Missed Call Math for Contractors: What Each Unanswered Call Actually Costs

On October 1, 2026, Google began charging Local Services Ads advertisers for calls they did not answer. That turns missed call math from a motivational statistic into a line item: per Google's lead charge documentation, a missed call during your business hours is billed as a valid lead if the caller stays on the line more than 20 seconds. A voicemail greeting counts as a connected call.

A miss now costs twice: the job you did not book, and the lead you paid for anyway. Most contractors have no real number for either, just a shrug or a figure from a vendor selling answering software. Neither tells you whether fixing the phone is worth what fixing it costs.

What follows is the arithmetic: a formula, benchmarks to check your answer rate against, a worked annual model with every assumption stated, and the breakeven number of jobs. Invoca's Home Services Lead Conversion Benchmarks Report, published July 2026 from more than 70 million calls across ten industries, found that 52% of inbound home services calls reach a person — vendor data averaged across their customer base, with the home services sample undisclosed, so hold the precision loosely.

Table of Contents

1. Missed Call Math: The Formula and Where Every Number Comes From
2. A Worked Example at a Hypothetical HVAC Company
3. What the October 2026 Google LSA Change Does to the Math
4. Where the Math Misleads, and What Not to Automate
5. A 90-Day Plan to Move the Number
6. Frequently Asked Questions
7. Ready to Put a Real Number on Your Missed Calls?

Missed Call Math: The Formula and Where Every Number Comes From

Missed call math is six multiplications, and the discipline is in the subtractions you make first. The answer you want is lost gross profit, not lost revenue, because gross profit is what a fix has to beat. Work it in this order:

1. Total inbound calls. From call tracking or carrier call detail records, not from memory.
2. Minus non-opportunity calls. Misdials, robocalls, suppliers, existing customers rescheduling, your own technicians. Almost everyone skips this step, and skipping it inflates everything downstream.
3. Times the miss rate on what is left: the share that never reached a person.
4. Minus the ones you recovered. A voicemail you returned, a caller who tried again, a form submitted later. A miss is only a loss if the customer never came back.
5. Times the share that were genuine new-job opportunities. Invoca puts the lead rate, the portion of answered home services calls that are real leads, at 38%.
6. Times your booking rate, then average ticket, then gross margin. Invoca's on-call conversion rate for home services leads is 45%; use your own if you track it.

Two rules keep this honest. Count gross profit, since a $600 job at 40% margin is $240 of decision-relevant money. And treat every input as a range: if the fix pays at your pessimistic end, the decision is made.

On step three, your answer rate is almost certainly lower than your office believes, and the gap between trades is enormous. In the Invoca July 2026 data, the share of inbound calls reaching a person runs from 32% in construction and 34% in HVAC up to 73% in pest control and 74% in plumbing, with restoration, doors and windows, lawn care, home security and carpeting between 51% and 65%. The headline 52% overstates it, and the report supplies the correction: look only at calls lasting more than 15 seconds, which screens out misdials and instant hang-ups, and the rate rises to 65%; past 30 seconds it is 73%. Use the 15-second figure as your benchmark, since it approximates the denominator you care about — people who meant to call and waited.

Only one of the three usual causes is a staffing problem. Concurrency means two calls arriving at once with one person to take them; routing means a phone tree feeding a mailbox nobody has opened since spring. Both are configuration, so fix them before buying anything. Coverage is the real gap. Demand is not: Jobber's Home Service Economic Report for Q2 2026, drawn from more than 100,000 service businesses, reported contracting revenue up 6.4% year over year in June.

A Worked Example at a Hypothetical HVAC Company

Consider a hypothetical six-truck residential HVAC company taking 4,200 tracked inbound calls a year. On conservative inputs, unanswered calls cost it about $26,000 in gross profit annually, and a fix needs to save only three jobs a month to pay for itself. Substitute your own numbers for these:

• 4,200 inbound calls, about 350 a month.
• 22% non-opportunity calls, or 924, leaving 3,276 opportunity calls.
• 65% answer rate, the Invoca over-15-second benchmark: 2,129 answered, 1,147 unanswered.
• 45% of misses recovered via returned voicemail or a second attempt: 631 lost for good.
• 38% were genuine new-job leads, Invoca's lead rate: 240 lost leads.
• 45% would have booked, Invoca's on-call conversion rate: 108 lost jobs.
• $540 average ticket at 45% gross margin, so $243 gross profit per job.

That is $58,320 in lost revenue and $26,244 in lost gross profit, from 108 jobs out of 3,276 opportunity calls — just 3.3% leakage, which is why the conservative version persuades and the scary version does not.

Now the decision. A tracked line with proper ring groups and overflow, plus after-hours coverage, lands most small contractors between $6,000 and $12,000 a year all in. At $9,000 and $243 of gross profit per job, breakeven is 37 jobs a year, about three a month. Against 108 lost jobs, that is a clear yes.

At lower numbers it flips. A handyman at a $300 ticket and 35% margin earns $105 per job, so the same $9,000 needs 86 recovered jobs; if the model shows only 40 lost jobs, fix routing for free and buy nothing. Ticket mix decides it — Angi's 2025 State of Home Spending Report put average household spend at $12,472 across about ten projects.

What the October 2026 Google LSA Change Does to the Math

If you advertise on Local Services Ads, add a seventh term: leads you are billed for and never spoke to. That is money leaving, not opportunity cost, and it appears on a statement.

Google's documentation sets out the mechanics precisely:

• A missed call is charged if the caller holds more than 20 seconds inside the business hours on your Google Business Profile, and where your ad schedule differs, only calls inside both windows count.
• A voicemail greeting counts as a connected call, whether or not a message is left.
• In a phone tree the 20-second timer starts at the first key press, so a caller who hangs up before pressing anything is not billed. Follow-up contact from the same number or email is not charged again within 15 days.

PPC Land reported that the notice reached advertisers on August 24, 2026, and that Google published no dispute process; the help page points instead to the per-lead Feedback Survey.

Extend the example. Say 30% of those 3,276 opportunity calls came from LSA, so 983 calls, of which 344 go unanswered at the same 35% miss rate. Not all become charges, since the caller has to hold past 20 seconds during business hours. At 55% qualifying, that is 189 newly billable missed leads. At a cost per lead of $65, that is $12,285 a year for calls nobody picked up, taking total exposure to roughly $38,500.

Two configuration changes are worth making this week. Tighten Business Profile hours to the hours you genuinely answer, because hours you cannot cover are now a billing window. And reconsider the voicemail greeting on your LSA number, which converts a free miss into a paid one.

Where the Math Misleads, and What Not to Automate

The most common error is not pessimism, it is double-counting, and it discredits your case the first time someone audits it. Four traps, each with a check.

Double-counting the same customer. One homeowner calling three times in an afternoon is one opportunity, not three. Check: deduplicate by phone number inside a rolling seven-day window.

A denominator full of junk. Spam and misdials make your miss rate look catastrophic and your savings fictional. Check: exclude calls under 15 seconds and tag recurring supplier and robocall numbers. If step two removes less than 10% of your volume, you have not actually looked.

Treating every miss as a lost job. It is lost only if the customer never returned and would have booked. Check: sample 50 unanswered calls from last quarter against your CRM and invoicing history. Most shops find recovery well above the 45% used above.

Measuring the phone and ignoring the handoff. Invoca shows 38% of answered calls are leads and 45% of those convert, so more than half the people you reached did not book. Check: pull your booking rate on answered calls before blaming the misses.

What not to automate: pricing, phone diagnosis, anything involving gas, water, electrical or a safety risk, and any call from a customer with an open complaint or warranty claim. AI should capture, qualify and route, enhancing the judgment of whoever takes it from there rather than replacing it. Our analysis of AI phone answering for contractors covers that boundary, and home services dispatch covers what happens after the booking.

A 90-Day Plan to Move the Number

Measure for 30 days, fix what is free for 30, then buy coverage for 30. The first phase is the only thing that will later prove whether any of it worked.

Days 1–30: get the denominator right. Put call tracking on every advertised number, including LSA. Export 90 days of call detail records and hand-tag 200 calls into opportunity and non-opportunity. Record four baselines: total calls, miss rate on calls over 15 seconds, recovery rate on misses, and booking rate on answered calls. Then count how many charged LSA leads last month went unanswered. Buy nothing yet.

Days 31–60: fix routing and concurrency for free. Rebuild the ring group so a second simultaneous call rings somewhere real. Flatten the phone tree. Set Business Profile hours to the hours you actually answer. Route after-hours calls to a cell phone for two weeks and log what arrives; most shops find evening volume smaller and more urgent than assumed, which changes what they buy. Kill criterion: if the miss rate on calls over 15 seconds is already under 10% after this phase, stop.

Days 61–90: add coverage, then re-measure. Buy against the gap the baseline showed: a live answering service, an AI receptionist, or one part-time person covering 5 to 9 p.m. Every captured call must land in your scheduling system with name, address, problem and callback number, because a recorded message nobody acts on is not coverage. At day 90, re-run the four baselines and the model. If the miss rate fell but booked jobs did not, the leak is in the handoff, which our overview of AI for home services maps.

Frequently Asked Questions

Ready to Put a Real Number on Your Missed Calls?

A number built from your own call records beats any benchmark, and it turns an argument about the phone into a decision with a breakeven you can check in ninety days. Process first, technology second: the free routing fixes usually move the number further than software does. ANtreliX is a strategy-first AI and business intelligence consultancy based in North Myrtle Beach, South Carolina, working with home services companies across the Grand Strand and remotely nationwide. See our case studies or Talk to ANtreliX.