About this guide: This is an evidence audit and a pre-registered study plan, not a vendor comparison. It traces every HVAC missed-call statistic we could find to its origin, grades each one, and sets out, before any data is pulled, how Futuro Corporation will measure HVAC call handling from its own client data. No vendors are ranked. The only price on the page is Futuro AI’s own, and like every price we publish it has a dated screenshot on our vendor pricing index (see the weekly pricing note). External sources are linked at the claim.
Pricing for every vendor we cover is checked weekly, and a dated screenshot of the pricing each vendor publishes is updated every Friday at 9 AM ET. We keep these on one public page so you can see exact, current pricing for each vendor: Vendor pricing index.
Has anyone actually measured how many calls HVAC companies miss?
Ask ten HVAC marketing blogs how many calls a shop misses and you get one number repeated ten times. The figure changes depending on who's selling — 62% of calls come after hours, 22% go unanswered during business hours, 30–50% miss rate in peak season, 85% of callers won't call back, $45,000 to $120,000 lost every year — but the structure never does: a confident number, a vague attribution, and no link to anything you can actually open and read.
We tried to trace every one of these numbers to a primary source. The full audit is below, but the short version is simple: the HVAC phone-stat ecosystem is an echo chamber of AI-answering and missed-call-tracking vendors citing each other, sometimes with a real-sounding attribution ("ServiceTitan, 2024," "ACHR News, HVAC Call Pattern Analysis," "multiple home services studies") attached to a study nobody can produce. We could not locate a single primary dataset behind any of the six famous claims.
This matters because the numbers aren't decoration. Shops make five-figure staffing and software decisions on them. A stat that can't survive a source check shouldn't be setting anyone's budget — including ours.
The audit: six famous numbers, six dead ends
| The claim | Where it circulates | Claimed origin | What we found | Grade |
|---|---|---|---|---|
| "HVAC companies lose $45,000–$120,000 a year to missed calls" | AI-answering vendor blogs, agency posts, LinkedIn round-ups | "Data from 1,200+ contractors (ServiceTitan, 2024)" | No study, report, or dataset with these figures is locatable; the attribution appears only on sites selling missed-call solutions | Vendor-circulated |
| "22% of calls are missed during operating hours" | Missed-call-textback vendor blogs | Attributed to a voicemail-software vendor's blog listicle | A number on a vendor blog, citing no methodology; the underlying page offers no sample or date | Vendor-circulated |
| "62% of HVAC calls come after hours" | AI-receptionist vendor blogs and sales decks | "ACHR News, HVAC Call Pattern Analysis" | No such analysis is locatable. The 62% figure matches a different real study — 411 Locals, 2016 — which measured something else entirely (below) | Vendor-circulated |
| "HVAC shops miss 30–50% of calls in peak season" | AI-answering vendor blogs | Variously "ServiceTitan," "industry data," or nothing | The number mutates by site — 27%, 30%, 32%, 35–45%, "exceeds 70%" — often on the same domain. Mutation is the signature of a stat with no source | Vendor-circulated |
| "85% of callers who don't reach a person won't call back" | Answering-service and AI-vendor blogs industry-wide | "Multiple home services studies" / attributed to AT&T | No study locatable at any attribution; the AT&T version has circulated for decades without a document behind it | Unverifiable |
| "78% of customers hire the first contractor who responds" | Vendor blogs — and, until now, our own marketing | Unnamed “industry research” | No primary study found. We grade our own usage first-party and flag it the same way | First-party |
The one true origin story: how 62% became "after hours"
The 62% figure is the most instructive failure, because there is a real study at the bottom of it — just not the one being claimed. In January 2016, 411 Locals, a local-SEO agency, published a field study: "we monitored the phone calls of 85 businesses, operating in 58 industries, for a period of 30 days" and found "37.8% of calls do get answered, another 37.8% get forwarded to voicemail and 24.3% don't get any response," according to 411 Locals (January 18, 2016). Add the last two buckets: 62.1% of calls were not answered live — across all hours, across 58 industries, at 85 small businesses, a decade ago.
Watch what the echo chamber did with that. "Not answered live" became "unanswered." "85 businesses in 58 industries" became "HVAC." And "all hours" became "after hours" — a claim the original study never made and its data could not support. Each hop made the stat more specific and more useful to whoever was selling phone coverage, and less connected to anything that was ever measured. That's how a real number becomes a fake one without anyone technically inventing it.
What we can stand on while the benchmark is built
The honest inventory isn't empty. Vendor platform data — published, defined, and drawn from tens of millions of real conversations — gives a defensible floor. "Our research found that 27% of calls to home services businesses are not answered," according to Invoca's platform data (May 2024) — all seasons, all home services, explicitly not HVAC-specific. Invoca's 2026 benchmarks, drawn from "over 70 million calls and 600 million minutes of phone conversations," found 52% of home-services callers speak with a person, according to the Invoca Lead Conversion Benchmarks Report 2026. And the market these numbers describe is enormous: 440,900 HVAC mechanics and installers held U.S. jobs in 2025, with employment "projected to grow 11 percent from 2025 to 2035," according to the U.S. Bureau of Labor Statistics, spread across 111,207 plumbing and HVAC employer establishments per County Business Patterns 2023 at the U.S. Census Bureau.
What's missing is the layer in between: HVAC-specific, season-aware, definition-exact measurement. That's the layer this study builds.
What will the benchmark measure, and how does it define a missed call?
A benchmark is only as good as three things: what it measures, how it defines a miss, and how honestly it states its limits. All three are locked here, in public, before the data pull runs — the pre-registration commitment described in the appendix.
The dataset
The study measures anonymized, aggregate call-platform data across Futuro AI’s trades client base, restricted to clients whose primary trade is HVAC. No call recordings, transcripts, caller identities, or business identities are used; the aggregates below are computed from call metadata (timestamps, durations, dispositions) and published only as distributions and shares, never per-client figures.
The definitions
Most quoted stats fail before they start because "missed call" means whatever helps the pitch. This study separates three outcomes, each with a different cause and a different fix:
- Unanswered — the call rings without pickup until the carrier or phone system ends it, or hits a full queue and never reaches a greeting. Cause: nobody available. Fix: coverage.
- Abandoned — the caller hangs up while ringing, on hold, or in a menu. Cause: patience exhausted. Fix: speed and queue design.
- Voicemail termination — the caller reaches voicemail and hangs up without a message. Cause: callers treat voicemail as a dead end — "our platform data shows that less than 3% of callers who get pushed to voicemail leave a message," according to Invoca (May 2024). Fix: don't route revenue calls to voicemail at all.
A call counts as answered only when the caller reaches a person or an AI agent that completes an intake. "After hours" is measured against each client's own published business hours, not a fixed 9-to-5 assumption. Emergency calls are classified from intake disposition (no-cool, no-heat, gas smell, water actively leaking), not from caller self-labels. Misdials and sub-15-second hang-ups are excluded from miss-rate denominators — the same filter Invoca applies, which lifts its all-industry answer rate from 56% to 65%, so our figures stay comparable to the published floor.
The limits — published, not buried
Publish the caveats or the study is marketing. Ours, in full: (1) The panel is self-selected — shops that adopted an AI receptionist may differ systematically from shops that didn't, in call volume, growth posture, or prior phone pain. (2) Survivorship: the panel includes active clients; businesses that churned during the window are excluded, which can inflate measured performance. (3) Trade and geography mix reflects Futuro AI's client base, not the national distribution of 111,207 plumbing/HVAC establishments. (4) Classification rules above are ours; other platforms define "answered" and "abandoned" differently, which is exactly why cross-study comparisons keep failing. Where our definitions differ from a published source's, we say so next to the figure.
Which six questions will the findings edition answer?
Six questions, each one aimed at a claim the industry already quotes. The plan is locked; the measurements land with the early-2027 data update. Every result will publish as a distribution with its sample size and definition attached — never a bare number.
1. What share of HVAC calls actually reach a person — by shop size?
The circulated claims (22% missed in-hours; 30–50% missed in peak season) never say how they vary with shop size, which is suspicious: a two-tech shop where everyone is on a roof and a 40-tech shop with a CSR team should not have the same answer rate. We measure the three-way miss rate (unanswered / abandoned / voicemail-terminated) and the answer rate, segmented by technician count.
2. What share of HVAC calls actually come after hours?
The claim under test: "62% of HVAC calls come after hours" — graded vendor-circulated above, with no locatable source and a likely origin in the 411 Locals mutation. This is the single most load-bearing stat in the AI-answering pitch, and it has never been measured against clients' actual published hours. Until our number lands, the honest position is: nobody knows, and anyone quoting a figure is guessing with confidence.
3. How big are the June and January spikes?
HVAC demand is weather-driven, so call volume should hump in the first heat wave and the first hard freeze — but by how much, and how much worse does the miss rate get inside the spike? We publish a monthly call-volume index (each month's volume relative to the annual mean) and the miss rate by month. That curve publishes as a chart with the findings edition.
4. When do the emergency calls come?
Not all missed calls cost the same. A missed no-cool call in a heat wave is a same-day job for the next shop on the list; a missed maintenance-plan question can wait. We classify calls as emergency or routine from intake disposition and publish the mix by hour — the answer to "is the after-hours risk actually an emergency risk?"
5. How many callers actually leave a voicemail?
The claim under test: "85% of callers won't call back" — unverifiable as circulated. The published floor is Invoca's platform finding that under 3% of voicemail-routed callers leave a message. We measure the voicemail leave-rate on HVAC calls specifically, plus callback behavior within 72 hours where the data allows, so the "won't call back" claim finally gets a real denominator.
6. What is a missed call worth — with published inputs?
The revenue model gets its own question because it's where the phantom stats do the most damage. Ours is a formula with named, published inputs (below), and the findings edition fills in the HVAC-specific terms: missed-call rate, lead share, and close behavior by shop size. No "$126,000 a year" headlines — a range, with the arithmetic visible.
Which HVAC phone statistics can you cite today?
Every number on this page, one card each: the figure, the definition, the sample, the date, and the grade. These are the figures you can cite today; the benchmark’s own cards publish with the findings.
Part A — citable today: the published floor
"Our research found that 27% of calls to home services businesses are not answered," according to Invoca (May 2024). Scope matters: all home services, all seasons — not HVAC-specific, not peak-season. This is the real figure the "30–50% peak" mutation appears to be inflated from.
"Our platform data shows that less than 3% of callers who get pushed to voicemail leave a message," according to Invoca (May 2024). The voicemail box is not a safety net; routing a revenue call to voicemail is statistically indistinguishable from missing it.
Across all industries, "56% of calls to businesses are answered by a person," rising to 65% when misdials and sub-15-second hang-ups are excluded and 71% for calls over 30 seconds; home services lands at 52%, with sub-industry answer rates ranging from 32% to 74%, according to the Invoca Lead Conversion Benchmarks Report 2026. The 32–74% spread is the tell: trade and shop size move the number more than any slogan does.
"Across all industries, 38% of calls answered by a person are leads, and 42% of those leads convert on the call," with home services at 45% lead-to-appointment conversion, according to the Invoca Lead Conversion Benchmarks Report 2026. This is the published close-rate anchor our revenue model uses below.
"37.8% of calls do get answered, another 37.8% get forwarded to voicemail and 24.3% don't get any response," according to 411 Locals (January 18, 2016). Vendor-run, small, and a decade old — but the only structured field study of small-business phones anyone cites, and the true origin of the mutated "62% after hours" claim.
Median pay "$61,010 per year" in May 2025, employment "projected to grow 11 percent from 2025 to 2035, much faster than the average," and "about 40,600 openings" projected per year, according to the BLS Occupational Outlook Handbook. The labor shortage is exactly why the phone problem can't be solved by hiring alone.
County Business Patterns 2023 counts 111,207 employer establishments in plumbing, heating, and air-conditioning contracting, employing roughly 1.2 million people, according to the U.S. Census Bureau — before counting non-employer shops. That's the denominator every "HVAC companies lose…" headline is implicitly making a claim about.
The six circulated claims are graded in the audit table above; the grades are not repeated here.
The six benchmark cards that replace the circulated claims publish with the findings edition, one per question above, each with its sample and definition attached.
Why do the circulated HVAC phone stats fail?
The audit already settles some things without waiting for the data pull. Three failure modes explain every dead number on this page.
Phantom attribution. "ServiceTitan, 2024," "ACHR News, HVAC Call Pattern Analysis," "multiple home services studies" — real-sounding names attached to documents that don't exist. The tell: nobody who cites them links them, and no two citers describe the same study. (We apply the same standard to ourselves: Futuro AI's own 78% first-responder figure fails this test too, and it's labeled accordingly.)
Mutation. The 62% story above is the template: a real number with real scope limits gets re-quoted with the scope stripped, then re-quoted again with a new scope invented. Invoca's 27% — all home services, all seasons — is currently mid-mutation into "HVAC shops miss 30–50% in peak season" across the vendor ecosystem. Each hop adds specificity and subtracts truth.
Definition confusion. "Not answered live" (411 Locals), "not answered" (Invoca 2024), "didn't speak with a person" (Invoca 2026), and "missed" (everyone's sales deck) are four different measurements producing four different numbers from similar underlying behavior. Half the apparent contradiction in this niche — 14% here, 62% there — is definitions colliding, not data. That's why the definitions section of this study is longer than most studies' findings.
What the audit can't settle is what the true HVAC numbers are. That verdict is deliberately parked: the findings edition publishes each circulated claim next to its measured replacement, claim by claim, in early 2027.
How can you measure your own shop’s missed calls?
You don't need to wait for our benchmark to know your own number. You need four weeks of call logs and honest arithmetic.
The revenue model — formula first, assumptions named
Every term is yours to set except two, and both have published anchors. Lead share: "38% of calls answered by a person are leads," according to the Invoca Lead Conversion Benchmarks Report 2026. Close rate: the same report puts lead conversion at 42% across industries and 45% in home services. Illustrative example, with every input labeled: a shop taking 300 calls a month, missing 30% of them (bracketing Invoca's 27%), with 38% of those being leads, closing at 42% (Invoca's cross-industry rate, a little under its 45% home-services figure), on a $450 average ticket — that's 300 × 0.30 × 0.38 × 0.42 × $450, or roughly $6,500 a month walking to the next shop on the list. Change any input and the answer changes; that's the point of showing the formula instead of quoting a headline.
The four-week DIY audit
| Week | Total inbound | Rang out | Abandoned on hold | Voicemail, no message | Miss rate |
|---|---|---|---|---|---|
| Week 1 | — | — | — | — | — |
| Week 2 | — | — | — | — | — |
| Week 3 | — | — | — | — | — |
| Week 4 | — | — | — | — | — |
Miss rate = (rang out + abandoned + voicemail-no-message) ÷ total inbound. Then break it down by hour and day: most shops find their misses cluster — the lunch hour, the last hour of the day, Monday mornings — which turns "we miss too many calls" into a specific coverage gap with a specific fix. If your logs can't split these categories, that's the first finding: you can't manage what your phone system won't count. The analytics layer we ship with Futuro AI exists precisely to produce this table automatically — the same aggregates this study publishes at panel level (how the analytics work).
For the deeper playbook on what happens after the miss — speed-to-lead, callback windows, and the contractor-specific numbers — see our trade-contractor missed-call analysis and the inbound speed-to-lead breakdown. And if you want your own baseline without touching a spreadsheet, the trades revenue calculator runs the formula above on your inputs — now with its assumptions labeled.
What should an HVAC shop do about missed calls in each situation?
The right fix depends on where your misses happen. Eight common situations, each with the first step that fits it.
A two- or three-tech shop where the owner is on the roof
The miss rate here is set by who is free to pick up, which during a service call is often nobody. Run the four-week audit first; if misses cluster during job hours rather than after hours, the gap is daytime coverage, and forwarding unanswered calls after a set number of rings fixes more than an after-hours line would.
A shop with a CSR or dispatcher on the phones
Your misses are more likely abandoned calls than unanswered ones: callers hang up on hold while the one person on the phones is booking someone else. Count abandoned-on-hold separately in the audit. If that column dominates, the fix is a second line of coverage for overflow, not a longer greeting.
After-hours calls forwarded to the owner’s cell
Forwarding moves the miss rather than ending it. Count how many forwarded calls went unanswered or to the cell’s voicemail, where, per Invoca’s platform data, fewer than 3% of callers leave a message. If the owner is asleep for most of them, the after-hours line is a voicemail box with extra steps.
A shop already paying an answering service
Ask the service for its own logs: calls answered, average ring time, and how many messages were relayed versus booked. A message-taking service turns a missed call into a callback, and the callback still races the next shop on the caller’s list; the HBR research on lead response time is the reason that race matters.
The first heat wave or hard freeze of the season
Volume spikes are when circulated stats are least useful and your own logs are most useful. Pull the audit for the spike weeks separately from a normal week. A missed no-cool call in a heat wave is not only lost revenue: the CDC lists people without air conditioning and adults 65 and older among those at higher risk of heat-related illness. If the miss rate doubles inside the spike, plan overflow coverage for those weeks rather than staffing the whole year for them.
A shop that only wants emergency calls after hours
Decide what counts as an emergency before choosing coverage: this study classifies no-cool, no-heat, gas smell, and active water leaks as emergencies. Whoever answers after hours, person or AI, needs that rule in writing, plus a clear script for gas smells: PHMSA’s guidance is to leave the area, avoid switches and phones near the leak, and call 911 from a safe distance.
A multi-location or multi-trade company
Run the audit per location and per line. Blended miss rates hide the branch that misses a third of its calls behind the one that misses almost none. Invoca’s 2026 benchmarks show home-services sub-industry answer rates ranging from 32% to 74%, so a single company-wide figure can hide a wide spread.
A phone system that cannot export call logs
That is the first finding: you cannot manage what your phone system will not count. Most carriers can supply a call-detail record on request. Failing that, a tally sheet by the phone for two normal weeks, marking every call that rang out or hung up, gives a rough miss rate to start from.
How is this study registered, updated, and cited?
Pre-registration statement
This edition was published on November 1, 2026, before the aggregate data pull it describes. Locked at publication: the panel definition (HVAC-primary clients of Futuro AI’s trades client base), the three-way missed-call definition, the answered-call definition, the after-hours rule (client-published hours), the emergency-classification rule, the sub-15-second exclusion, the six questions, and the six reserved stat cards. Any deviation in the findings edition will be disclosed in the change log with its reason. The known limits — self-selection, survivorship, client-base trade and geography mix — are stated in full in the methodology section.
Update cadence
The findings publish with the early-2027 data update, and the study re-runs annually every January: new panel aggregates, every external source re-verified, every grade re-checked, and a dated change-log entry describing what changed. If a source we grade well-sourced is retracted or superseded, the grade changes at the next review — or immediately, if it affects a load-bearing figure. Spot an error? Email editorial@futurocorp.com; corrections are dated on the page.
How to cite this study
Copy-ready citation:
Futuro Corporation (2026). "How Many Calls Does an HVAC Company Actually Miss? Every Circulated Stat Audited, and the Benchmark Plan." Pre-registered methodology edition, November 1, 2026; findings edition early 2027. https://www.futurocorp.com/blogs/hvac-missed-call-benchmark
Cite the edition, not just the page — each January update carries its own date, and figures may be revised as the panel grows. If you cite a circulated-stat grade from this page, please cite the grade, not just the stat: the grade is the finding.
For journalists and researchers: the aggregate tables behind the findings edition are available on request at publication. We don't share per-client data at any aggregation level.
Frequently Asked Questions
Nobody has published a primary measurement of it, which is the gap this study exists to close. The most defensible proxies are call-tracking platform data: Invoca measured 27% of calls to home services businesses going unanswered, and its 2026 benchmarks found only 52% of home-services callers speak with a person. Neither figure is HVAC-specific or season-adjusted. Futuro Corporation plans to publish an HVAC benchmark from its client data in early 2027.
We could not trace it to any primary source. It circulates on AI-answering vendor blogs with attributions like "ACHR News, HVAC Call Pattern Analysis," but no such analysis is locatable. The likely origin is a 2016 field study by 411 Locals, which found 62% of calls to 85 small businesses across 58 industries were not answered live: a decade-old, cross-industry, all-hours figure, not an HVAC after-hours measurement.
This study separates three outcomes most quoted stats blur together: unanswered calls (the phone rings out or hits a full queue), abandoned calls (the caller hangs up while ringing, on hold, or in a menu), and voicemail terminations (the caller reaches voicemail, where fewer than 3% leave a message, per Invoca platform data). Each has a different cause and fix, so each gets its own line in the findings.
No. Invoca reports home services as a group, which includes HVAC alongside plumbing, roofing, and other trades, and its figures cover all seasons. Its 2026 report shows home-services sub-industry answer rates ranging from 32% to 74%, so an HVAC shop could sit well above or below the 52% average. Treat them as a floor to check your own logs against, not as your number.
Pre-registration means publishing the dataset, definitions, and analysis plan before examining the data, so the questions cannot be reshaped to fit whatever the numbers say. This edition is the pre-registration: the three-way missed-call definition, the exclusion rules, the six questions, and the limits of the client-base sample are locked in public. The findings edition fills in the measurements against that plan.
The findings are planned for early 2027, once Futuro Corporation completes the aggregate pull across its HVAC clients: answer rate by shop size, after-hours share, the seasonal call-volume curve, emergency-versus-routine mix by hour, voicemail abandonment, and the revenue model. The study then updates every January with a dated change log.
Pull four weeks of call logs from your phone system. Count total inbound calls, then count how many rang out, were abandoned on hold, or went to voicemail without a message. Divide missed by total. Then repeat the count by hour and by day; misses often cluster in specific windows, which tells you which coverage gap to fix first. The worksheet on this page walks through it.
No primary dollar figure exists; the large per-call and per-year numbers circulating online are vendor arithmetic built on unverifiable inputs. The honest model is a formula: missed calls times the share that were leads, times your close rate, times your average job value. For the close-rate term, Invoca's 2026 benchmarks give a published anchor: 42% of leads convert on the call across industries, and 45% in home services.
Treat it as a first-party claim, not a settled fact. The 78% figure used in our own marketing traces to vendor circulation, not a locatable primary study, so this page grades it with the same rubric as everyone else's numbers. What survives a source check is the direction: Harvard Business Review found firms responding within an hour were nearly seven times as likely to qualify a lead.
Cite it as: Futuro Corporation (2026). How Many Calls Does an HVAC Company Actually Miss? Every Circulated Stat Audited, and the Benchmark Plan. Pre-registered methodology edition, November 1, 2026. futurocorp.com/blogs/hvac-missed-call-benchmark. Each January update carries its own date, so cite the edition you read.