Search for missed call statistics and you will meet the same handful of numbers on every page. Sixty-two percent of calls to small businesses go unanswered. Eighty-six percent of callers who reach voicemail hang up. A six-figure sum walks out the door every year. The pages quoting them rarely name a study, and when they do, the study is usually a link to another page quoting the same number.
We went looking for the originals. This page records what we found: which figures trace to a real piece of research, how old that research actually is, and which ones we could not source at all. Where we could not verify something, it says so, because a statistics page that hides its own failures is the problem it claims to solve.
Last checked on the date at the top of this page.
The sixty-two percent figure: real, and older than you think
The number. Sixty-two percent of calls to small businesses go unanswered.
Where it comes from. A study by 411 Locals, a local search marketing firm, covering 30 days of calls to 85 businesses across 58 industries. The breakdown is more useful than the headline: 37.8 percent of calls were answered by a person, 37.8 percent went to voicemail, and 24.3 percent got no response at all. The sixty-two is the second and third numbers added together.
The catch. It was published in January 2016. A large number of pages citing it carry a 2023 or 2024 date, which appears to come from the date those pages were written rather than the date of the research. Phone behavior has not stood still in a decade, and a sample of 85 businesses was never meant to carry the weight now placed on it.
Verdict. A real study with a real method, usable if you cite the year. Not usable as a current figure.
The eighty-six percent figure: we could not find a source
The number. Eighty-six percent of callers who reach voicemail at a service business hang up without leaving a message.
Where it is said to come from. Most pages attribute it to Invoca, a call analytics company. Some attribute a similar figure in the low eighties to Hiya's State of the Call report.
What we found. Searching for the original, we could not locate an Invoca publication containing that number. The closest verifiable Invoca figures point in a different direction: a 2021 aggregate reporting that only around two percent of callers leave a voicemail at all, and a more recent benchmark, quoted second-hand, that forty-four percent of callers never reach a person on an inbound call. Those measure different things. The Hiya figures we could find were also second-hand, described a 2026 finding that forty-one percent of people let an unknown number go to voicemail and fourteen percent answer immediately, and were about unknown callers rather than voicemail abandonment.
Verdict. Unverified. It may trace to a report we did not reach, or it may be a misquote that acquired a source along the way. We would not use it, and we would be cautious about any page that does without naming a document.
The six-figure annual loss: arithmetic, not measurement
The number. A specific six-figure sum lost per year to missed calls, quoted to the nearest thousand.
Where it comes from. Vendor calculation. The method, where it is disclosed, multiplies monthly missed calls by an average customer value, by twelve, and then by an assumed conversion rate. Every term after the first is an assumption, and the assumed conversion rate is doing most of the work.
Verdict. Illustrative, not evidence. The precision is the giveaway: a figure derived from three estimates cannot be accurate to the nearest thousand. It is fine as a worked example on your own numbers and misleading as a statistic about an industry.
The home services range: no named study
The number. Seventy-five to eighty-five percent of callers who reach voicemail at a home services business hang up.
What we found. The sources describing this range attribute it to industry data from call software providers, without naming one. We could not find the underlying data.
Verdict. A vendor estimate presented as a finding. Treat it as a guess, and a self-interested one.
The one that holds up better than most
The number. Forty-four percent of calls to small businesses are answered on the first or second ring.
Where it comes from. An analysis attributed to CallRail covering around 100,000 small business phone lines. The sample size is in a different league from 85 businesses, and the thing being counted, speed of answer, is something a call platform can measure directly rather than infer.
Verdict. Credible and worth citing with the source named. Note that "answered on the first or second ring" is a narrower claim than "answered", and some pages quietly widen it.
Why this matters more than the numbers do
Every figure above is being used for the same purpose: to persuade a business owner that the phone is leaking money. The owner usually already knows that. What the owner does not know is their own number, and no industry statistic can supply it.
A sample of 85 businesses from 2016 cannot tell a plumbing company in a hail-prone county what happened to its phone during last August's storm week. Only that company's own call log can.
How to get your own number instead
You can do this in a week without buying anything.
- Count the calls that arrive, not the ones you remember. Your phone provider's log has this.
- Count how many were answered by a person, and treat voicemail as unanswered, which is what the 411 Locals study did and what makes its breakdown honest.
- Split it by hour and by day. The average is rarely the problem. A business that answers nine calls in ten on a Tuesday morning and two in ten on a Saturday has a Saturday problem, not an answering problem.
- Count the repeat numbers. A caller who rings twice in ten minutes and never rings again told you something.
- Work out what one booked job is worth to you. Now you can do the arithmetic the vendors do, with your own figures in it, and the answer means something.
If the gap is in the hours when nobody is there, that is a coverage problem and it has several solutions, of which an answering service is one and an AI receptionist is another. If the gap is in the middle of a working Tuesday, the problem is capacity and no software will fix how many hands you have.
What we do with this
MairyAI writes up every call, which means a business running it has the log described above without keeping a tally by hand: how many calls arrived, what time they came, what the caller wanted, which ones ended in a booking, and which callers had rung before. We publish our own prices on the pricing page and we do not publish a loss figure, because we would have to make up two of the three numbers in it.
If you quote anything from this page, quote the source and the year with it. That is the whole point.