Most home-services operators we talk to estimate they miss "maybe 10–15% of calls." When we measure, the real number is almost always more than double that. We wanted to know what the real distribution looked like, so we pulled 90 days of inbound call logs from 312 shops running IzzyOps in Q1 2026 and ran the numbers.
The headline: the average shop misses 38% of inbound calls. The median is 34%. The 90th percentile shop misses 62%. The shops with the best phone coverage still miss 11%.
1. The 5pm cliff is real
Call volume in home services peaks between 8am and 11am, dips at lunch, runs flat until 5pm, and then drops sharply. But missed calls do the opposite — they spike after 5pm and stay elevated through the night. 71% of missed calls happen between 5pm and 8am. Not surprising, but the magnitude is.
Saturday evening is the single worst window — 14% of weekly missed calls happen between 6pm Saturday and 8am Sunday. Some of these are casual enquiries that wouldn't have converted anyway. Many are emergencies, where caller intent is the highest you'll see all week.
2. After-hours calls convert at 1.6× the daytime rate
This is the one that surprised us. Calls answered after 5pm convert to booked jobs at 1.6× the rate of daytime calls in the same shops. The intuition is straightforward: people who call you at 9pm on a Tuesday already have a problem they need solved. Daytime callers are price-shopping, comparing, and generally less committed.
If you assume after-hours callers convert at the daytime rate, you under-count the cost of voicemail by about 60%.
3. The cost looks like this
For a median HVAC shop running ~520 inbound calls/month with an average ticket of $420 and a daytime close rate of 38%:
- Missed calls: 177/month (34%)
- Daytime-equivalent lost revenue: $28,260/month
- Adjusted for after-hours conversion lift: $36,400/month
- Annual: $436,800
For Plumbing the number is higher because ticket sizes run larger; for Electrical it runs lower because emergency volume is smaller. Across the cohort, the median lost-revenue figure is $31,000/month.
4. The shape of "what was the call about"
We classified the first 60 seconds of each missed-then-recovered call (i.e. calls IzzyOps picked up after a previous voicemail attempt):
- 49% emergency repair (active leak, no AC in summer, no heat in winter)
- 22% appointment booking for non-urgent service
- 14% follow-up on a previous visit or quote
- 9% price enquiry
- 6% wrong number, sales calls, miscellaneous
The takeaway: 71% of missed calls are revenue-or-relationship critical. Only 6% are noise.
5. Voicemail conversion is worse than you think
We tracked what happens to a missed call that goes to voicemail. Of callers who leave a voicemail (only 28% do), 41% get a callback within 24 hours. Of those callbacks, 19% reach the caller live. Of those live reconnections, 47% book. Multiply it out: 1.1% of missed calls turn into bookings via the voicemail-callback path.
Compare that to a live answer, which books at 38%. Voicemail isn't a backup plan. It's a slow leak.
What this means in practice
Three things, in order of impact:
- Fix the after-hours hole first. If you can only do one thing this quarter, get coverage from 5pm to 8am, all seven days. The math on that single window is bigger than any marketing campaign you'll run this year.
- Stop using voicemail as a buffer. Either you answer live or you have a system that books on the first attempt. The voicemail-callback loop costs you 90%+ of those callers.
- Measure your real miss rate. Whatever you think it is, double it, then check. Most call-tracking software under-counts because it can't see calls that ring out without leaving a voicemail.
If you want the underlying dataset (anonymised, by industry and shop size) drop us a note. We're happy to share — the goal is to make this less of a black box for the whole space, not to keep it as a sales prop.
Methodology
312 IzzyOps customers in the home-services category, US + AU, Q1 2026. Anonymised call logs, ANI removed, only the call metadata (timestamp, duration, outcome, intent classification) used. Conversion data joined from the customers' DMS (ServiceTitan, Housecall Pro, Jobber) where connected.