driving between homes, 35-70 is realistic. Because the spread is so wide, doors-per-hour during productive field time is a fairer comparison than raw daily doors, and you should segment reps by territory type before comparing them.

What's a good close rate for a roofing sales rep?

Skilled residential reps commonly close 25-50 percent of qualified, properly-set sits, with storm-restoration contingency sign-ups running higher and cold retail running lower. But close rate alone is misleading: a rep closing 60 percent of two sits produces fewer jobs than one closing 30 percent of ten. Always read close rate next to sit volume and doors-per-signed-job, and only count a close as real after the cancellation window passes and the deal funds.

Why track contact rate and conversation rate instead of just closed deals?

A closed deal is the slowest, noisiest signal you have; by the time it lands, the work that produced it happened weeks earlier. Upstream metrics like contact rate and conversation-to-inspection rate move within days and tell you exactly which funnel stage a rep is leaking at, so you can coach the right thing. Closing harder does nothing for a rep whose real problem is a low contact rate from knocking at the wrong time of day.

How do I set a fair canvassing quota for my reps?

Work backward through your own historical conversion rates. Start from the rep's revenue target, divide by average funded job value to get funded jobs, gross that up for your cancellation rate to get signed jobs, then divide successively by sit-to-close, inspection-to-sit, and doors-per-inspection to land on a daily door or productive-hour target the rep fully controls. Because it's arithmetic built on real ratios, you can defend it line by line, and it instantly reveals when a quota exceeds the territory's actual door supply.

What is the doors-per-inspection metric and why does it matter?

Doors-per-inspection is total doors knocked divided by inspections generated. It rolls contact rate, conversation rate, and inspection conversion into one number that shows how hard a rep works to get a single roof to look at. It's the best early forecasting metric because inspections are the first stage with real predictive power for revenue and they happen early enough to act on. A rep far above your team average on this number has a specific early-funnel problem you can diagnose.

Should storm and retail canvassing be measured with the same benchmarks?

No. The metric names are the same but the numbers behave completely differently. Storm restoration runs on urgency, so contact and conversion rates spike and you add a speed-to-door metric (hours from event to first knock, percentage of fresh list worked within 72 hours). Cold retail manufactures demand, so conversion is lower at every stage and pipeline metrics like scheduled follow-ups matter more. Segment them or your averages become meaningless.

How does roof-targeting data change canvassing KPIs?

When reps knock a list pre-ranked by roof-age range and per-roof storm exposure instead of random doors, fewer knocks are wasted on roofs unlikely to convert. That typically lowers doors-per-inspection and raises conversation-to-inspection rate, which means you should re-baseline your benchmarks whenever you change lead sources. Targeting raises the odds and tightens routes, but roof age is a range not a date and storm modeling is odds not proof, so the rep still has to knock, inspect, document, and close.

How do I keep reps from inflating their canvassing numbers?

Write a precise definition of what counts as a knock, a contact, and a conversation, then audit. Re-knock a sample of doors each rep marked not-home or not-interested and compare reality to their log. Track productive field hours separately from clock hours, and never punish an honestly logged bad day, because the moment reps fear honesty they start gaming the data and your whole dataset rots. Spot-check disposition discipline the same way every week.

Why is cancellation rate a canvassing KPI rather than only an office problem?

Home-solicitation sales carry a legal right to cancel within a set window (commonly three business days under federal and state cooling-off rules), and some reps rely on high-pressure closes that sign on Friday and unwind by Tuesday. A rep with a high close rate and a quietly high cancellation rate isn't your best closer, they're your best presser, and pressing produces chargebacks, complaints, and bad reviews. Tracking cancellations per rep is the only way to tell a durable close from a fragile one.