Three clocks decide whether a homeowner picks up. The law keeps one, the homeowner keeps another, and your dialer logs a third that most floors never read. The pages that rank for this query usually publish a single hour and cite nobody. This one does something more useful: it shows the one hour-of-day dataset we can actually source, explains why it is not a phone number, and gives you the procedure for finding your own.
The calling-window section describes operational practice, not legal advice. Rules change and state law varies — confirm your dialing setup with compliance counsel.
The first clock: the law’s
The federal Telemarketing Sales Rule restricts outbound sales calls to 8 a.m. to 9 p.m. in the called party’s local time. Several states narrow that — later starts, earlier stops, no calls on certain days or holidays — and if you dial nationally you inherit the tightest rule that applies to each individual call.
The operational trap is the phrase called party’s local time. Floors read it off the area code, and area codes stopped being a location the day numbers became portable. If you are calling a homeowner about their house, you already hold a better clock: the property address. Derive the window from that, load callback rules that respect it, and the evening block — the one the folklore says is best — stops being the block where a national list quietly breaks the rule on one coast. DNC compliance for outbound solar covers the rest of that discipline.
Everything below happens inside this window. There is no best time outside it.
The second clock: the homeowner’s
We build a canvassing app rather than a dialer, so the hour-of-day data we can put a methodology behind is for the door. Across 237,329 logged door outcomes on 7,241 real field days, derived from anonymised logs on 17 August 2026, the share of knocks someone answered moved like this:
- 31.5% in the 9 a.m. hour, rising through the morning
- A flat plateau of roughly 36–38% from noon to 7 p.m.
- 40.8% in the 7 p.m. hour and 46.2% in the 8 p.m. hour
- By day: Monday highest at 38.1%, Saturday lowest at 34.7%
Two things in that pattern are worth carrying to a phone floor, and two things are not.
Worth carrying: the shape. Presence at home rises through the day and peaks after dinner, the evening edge is smaller than folklore claims until the last permitted hours, and Saturday — the default outbound day — is the weakest of the seven. The other transferable finding is about behaviour rather than homeowners: knock volume in that data peaks between 1 and 4 p.m. and collapses after 7, so the best hours were the least worked. Floors do the same thing with a shift pattern.
Not worth carrying: the numbers. A knock is a fixed address. A dial is a number that may have been ported to another state, screened by a handset, labelled spam by a carrier, or reassigned to a stranger. None of those happen to a front door. The door data tells you the homeowner’s clock; it cannot tell you what fraction of dials will reach it. That fraction is set by your file and your caller-ID health before the hour of day gets a vote — what is a good contact rate puts time of day fifth on the list of what moves it, behind match quality, list age, line type and caller-ID reputation.
The third clock: your dialer’s
Every dial you have ever placed is logged with a timestamp and a disposition. That is an hour-of-day study of your own homeowners, on your own file, with your own caller IDs, and it is almost never read. Here is the procedure.
- Dial every permitted hour for two weeks. Not the hours the shift pattern already covers — all of them, including the 8 a.m. and 8 p.m. hours that are usually empty. A report on hours you never dial cannot tell you anything about them.
- Spread each record across hours. If a record is only ever attempted at 2 p.m., you have learned about 2 p.m. and nothing else. A fixed attempt policy that rotates each record through morning, afternoon and evening is both the fairer test and the better cadence.
- Report right-party contact per attempt, by local hour. Not connects — a connect with the wrong household at 6 p.m. is not evidence that 6 p.m. is good. Local hour means the property’s hour, not the floor’s.
- Split by line type. Mobiles travel with the person and landlines stay with the house. Their curves differ, and a blended curve describes neither.
- Split by segment age. A worked segment has already lost its easy-to-reach records, so its hour curve is flatter and lower everywhere. Compare fresh with fresh.
Two weeks of that produces a table nobody else can publish for you, because it is specific to your file, your caller IDs and your geography. It will also usually show the same thing the door data shows: the best hours are the ones you were not staffing.
What changes between a knock and a call
| Factor | At the door | On the phone |
|---|---|---|
| Location certainty | Absolute — the rep is standing there | Only as good as the match; a ported mobile can be anywhere |
| Screening | A peephole | Caller-ID display, spam labels, silenced unknown callers |
| Legal window | Local ordinances and posted hours, where they exist | 8 a.m.–9 p.m. called party’s time federally, tighter by state |
| What “answered” means | Someone in the household opened the door | Someone holding the number picked up — not always the owner |
| Repeat attempts | A second visit, days later | Several attempts in a day are possible and usually a mistake |
The last row is the one that bites. Because redialling is cheap, floors chase the evening peak by stacking attempts on the same number, and the number gets flagged. A good hour worked badly is worse than an average hour worked once. Outbound dialer setup covers attempt limits and caller-ID health, which decide whether the best hour stays the best hour.
Scheduling against what you find
Once the table exists, the decision is mechanical. Put the most reps on the hours with the highest right-party contact per attempt, keep a thin presence on the rest so the data keeps refreshing, and hold callbacks for the hour the homeowner was reached the first time — a person who picked up at 7:40 p.m. on a Tuesday is more likely to pick up then again than at any hour a study could recommend.
Then re-run the report quarterly. Daylight-saving shifts, a new file source, a change in caller-ID reputation and a school calendar all move the curve, and a schedule set once in spring is a guess by autumn. The wider operating rhythm this sits inside — data supply, telephony, compliance and measurement — is laid out in how to run a solar call center.
Disclosure: Scout Data is our product. We sell the file this page keeps pointing at, matched by name to the owner on title with the property address on every record, and we do not publish a phone contact rate by hour for it, for the same reason we will not quote one here: it would be your floor’s number, on your caller IDs, not ours.
Frequently asked questions
What is the best time of day to call homeowners?
Within the legal window, the honest answer is “the hour your own dispositions say,” and it takes two weeks of dialling every permitted hour to find out. We do not publish a phone benchmark because we do not run the dialers. The one hour-of-day dataset we do publish is for door knocks, where contact rises from 31.5% in the 9 a.m. hour to 46.2% in the 8 p.m. hour — and a knock is not a call.
What hours are you legally allowed to call homeowners?
The federal Telemarketing Sales Rule restricts outbound telemarketing calls to between 8 a.m. and 9 p.m. in the called party’s local time, and several states narrow that window or add days on which calls are not permitted. Derive local time from the property address rather than the area code, and confirm the applicable rules with compliance counsel — this page is operational practice, not legal advice.
Is it better to call homeowners in the evening?
Folklore says yes by a wide margin. Our door data says the evening edge is real but smaller than the folklore through dinner, and only becomes large in the last permitted hours — which are also the least worked. On the phone, the evening block is also where a national list dialled on the floor’s clock breaks the calling window on one coast. Test it rather than assume it.
Does the best time to call differ for mobiles and landlines?
Almost certainly, and your own file is the only place to find out. A mobile travels with the person and a landline stays with the house, so the two follow different daily patterns and are governed differently for automated dialling. Split every hour-of-day report by line type before drawing a conclusion from it.