After-hours calls and what happens to them now

Every contractor believes they know their after-hours volume. Ask them and they will say it is light, maybe a few a week, mostly wrong numbers.

Then they pull the log and it is a different business. The calls are there. They have always been there. Nobody was looking, because a call that nobody answered does not generate a task, a note, or a complaint.

How many are you actually missing?

Pull the log and find out, because the industry averages will not tell you.

CallRail's benchmark of 1.1 million de-identified conversations, announced in January 2025, put the missed call rate for home services at 14%. That was the best of the industries they measured. Health care sat at 32%, legal at 28%, real estate at 9%. The spread between 9% and 32% is the actual finding. A single cross-industry number would be useless, and so would applying the home services average to your specific phone.

Two things are worth knowing before you use that 14% as a benchmark. CallRail defines a missed call as an unanswered inbound call that resulted in a detected voicemail or was detected as abandoned, which is a narrower definition than most people assume. And the sample is drawn from businesses that had already bought call tracking software, which is to say businesses that were already paying attention. Your shop may be better or considerably worse.

The second reason 14% looks low against the numbers vendors quote at you is that vendors conflate answering with booking. ServiceTitan's platform analysis found a typical shop books 42% of the calls it takes. A 42% booking rate is not a 58% miss rate, and the two get mixed constantly.

What should the call log actually show you?

Four columns, split by hour and by day of week.

Total inbound calls. Calls answered by a human. Calls that rolled to voicemail. Calls abandoned before anything picked up. That last group is the interesting one, because those callers did not even wait for the beep.

A voicemail is a lead with a delay. An abandoned call is a lead that already hired someone else.

There is no credible independent research on what share of local business calls arrive outside staffed hours. Every figure in circulation comes from a company selling after-hours coverage, and the honest planning assumption is a range rather than a number: somewhere between a fifth and a third of inbound calls to a consumer-facing local business land outside staffed hours, weighted higher for emergency trades. For a plumber or an HVAC shop, the value skew matters more than the volume skew anyway. After-hours calls are disproportionately emergencies, and emergencies are disproportionately high ticket.

That is the point of counting your own. Your number depends on your hours, your trade, your dispatcher, and whether the line rolls anywhere at 7pm.

Why does an after-hours call convert differently?

Because of who is calling at that hour and what they are willing to do.

An 8pm caller is either in trouble or is finally home from work and dealing with the thing they have been ignoring. Both are high intent. Neither will call back tomorrow.

The Harvard Business Review audit of 2,241 US companies found 23% never responded to an inbound lead at all, 24% took more than a day, and only 37% responded within an hour. The average response time among companies that answered within thirty days was 42 hours. A separate dataset in the same piece, covering 1.25 million leads across 42 companies, found firms contacting within an hour were nearly 7x more likely to qualify a lead than those trying an hour later, and more than 60x more likely than those waiting 24 hours or longer.

Cite that carefully, because one of its co-authors was the founder and chief executive of the software company whose platform produced the data. It is high quality vendor research published under an editorial masthead, not a peer-reviewed result. It was also about online lead response rather than emergency plumbing. Nothing about a 9pm caller makes them more patient than a B2B prospect, but the study does not prove that either.

The other reason the callback plan is weaker than it used to be: a TNS survey in July 2022 found 75% of Americans never answer calls to their wireless phone from a number they do not recognize. Capturing the number and ringing back at 8am is a strategy built for 2016. Resolving on the inbound call is the strategy that still works.

There is a competitive dimension too. The national lead aggregators answer around the clock, because that is their entire product. When your line rolls to voicemail at 7pm you are handing them the call directly, and competing with the aggregators in your own market starts with not gifting them the hours they are strongest.

What is an abandoned after-hours call actually worth?

What are the actual options?

Four, and each has a real cost.

Voicemail. Free, and it works only for callers who already decided on you. For a homeowner comparing three companies, voicemail is a no.

Forward to a cell. Cheap and it works, until it does not. The owner answers at 8pm on Tuesday, not at 11pm on Saturday, and the inconsistency is worse than a policy. Callers cannot tell which night they got.

An answering service. A person answers, takes a message, and passes it along. Better than voicemail. The weakness is that the caller still has not booked anything, and the person answering usually cannot say whether you service their town or handle their equipment.

An automated answer that qualifies and books. Picks up on the first ring every time, gets the address and the problem, books it or escalates to the on-call tech. What an AI receptionist actually does in 2026 is narrower than the marketing suggests, and the narrowness is the point.

Be skeptical here in a specific way. Coval, a company running roughly 3.1 million voice agent evaluation metrics weekly, reports that about 95% of voice agents work in a demo while about 62% survive their first week live. That is a vendor figure from a company selling evaluation tooling, so weight it accordingly. It matches what the gap between a demo and a Saturday night looks like.

What has to be true before you automate?

The system needs to know things your dispatcher knows and nobody wrote down.

Your service area, by town, not by radius. Your after-hours pricing. Which jobs you do not take. Which equipment you will not touch. Your on-call rotation and how to reach the person on it. What counts as a real emergency and what waits until morning.

Then the escalation rules, written by you. Gas smell, no heat with an infant in the house, active flooding, sewage in the living space. Those interrupt a human immediately.

And one more thing that costs nothing. Whoever or whatever answers should be able to state your license number and confirm you carry insurance, because licensing and insurance are the trust signals that convert a stranger at 10pm more than any script.

Where does this argument break down?

Three places, and the first one is the one nobody says out loud.

When the problem is capacity, not discipline. ServiceTitan's gradient is the honest version of this: 24% booking for shops under five technicians against 59% for shops with 25 or more. That gap is not a training gap. The small operator is on a roof, in a crawlspace, or under a sink, and no amount of accountability changes that. It is also why the same analysis found that every five point improvement in booking rate is worth roughly $100,000 in additional revenue for a shop with five to fourteen technicians, achievable with less than one additional booked call per weekday. The lever is small and the number is large because the base is so low.

When the after-hours volume is genuinely thin. Some trades have almost none. If ninety days of log shows eleven abandoned after-hours calls, you have found the answer, and the answer is that this is not your constraint.

When answering more calls makes the operation worse. Booking a 1am job you cannot staff produces a missed appointment, and a missed appointment produces a review. Coverage without capacity is a slower way to lose.

What to do this week

Log into your phone provider and export ninety days of call detail. Filter to calls outside your business hours. Count three things: total, answered, abandoned.

Run the five step arithmetic above with your own booking rate and your own average ticket. That is the only number that will get this project funded in your own head.

Then call your own line at 9pm and at 7am Sunday and write down exactly what a customer hears.

Do that monthly from now on. It belongs on the short list of numbers a home services owner should watch weekly, because it moves when your staffing moves and nobody tells you.

And before you add ad budget to generate more calls, fix the hours you already have. Paying for clicks that ring into voicemail is the most expensive mistake in the category, which is worth remembering when you calculate the minimum viable ad spend for a local business.

Questions about your after-hours volume?

Email me at eric@seod.com with a ninety day call detail export from your phone provider, any format. I will count your after-hours calls, split answered from abandoned, and send you back the annualized dollar figure using your own average ticket. It takes me about ten minutes and there is nothing attached to it.

More on lead flow and phone economics sits in the home services library.