Monday, 7:10 AM. An electrician scrolls his phone in the truck: eleven missed calls since Friday afternoon, four voicemails. Two are the supply house, one is a robocall, one is a customer asking about an invoice. The other seven numbers left nothing. Were they jobs? He'll never know unless he counts.
Industry-wide figures about missed calls are easy to find and hard to trust, and none of them describe your phones. Your own call log, your close rate and your job profit give a number you can act on, and one you'll believe.
Skip the math: use the missed call calculator.
The formula
- Unrecovered new-customer calls: distinct people who weren't already customers, whose call went unanswered, and whom you didn't reach and book afterward.
- Close rate: of the new-customer calls you do answer, how many become a booked job. Use your own records, even rough ones.
- Gross profit per first job: what's left after materials and direct labor, not the invoice total. Using revenue overstates the loss.
Repeat work, referrals and the service plan a new customer might join are real, but keep them out of the first pass. A conservative number you trust is more useful than a big one you don't.
Step 1: pull and clean your call log
Most business phone systems let you download call history: date, time, caller number, duration and whether the call was answered. If the business runs on a mobile phone, be careful: a carrier's usage log often lists only calls that connected, including calls that went to voicemail, and may not show calls that rang out or hung up early. Check that missed calls actually appear before you trust it; if they don't, use the phone's own recent-calls list or move the line to a system that reports them. Take four full weeks, so one strange day doesn't skew the result.
Then clean it. A raw count of unanswered calls overstates the problem, and this is where most guesses go wrong.
Flowchart
Sorting one missed call in your log
- StartA number shows as unanswered in your four-week export
- DecisionDid the same number call again within the hour?YesStepMerge the calls: one person, one opportunityNo
- DecisionRobocall, vendor or wrong number?YesStepDrop it from the countNo
- DecisionAlready a customer?YesStepTrack separately: a service problem, not lost new revenueNo
- StepCount it as a missed prospect
- DecisionDid you reach and book them within a day?YesA personRecovered by your callback; note how long it tookNo
- DoneIt goes into the formula as an unrecovered prospect
- Match numbers against a customer export from your invoicing or field-service software to mark existing customers quickly.
- Search a handful of unknown numbers that left no message. Many turn out to be businesses or known spam; the rest are probably people.
- Mark a prospect recovered only if you actually spoke and booked. A returned call that went to their voicemail doesn't count.
- If your system has a queue or auto-attendant, check its abandoned-call report too. A caller who hung up on hold can be logged as answered.
Step 2: run the numbers on a worked example
Picture a two-van electrical shop. Here's how its four weeks might break down. The numbers are hypothetical and round; swap in your own.
| Line | Example shop | Yours |
|---|---|---|
| Unanswered calls in four weeks | 70 | [ ] |
| Minus repeat dials from the same number | −12 → 58 | [ ] |
| Minus robocalls, vendors and wrong numbers | −18 → 40 | [ ] |
| Minus existing customers (track separately) | −15 → 25 | [ ] |
| Minus prospects reached and booked on callback | −9 → 16 | [ ] |
| = Unrecovered prospects | 16 | [ ] |
| × Close rate on answered new-customer calls | × 0.5 (1 in 2) | [ ] |
| × Gross profit per first job | × $450 | [ ] |
| = Monthly cost of missed calls | about $3,600 | [ ] |
That leaves 16 unrecovered prospects. The shop books about one in two new-customer calls it answers, and a typical first job leaves around $450 in gross profit after materials and labor.
Hold that figure loosely. Not every unrecovered caller would have booked at your usual rate: someone calling at 9 PM about a dead outlet may be more motivated than average, or just shopping. And the prospects you recovered still cost you something in phone tag.
The plumbing version
For a plumbing shop, run the formula twice: once for emergencies and once for routine work. A burst pipe at 10 PM and a dripping faucet carry very different job values and close rates, and averaging them hides where the money actually goes. If most of the lost value sits in a few after-hours emergencies, that points straight at night coverage.
The restaurant version
Restaurants use the same idea with different inputs. Say a 70-seat place misses about 20 calls between 6 and 9 PM each weekend, and the host's best estimate is that half were people trying to book. Over a month that's around 40 lost booking attempts. At three guests per booking and $45 spent per guest, that's about $5,400 in sales, which you'd then cut down to your margin and to the guests who would have walked in anyway.
Step 3: find the hours you miss most
The total tells you whether to act. The timing tells you how. Drop the unrecovered prospects into a simple grid by time of day.
| Time of day | Weekdays | Weekends |
|---|---|---|
| 7–9 AM, crews loading up | 4 | 0 |
| 9 AM–noon | 1 | 1 |
| Noon–1 PM | 3 | 0 |
| 1–5 PM | 1 | 1 |
| After 5 PM | 3 | 2 |
In this example, evenings account for 5 of the 16, but 7 fall in just three weekday hours: the two when every truck is loading up and the lunch hour, when nobody is near the office phone. That changes the fix. Covering those short daytime windows is a smaller, cheaper decision than round-the-clock answering, and it may be worth doing first, with evening coverage judged on its own numbers.
Before and after
Guessing vs. counting
Going by gut
- 01Count the voicemails, not the calls
- 02Assume anyone who left no message wasn't serious
- 03Treat every missed call as a lost job, or none of them
- 04Pick phone coverage from a vague sense of when it's busy
ResultA number you can't defend in either direction
Counting from the log
- 01Export four weeks of calls with times
- 02Merge repeats, drop spam, set existing customers aside
- 03Apply your own close rate and gross profit
- 04Map unrecovered prospects by hour and day
ResultA monthly dollar figure and the exact hours to cover
What to do with the number
- Small number, scattered hours: tighten your callback habit before buying anything. The speed-to-lead guide covers how to make fast replies routine.
- Bunched into a few windows: cover those hours. No-answer forwarding to a colleague or an assistant during the morning rush and lunch is a small change; here's how forwarding works.
- Large and spread over evenings and weekends: after-hours coverage is the real decision. Compare options by what they finish on the call, not just whether they pick up, as in AI receptionist vs. answering service.
- Lots of existing customers in the pile: that's a service gap, not a sales one. They're often asking about schedules or invoices, which a quick text update could answer before they ever call.
Whatever you change, run the same count a month later with the same rules. If the unrecovered number doesn't fall, the fix isn't working, whatever a dashboard says.
Mistakes that skew the count
- Using invoice totals instead of gross profit.
- Measuring only a slow month, or only your busiest one. A plumber's January and June can look like different businesses.
- Forgetting calls that ring a second line or someone's personal cell.
- Counting a returned voicemail as a recovered customer.