What Percentage of Callers Leave a Voicemail? (2026)
Ask the internet how many callers leave a message when your business line rings out, and you will get five different answers, each stated as settled fact. Depending on which page you land on, the share of callers who hang up without leaving a voicemail is 58, 67, 80, 85 or 90 out of every hundred. Those cannot all be right. As far as we can tell, none of them is sourced.
One disclosure before we start, because it matters here more than usual. NoMissedCall publishes this blog and sells an AI receptionist. A large, frightening voicemail-abandonment number is good for our marketing — it is the single most useful statistic anyone in our category could own. We spent a working day in August 2026 trying to find the study behind it so we could quote it properly. There isn't one. What follows is us arguing against our own commercial interest, with every dead end linked so you can check the work.
The short answer
No published study measures what share of callers to a small business leave a voicemail. Not a survey, not a carrier dataset, not an academic paper. The figures in circulation are assertions that acquired authority by being repeated, and the repetition is now dense enough that they appear in patent filings and press releases as though they were established.
The strongest evidence that nobody knows is the spread itself. If five vendors had each measured the same real quantity, their answers would cluster. Instead they range across thirty-two points, and every one of them is presented without an error bar. A number that varies that much between sources is not a measurement. It is a rhetorical device.
Tracing the 80 percent
The most widely repeated version is that 80 percent of callers sent to a recorded greeting hang up without speaking. It is quoted on vendor pages, in social posts, and — we were surprised to find — inside granted US patent applications describing call-handling systems.
The oldest reachable appearance we could find is a Forbes contributor column from 31 July 2014 about the call-centre industry, which states in passing that eighty percent of callers sent to voicemail do not leave messages, because they do not believe anyone will hear them (Forbes, 2014). Two things about that sentence matter. It is a contributor byline, not Forbes newsroom reporting. And the only thing it points to is a 2009 New York Times piece about consumers drifting away from voicemail as a habit — a culture story about people ignoring their inboxes, not a measurement of what callers do when a plumber does not pick up.
So the chain is: a 2026 vendor page, citing a 2014 opinion column, citing a 2009 newspaper feature, about a different question. There is no sample size in it anywhere. There is no dataset. There is nothing to check.
The dollar figure that changes when you click through
The same pattern runs through the money claims. The best-known is that missed calls cost the average small business around $126K a year. We picked one prominent page carrying it and followed its citation: it credits Ambs Call Center, an answering service, and links there.
Ambs's own page does not contain that figure. What it publishes is a per-call estimate of $12.15 and a worked example putting a business that misses roughly six calls a day at more than $26,000 a year (Ambs Call Center, checked 18 August 2026). Ambs describes the basis as an industry-backed cost-per-call model and conservative revenue estimates, and publishes no sample, no methodology and no underlying dataset on that page.
Set aside whether $12.15 is right. The point is narrower and easier to verify: a widely quoted six-figure number is being attributed to a page that states a figure roughly five times smaller. Nobody in the chain clicked through.
The claim whose study is no longer there
Then there is the assertion that 62 percent of calls to businesses ring out — the one that gives the whole category its urgency. It traces to a blog post by 411 Locals, a local-marketing firm, reporting a small audit in which 37.8 percent of calls were picked up by a live person, the rest going to voicemail or nothing at all. Two figures, one subtraction, one headline.
The vendor pages that cite it still link to that post. As of 18 August 2026 the link no longer resolves to a study; it lands on the company's marketing homepage, where there is no research section, no methodology and no report to download. Whatever was measured, and however few businesses were in it, you can no longer read it. The same is true of a PATLive blog post cited across the niche for the claim that most callers never ring back: that URL now returns a 404.
Link rot is ordinary on the web. What is not ordinary is an entire commercial category resting its pitch on two pages that have quietly stopped existing, and nobody noticing because nobody was checking.
Even the corrections have a sourcing problem
We are not the first to look at this. OnCrew, another vendor in the space, published a careful audit of these same numbers, reached broadly the conclusions above, and deserves the credit for getting there first. It is a genuinely good piece of work and better than most of what the category publishes.
But its replacement figure runs into the same wall. In place of the folklore, it offers a 2025 CallRail consumer survey reporting that 42% of consumers leave a voicemail, that 78% have abandoned a business over an unanswered call, and that 82 percent will ring a rival instead. We went looking for that survey on 18 August 2026 and could not find it published on CallRail's own site. It is not on the company's home services statistics roundup, which is where CallRail collects its first-party findings and is scrupulous about labelling which numbers are its own.
What CallRail does publish, on its own missed-calls page, is that 28% of calls to businesses go unanswered, drawn from customer data in its beta programme. That is a real, self-described, first-party measurement — and it is less than half the figure the category attributes to the same company. If you are going to cite CallRail, cite the number CallRail actually stands behind.
What you can cite
The category's real problem is not that there is no evidence. It is that the honest evidence is duller than the folklore, so nobody uses it. Here is what survived our checking.
| Finding | Source and basis | Year |
|---|---|---|
| 27% of calls to home services businesses go unanswered | Invoca — the vendor's own platform data, disclosed as such | 2024 |
| 28% of calls to businesses go unanswered | CallRail — first-party customer data from its beta programme | 2026 |
| 86% of calls from unknown numbers go unanswered | Hiya, State of the Call 2026 — survey of 12,000+ consumers across six countries | 2026 |
| Contacting a lead within an hour makes qualifying it about 7x likelier than waiting one more hour | Harvard Business Review — audit of 2,241 US companies | 2011 |
| Average first response to a web lead was 42 hours; 23% of firms never responded | Same HBR study | 2011 |
Two caveats you should apply yourself. The Invoca and CallRail numbers come from companies selling call tracking, measured across businesses that had already decided they had a call problem — that is a self-selected sample, and both firms say plainly that it is their own platform data rather than dressing it up as research. The HBR audit is fifteen years old and about web form leads, not phone calls. Cited with those limits attached, all of them are still worth more than a confident number with nothing behind it.
Work out your own number instead
The useful move is to stop shopping for a statistic and spend twenty minutes on your own call log, which is the only dataset that describes your business.
- Pull one month of calls from your phone bill, mobile log or call-tracking tool, and count the ones nobody answered.
- Strike out spam, suppliers, and anyone who rang back and got through. What remains are the enquiries you genuinely lost.
- Apply your own conversion rate — of the enquiries you do answer, what share become paying work? Most owner-operators know this within ten points.
- Multiply by your average job value.
The result will almost certainly be smaller than the figures on vendor pages, and it will be defensible, which the vendor figures are not. It also tells you something they cannot: whether an answering solution is worth buying for you. If the honest arithmetic comes out at two lost enquiries and a few hundred dollars a month, the correct decision might be to change nothing.
What this does to the pitch, including ours
Strip out the unsourced numbers and the argument for having your phone answered gets narrower, slower and more honest. It stops being "you are haemorrhaging six figures" and becomes something closer to: some share of your unanswered calls are real enquiries, response speed measurably affects whether leads convert, and answering machines are a slower response than answering.
That is a weaker sales pitch. It is also the one we can defend, which is why our cost comparison of AI receptionists, answering services and call centres is built entirely on published price cards and skips the missed-call statistics that would have made it more dramatic. Same with our rundown of AI receptionist apps: prices from vendors' own pages, features we could verify, and no invented urgency.
Where this argument is weakest
Fairness cuts both ways, so here is the case against what you have just read.
Absence of a study is not proof the folklore is wrong. It is entirely plausible that most people really do hang up on voicemail — the behaviour matches how most of us treat our own phones, and the 2009 reporting the Forbes column pointed at was describing a real cultural shift. Our claim is narrow: no one has measured it for business lines, so no one should be quoting a decimal point.
The vendors repeating these numbers are also mostly not lying. They are doing what everyone does with a statistic that has been in circulation for a decade — assuming that something quoted this often must have been checked by somebody. We assumed the same thing until we started clicking.
And our replacements are imperfect. Two of the five are vendor platform data with obvious selection effects. One is from 2011 and about a different channel. A category with better evidence would not need this article.
The fix is not a better statistic. It is a lower tolerance for uncited ones — starting with the ones that would have helped us. If you want to see what our AI receptionist actually does with the calls you miss, the Help Center walks through setup, and the call forwarding guide covers wiring it to your existing number without losing your voicemail.
Frequently asked questions
What percentage of callers leave a voicemail?
There is no published study that measures this for small business phone lines. The figures that circulate — that 58, 67, 80, 85 or 90 out of every hundred callers skip the message — all trace back to vendor blogs and a 2014 magazine column rather than to a survey or a dataset. The honest answer is that nobody has measured it, and any vendor quoting you a single confident number cannot show you where it came from.
Where did the 80 percent voicemail statistic come from?
The oldest reachable appearance is a July 2014 Forbes contributor column about call centres, which states the figure in passing and attributes it to a 2009 New York Times piece about consumers drifting away from voicemail generally. There is no sample size, no survey and no business-call dataset anywhere in that chain. Everything published since has been quoting the column, or quoting someone who quoted the column.
Is there any reliable data on unanswered business calls?
Yes, but less of it than the marketing suggests, and it measures answer rates rather than voicemail behaviour. Invoca reported in May 2024 that 27% of calls to home services businesses go unanswered, based on its own platform data. CallRail reports 28% across its customer base. Hiya's State of the Call 2026, a survey of more than 12,000 consumers in six countries, found 86% of calls from unknown numbers go unanswered.
How should I estimate what missed calls cost my own business?
Use your own numbers instead of an industry average. Take the missed calls on your phone bill or call log for one month, estimate what share were genuine enquiries rather than spam or suppliers, multiply by the rate at which enquiries normally become paying jobs for you, and multiply that by your average job value. The result is specific to your business, and unlike any published figure you can actually defend it.
Why does NoMissedCall not quote these statistics?
Because we could not source them, and we sell an AI receptionist, which means the big scary numbers would work in our favour. Publishing a figure we cannot trace to a study would make every other number on this site less trustworthy. We would rather show you a smaller set of citable facts and let you do the arithmetic on your own call log.
Try it on your own phone
No Missed Call is an AI receptionist that answers your business line 24/7, books appointments straight into your calendar, and sends you a transcript and summary of every call. Setup takes minutes from your phone, and there is a 7-day free trial.
Download on the App Store · Get it on Google Play · Read the Help Center
