Search “door knocking statistics” and you’ll find the same uncited numbers on every sales-software blog — a “2% close rate” here, “knock 100 doors a day” there, no source anywhere. We build a canvassing app, so we measured instead. This page reports what 237,329 logged door outcomes across 7,241 real field days — solar, roofing, and HVAC teams knocking U.S. homes from 2022 through 2026 — actually show: how many doors answer, when they answer, and what a real day of knocking looks like. All figures derived August 17, 2026; methodology (and the numbers we deliberately won’t publish) at the bottom.
63% of knocks reach nobody
On field days, someone answers at 37.0% of logged doors — the other 63.0% are marked no-answer or come-back-later. That’s the honest denominator under every door-to-door metric: roughly two of every three doors a rep walks to are a dead knock, before a single word is spoken. And since reps skip logging some empty doors, the true unanswered share is likely higher still. Renters answer 11.8% of all logged knocks — which means nearly a third of the conversations a homeowner-products rep starts are with someone who can’t sign.
Put differently: a conversation with a decision-capable human costs about 2.7 doors, and reps report an appointment once per 17.2 doors. Territory selection — knocking doors that are worth 17 knocks — is the whole game, which is why door knocking apps that only count knocks miss the point.
The best hours are the ones reps don’t work
Contact rate is the share of logged knocks where the rep talked to someone (interested or not) rather than logging no-answer. Hours are local time at the door.
| Hour (local) | Knocks logged | Someone answered |
|---|---|---|
| 8–9 a.m. | 5,057 | 32.8% |
| 9–10 a.m. | 8,347 | 31.5% |
| 10–11 a.m. | 13,271 | 34.9% |
| 11 a.m.–noon | 19,882 | 34.3% |
| Noon–1 p.m. | 24,711 | 37.3% |
| 1–2 p.m. | 28,859 | 37.0% |
| 2–3 p.m. | 28,784 | 36.8% |
| 3–4 p.m. | 27,962 | 36.8% |
| 4–5 p.m. | 25,172 | 38.2% |
| 5–6 p.m. | 21,271 | 36.9% |
| 6–7 p.m. | 13,573 | 36.6% |
| 7–8 p.m. | 5,184 | 40.8% |
| 8–9 p.m. | 3,106 | 46.2% |
Two findings here contradict the folklore. First, the evening edge is real but small through dinner: 4–8 p.m. runs a 37.6% contact rate against 36.3% for 10 a.m.–3 p.m. — 1.3 points, not the doubling the blogs promise. The plateau from noon to 7 p.m. is remarkably flat. Second, the real lift arrives after 7 p.m. — 40.8%, then 46.2% in the 8 p.m. hour, half again better than the 9 a.m. hour’s 31.5% — exactly when most reps have quit: knock volume peaks between 1 and 4 p.m. and collapses after 7. The best knocking hours of the day are the least worked.
Monday beats Saturday
| Day | Knocks logged | Someone answered | Appointment set |
|---|---|---|---|
| Monday | 41,702 | 38.1% | 5.6% |
| Tuesday | 46,726 | 36.7% | 5.2% |
| Wednesday | 42,680 | 37.5% | 6.0% |
| Thursday | 41,269 | 37.0% | 5.9% |
| Friday | 34,305 | 36.6% | 6.4% |
| Saturday | 23,054 | 34.7% | 5.7% |
| Sunday | 7,593 | 37.7% | 7.4% |
Saturday — the default canvassing day — has the worst contact rate of the week at 34.7%. Monday has the best at 38.1%. The spread is modest, but its direction matters: weekday evenings beat weekend daytime, and the teams stacking every rep on Saturday are paying a measurable answer-rate tax for it. Sunday posts the best contact and appointment rates on the board (37.7% and 7.4%) on the week’s lowest volume — a small sample, but a consistent one.
What a real field day looks like
Across 7,241 field days, the median rep-day has 23 logged door outcomes. The top quartile logs 40+, the top decile 65+. The “knock 100 doors a day” standard the industry quotes describes the far tail, not the working norm — and since logging is voluntary, these are floors on doors approached, not ceilings. A rep at the median pace, at a 5.8% self-reported appointment rate, books roughly one to two appointments per field day — which squares with how field teams actually experience the grind.
The close rate we won’t publish
You may notice a number missing. Sites that publish door-to-door “close rates” are guessing, and our data explains why: sale outcomes in canvassing logs are recorded as a door’s current status, often applied or updated long after the knock — including teams bulk-tagging existing customer lists. In our raw data, naive math would put “sold” at over 8% of knock outcomes, which no one who has knocked a door believes. Rather than launder that into a plausible-sounding figure, we publish the funnel only as far as the data honestly supports: doors → answers → conversations → self-reported appointments. Anyone quoting a precise national door-to-door close rate without a methodology section is reciting folklore.
Methodology
- Source. Anonymized, aggregated door-outcome logs (“dispositions”) recorded by door-to-door sales teams in Scout’s canvassing app — 912,238 status-tagged doors logged by 1,050 reps across 168 teams, January 2022 through August 17, 2026. No individual rep, team, or address is identifiable in anything published here.
- What counts as a knock. The five statuses reps log at the door in the moment: No Answer, Go Back, Not Interested, Renter, and Appointment Set — 445,788 outcomes. Planning pins (To Visit), teams’ custom statuses, desk-taggable statuses (Disqualified), and after-the-fact sale statuses (Sold, Installed) are all excluded from rate math.
- Field-day filter. Rates and timing use only “field days” — one rep’s local day with 10–200 logged outcomes — to exclude bulk imports and office data entry. That leaves 237,329 outcomes across 7,241 rep-days (53.6% of live outcomes). Without this filter, overnight data-entry sessions masquerade as 85%-contact miracle hours.
- Contact rate. Not Interested + Renter + Appointment Set, divided by all five live statuses. It measures “a human answered,” not interest. Sold is excluded even though a sale implies contact — including it would import the retro-tagging problem described above; the effect is a slightly conservative contact rate.
- Local time. Approximated from each door’s longitude (UTC offset = longitude ÷ 15, rounded), continental U.S. coordinates only. Daylight-saving shifts can smear an hour at timezone edges; they cannot create the 9 a.m.–8 p.m. gradient shown.
- Status timing. A door’s status is its current state — a door knocked at 2 p.m. and revisited at 6 shows its final status at its first-logged time. This smears some revisit outcomes toward earlier hours; it does not invert the pattern.
- Self-report limits. Reps log what they choose to log. Unlogged empty doors mean the unanswered share is a floor; “Appointment Set” is the rep’s claim at the door, not a verified calendar event.
Derived August 17, 2026. Spot something that looks wrong? Tell us — we’d rather correct a number than defend it.
Frequently asked questions
What percentage of doors answer when you knock?
On real field days in our data, someone answers at 37.0% of knocked doors — 63.0% of knocks reach nobody. The rate swings by time of day, from 31.5% in the 9 a.m. hour to 46.2% in the 8 p.m. hour. Reps don’t log every empty door, so the true unanswered share is, if anything, higher.
What is the best time to door knock?
Answer rates climb through the day: roughly 31–33% in the 8–10 a.m. hours, a flat 36–38% plateau from noon to 7 p.m., then 40.8% in the 7 p.m. hour and 46.2% in the 8 p.m. hour. The folklore that evening is dramatically better than midday is mostly false — 4–8 p.m. beats 10 a.m.–3 p.m. by just 1.3 points — but the last two hours of daylight really are the best, and they’re the least worked.
What is the best day of the week for door knocking?
Monday, at a 38.1% contact rate — and Saturday is the worst at 34.7%, despite being a canvassing staple. Sunday shows the highest contact and appointment rates in our data (37.7% and 7.4%), but it’s only 3% of logged knocks, so treat that as directional.
How many doors can a door-to-door rep knock in a day?
The median field day in our data has 23 logged door outcomes; the top quartile logs 40 or more and the top decile 65 or more. Logging is voluntary — reps skip some doors — so these are floors on doors physically approached, not ceilings.
What percentage of door knocks lead to an appointment?
Reps mark 5.8% of logged knocks “Appointment Set” — one appointment per 17.2 doors. That’s self-reported by the rep at the door, not verified against a calendar, and teams differ in how loosely they use the status, so treat it as what reps claim, measured consistently.
Can I cite these statistics?
Yes — cite freely with a link to this page. The figures come from Scout’s own anonymized canvassing logs, and the methodology section explains exactly what is and isn’t counted — including the numbers we chose not to publish.