Disclosure: Futuro Corporation publishes this page and sells an AI receptionist, Futuro AI. Every statistic below is traced to its origin and graded on a three-tier scale — including our own first-party number, which gets the same treatment as everyone else's. No category claim on this page is sourced to a company selling in this category, and vendor data is labeled as vendor data.
The voicemail is where leads die
Every small business runs the same quiet experiment thousands of times a year: the phone rings, nobody answers, and the caller decides what happens next. The comforting story owners tell themselves is that voicemail catches the overflow — the customer leaves a message, you call back, job saved. The data says that story stopped being true at least a decade ago.
Here is the actual chain, each link of which gets a full stat card below. First, the call goes unanswered: in the only structured field study of small-business phones, 62% of calls were never answered by anyone. Second, the caller hits voicemail and exercises their veto: in home-services platform data spanning tens of millions of conversations, fewer than 3% leave a message. Third, the caller — who is usually standing in front of a problem they want solved today — dials the next name in the search results, and whoever responds first inherits the job: firms responding within an hour were nearly seven times more likely to qualify the lead in Harvard Business Review's audit of 2,241 companies.
None of this means the phone is dying. The opposite: the phone call is the highest-intent action a local customer takes — 70% of mobile searchers call a business directly from search results (Google/Ipsos, 2013), and 38% of answered business calls are leads (Invoca, 2026). What died is the message-taking step. Customers still call. They just won't leave a voicemail — and in 2026, they're usually calling from a search results page with three more names under yours.
The core numbers, sourced
Each card below is self-contained: the number, the claim it actually supports, the real source with its year, and our confidence grade. Grades are explained in the provenance section — Well-sourced means an identifiable study with methodology, sample, and date; Vendor-circulated means named attribution but no locatable primary; Unverifiable means we could not find an origin at all. Cite the well-sourced ones freely. Treat the rest accordingly.
The only structured field study of small-business phones: 411 Locals "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." Honest limits: it was run by a marketing vendor, the sample is 85 businesses, and it's a decade old — but it has a stated methodology and a publication date, which is more than most stats in this niche can claim.
Call-analytics platform Invoca reported in 2024 that "our platform data shows that less than 3% of callers who get pushed to voicemail leave a message" — and that 27% of calls to home-services businesses go unanswered in the first place. Vendor platform data, labeled as such — but drawn from call recordings across its customer base, which is the closest thing to a direct measurement of voicemail abandonment that exists.
Invoca's 2026 benchmarks, built from more than 70 million phone conversations, put the average business answer rate at 56% across industries and 52% in home services — meaning a typical contractor misses nearly half of all inbound calls during measured hours. See also the full 2026 report.
The reason missed calls matter more than missed emails: over a third of the calls businesses do answer are sales leads, per Invoca's 2026 data, and home-services leads convert at 45% once reached. The unanswered half of your phone isn't admin noise — it's the highest-converting channel you have, ringing out.
The foundational call-intent study: Google and Ipsos surveyed 3,000 mobile searchers and found 70% have called a business straight from search results; 61% called click-to-call extremely or very important at the moment of purchase; 47% described feeling frustrated or annoyed — and more likely to explore other brands — when a business offered no way to call. Old, but never contradicted, and it explains why the missed-call problem starts at the search page.
The original "five-minute rule" research — often misattributed to Harvard. Analyzing three years of data across six companies (15,000+ leads, 100,000+ call attempts), the study found "the odds of contacting a lead if called within five minutes versus even 30 minutes are 100x greater," with qualification odds up 21x. The data covers web-generated leads, not inbound phone calls — but it established the steep decay curve every later study has confirmed directionally.
Oldroyd, McElheran, and Elkington audited how 2,241 U.S. companies actually respond to leads for "The Short Life of Online Sales Leads": firms that tried to contact potential customers within an hour were "nearly seven times as likely to qualify the lead" as those that tried even an hour later — "and more than 60 times as likely as companies that waited 24 hours or longer." The same audit found an average first response of 42 hours, and that 23% of companies never responded at all.
The current-expectation baseline, from Jobber's 2026 report: "Over 70% of customers expect a same-day response—more than half expect it within the hour." A vendor survey with a disclosed sample, and the modern confirmation that the speed-to-lead curve from 2007–2011 has only steepened since.
This is the most-quoted number in the answering-service industry — and no one can produce it. It appears on hundreds of vendor blogs attributed to "AT&T research," sometimes to market-research firms, sometimes to no one. No AT&T study, press release, white paper, or dataset containing this finding is locatable; one source-tracing effort followed the citation chain back to an uncited 2016 vendor blog post. It may be directionally consistent with Invoca's measured <3% voicemail rate — but as a citable statistic, it doesn't exist. Don't build a business case on it.
Contractor-marketing blogs attribute this to ACHR News, to "Lead Connect," or to Contractor Magazine, depending on which one you read. We could locate no primary study from any of them. The direction is well supported by the sourced research — HBR's 7×/60× finding and the 2007 100× study both say the first responder wins disproportionately — but the specific "78%" figure has no locatable origin. Cite HBR instead; it says nearly the same thing with a methodology attached.
Widely repeated in HVAC marketing, never produced. No ACHR News article containing this measurement is locatable. What is well-sourced: Jobber's finding that most customers expect same-day or within-hour responses regardless of when they call, and the structural fact that the 411 Locals study measured business hours — nights and weekends are when answering coverage is thinnest, not when demand stops.
The "one in five calls" figure traces — through Numa's statistics roundup — to a 2016 statement by UK phone-services company Liquid11, whose CEO referenced data from roughly 10,000 business customers. A named company and a stated sample put it a tier above pure folklore, but the primary release is no longer locatable, so it stays vendor-circulated. Note it also conflicts with the 411 Locals field measurement of 62% — treat both as bounds, not truth.
Three numbers that travel together through contractor-marketing content, none with a findable origin. No study has measured how long an emergency caller waits before dialing the next contractor; the dollar figures trace to answering-service marketing math with no published methodology. The behavioral claim is plausible — nobody with water pouring through their ceiling narrates it into a voicemail box, which is exactly what Invoca's <3% finding implies — but the specific numbers are invented. We print them here only so you know where not to cite them.
Same rubric, applied to ourselves: this figure comes from Futuro Corporation's own trades call data (see TradeWork for the product context). It is first-party vendor data — collected by a company that sells a solution to the problem it measures. We publish the methodology and label it everywhere we use it, but it is not independent research and shouldn't be treated as such. We'd rather you grade it harshly than trust it blindly; that standard is the whole point of this page.
Where these stats actually come from
Most "missed call statistics" pages are the same twenty numbers copied in a circle, each site citing the one before it. We did the boring part: followed every number above back to its origin and graded what we found. Three tiers:
- Well-sourced — an identifiable piece of research with a methodology, a sample, and a publication date. It may still be old, vendor-run, or narrow — we say so on the card — but you can read it yourself and check our math.
- Vendor-circulated — a named attribution exists (a company, a publication), but the primary source can't be located. Usable as directional context; never as the load-bearing number in an argument.
- Unverifiable — no locatable origin at all. The citation chain dead-ends in other people's blog posts. These numbers may still be true — they just aren't sources.
The pattern that emerges: the stats with real origins are mostly boring and specific (52% answer rate, 42-hour average response, fewer than 3% leaving messages), while the dramatic round numbers (85%, 78%, $1,200) are the ones that evaporate. That inversion is itself useful — when a stat sounds perfectly designed for a sales pitch, it usually was.
| The stat, as circulated | Actual origin (what we found) | Grade |
|---|---|---|
| 62% of small-business calls unanswered | 411 Locals field study, 85 businesses / 58 industries / 30 days, published Jan 18, 2016 | Well-sourced |
| <3% of callers leave a voicemail | Invoca platform data, published May 23, 2024 | Well-sourced |
| 52% home-services answer rate / 38% of answered calls are leads / 45% conversion | Invoca Lead Conversion Benchmarks, 2026 — 70M+ conversations | Well-sourced |
| 70% of mobile searchers call from search | Google/Ipsos survey, 3,000 mobile searchers, Sept 2013 | Well-sourced |
| 100× / 21× at five minutes vs. thirty | Lead Response Management Study, Oldroyd, MIT/InsideSales.com, 2007 | Well-sourced |
| 7× / 60× within the hour; 42-hour average | Harvard Business Review, March 2011; audit of 2,241 companies | Well-sourced |
| 70%+ expect same-day response | Jobber 2026 Home Service Trends Report, 1,000+ owners | Well-sourced |
| 85% of callers won't call back | "AT&T research" — no AT&T study locatable; chain dead-ends in vendor blogs | Unverifiable |
| 78% hire the first responder | Attributed to ACHR News / Lead Connect / Contractor Magazine; no primary located | Vendor-circulated |
| 62% of HVAC calls come after hours | Attributed to ACHR News; no primary located | Vendor-circulated |
| 22% missed-call rate | Liquid11 (UK), 2016, ~10,000 customers; primary no longer online | Vendor-circulated |
| 3-minute emergency window; $1,200 per missed call; $126,000/year | No locatable origin; vendor marketing math | Unverifiable |
| 78% try a competitor next | Futuro AI first-party trades call data, 2026 — graded like the rest | First-party |
One number deserves a footnote of its own. The 411 Locals study is the best structural measurement of small-business answering ever published — and it is a decade-old, vendor-run study of 85 businesses. We grade it well-sourced because it has a stated methodology and date, not because it's definitive. If a university lab ever replicates it at scale, we'll replace it the same week; the corrections policy covers exactly that case.
Why voicemail died
Voicemail didn't lose an argument; it lost a generation. The decline is one of the best-documented behavioral shifts in consumer communication — it just happened to everyone except the small-business phone setup.
The timeline, in sourced milestones. In March 2012, Pew Research found that "63% of all teens say they exchange text messages every day" while "26% of all teens ... talk daily with friends on their cell phone, down from 38% of teens in 2009." That September, the Christian Science Monitor reported Vonage's figures, prepared for USA Today: "the number of voice-mail messages for Vonage users was down 8 percent this July compared to the previous July," and retrievals — people even listening — fell 14% over the same period. By 2014, when Coca-Cola offered employees the chance to switch voicemail off, HuffPost reported "just 6 percent of workers opted to keep getting voice mails," and cited an eVoice survey finding only a third of workers listen to voicemails from business contacts at all.
Then the texters grew up and became your customers. BankMyCell's 2019 survey of 1,200+ Americans aged 22–37 (field dates confirmed by BGR's coverage) found 81% feel "apprehension anxiety" before making a call and 75% avoid calls as time-consuming. Those respondents are 29–44 years old today — the core booking demographic for home services.
Hold that against the paradox from the first section: the same consumers who won't leave a voicemail do still call businesses — 70% of mobile searchers have called straight from search results (Google/Ipsos), and 38% of answered calls are leads (Invoca). The customer behavior is not "don't call." It's "call, and if a person doesn't pick up, hang up and call the next one." Voicemail asked callers to perform effort on faith that someone would call back; texting trained them to expect a response in minutes; and search results made the next option one thumb-flick away. The message-taking step simply had no defenders left.
The emergency multiplier
Everything above gets sharper in the trades, because a meaningful share of trade calls are emergencies — and an emergency compresses the caller's patience from hours to minutes.
Consider the difference in caller psychology. A homeowner booking a quote for a bathroom remodel might leave a message and wait a day. A homeowner with a burst pipe at 9 p.m. is calling down a list, and the first live voice wins a job that routinely bills at a premium. Nobody has measured the exact "emergency window" — the famous three-minute figure is unverifiable, as its card above says — but the direction is not in doubt: Invoca's finding that under 3% of home-services callers leave any message at all was measured across ordinary calls. Urgency only pushes that number down.
The sourced pieces of the puzzle: most home-service customers expect a same-day response and more than half expect one within the hour (Jobber, 2026); speed-to-lead decay is brutal and immediate (HBR 2011; MIT/InsideSales.com 2007); and in our own first-party trades data — graded first-party, as noted on its card — 78% of callers who can't reach a trades business try a competitor next. The uncomfortable arithmetic: after-hours and overflow calls are precisely where emergency jobs concentrate, and precisely where small-trade answering coverage is thinnest. The 62% unanswered figure from 411 Locals was measured during the hours businesses were supposedly open.
If you remember one sentence from this section: the calls you miss are not a random sample of your demand — they're biased toward your most urgent, highest-value work.
What to do about it
This page is a data hub, not a pitch — so the action plan is short, and it works even if you never buy anything from anyone.
- Measure your own number first. Pull two to four weeks of call logs. Count calls no one answered and hang-ups under ten seconds. That's your missed-call rate, and it beats every statistic on this page — including ours — because it's yours. The walkthrough is in our contractor missed-call analysis.
- Know your speed-to-lead exposure. If your average callback takes longer than an hour, the HBR and MIT numbers above describe your funnel. The speed-to-lead breakdown maps the thresholds.
- Decide who answers every call. A person, a human answering service, or an AI receptionist — the honest comparison of what each costs and covers is in our vendor pricing index (69 vendors, re-checked weekly). Futuro AI is $200/month flat, every industry, which you can verify on the pricing page. Whatever you choose, test it the way a customer would: call your own business at 8 p.m. on a Saturday.
The canonical question behind all of it — can AI answer my business phone? — has its own plain-English guide. And if you want to skip the reading: call 855-490-5531 and let Futuro AI answer you the way it would answer your customers, in any of 53 languages. In the Futuro Corporation Human Indistinguishability Double-Blind Study (1,000 participants), 94% couldn't tell they were speaking with an AI; 3% were unsure and 3% identified it correctly.