What a missed call actually costs
Every article about this quotes the same unsourced statistic. You do not need it. Four numbers you already own will get you closer, and pulling them takes about twenty minutes.
Search for what a missed call costs a small business and you will find the same handful of numbers repeated across a few hundred agency blogs. Most callers will not leave a voicemail. Most of those never call back. Pick a percentage, multiply by something, publish.
The numbers might even be roughly right. The problem is that you cannot tell, and neither can the person who published them. A percentage drawn from a national survey of an industry adjacent to yours tells you nothing about whether your phone cost you money last month.
You do not need the statistic. Four numbers you already own will get you far closer than any survey, and pulling them takes about twenty minutes.
The four numbers
- How many calls come in
- How many go unanswered
- How many of those never come back on their own
- What a won customer is worth
Multiply them and you have your number. Everything below is about where each one actually lives.
Where to find each one
Calls in, and calls missed
If the business runs on a mobile phone, your carrier account has a call detail record. Sign in to the account portal rather than the app — the app usually shows a summary while the portal will let you pull the full log for a billing period, often as a spreadsheet. You want a month of it. Incoming calls with a duration of zero are your missed calls.
If you already have a business phone system, the admin console has this built in and usually exports to CSV. Look for a report named something like call log, call history, or call detail. Most of these systems can also show you missed calls by hour and by day of week, which is worth a look on its own — for a lot of businesses the losses cluster into two or three predictable windows rather than spreading evenly.
A month is enough to work with. A quarter is better, because one unusual week will not distort it.
How many never come back
This is the one that takes real work, and it is also the one that makes your number yours instead of somebody's survey.
Take the unanswered numbers from your log. For each one, check whether that number appears anywhere later — a return call, a text, a booked job, an invoice. Some will have called back within the hour. Some will turn out to be existing customers who reached you another way. What is left, the numbers that rang once and never appeared again, is the population you are actually trying to measure.
On a month of moderate volume this is an hour of tedious work. It is worth doing once, honestly, because the resulting fraction is the single number the published statistics are least likely to get right for you.
What a customer is worth
Two pieces. First, average job value: total revenue for a period divided by the number of jobs in it. Not your best job, not the one you quote in conversation — the average, including the small annoying ones.
Second, your close rate: of the enquiries you did answer, what fraction became paid work. If you have never tracked this, count a month by hand. Most owners guess high, and the gap between the guess and the count is usually the most interesting thing the exercise produces.
Working it
Suppose a month of your log looks like this. These are placeholders — the point is the shape of the arithmetic, not the values.
| Input | Example | Where it came from |
|---|---|---|
| Incoming calls, one month | 118 | Carrier call detail record |
| Unanswered | 24 | Zero-duration incoming |
| Of those, never seen again | 9 | Cross-checked against records |
| Close rate on answered enquiries | 35% | Counted by hand |
| Average job value | $2,400 | Revenue ÷ jobs |
| Monthly exposure | $7,560 | 9 × 0.35 × $2,400 |
| Annualised | $90,720 | × 12 |
Two things about that figure. It is larger than most owners expect, and it is built entirely from records you already have. Nobody had to tell you what percentage of callers leave voicemails.
What the number is not
It is an upper bound on an opportunity, not a promise, and treating it as recoverable revenue will make you buy the wrong things.
- Some of those callers were never going to be customers. Price shoppers, wrong numbers, vendors, someone calling every contractor on the first page of results.
- Answering a call does not convert it. The close rate in the arithmetic is your real close rate, not a hundred percent, and that is already the conservative choice.
- Capacity is real. If you are already turning work away, a call you catch is a call you may not be able to serve — in which case the fix is scheduling, not answering.
A number built this way is still worth having. It tells you the size of the category, which tells you how much attention the category deserves. That is the decision you are actually making.
What to do about it, cheapest first
Work down this list and stop when the cost of the next step exceeds what the arithmetic says the problem is worth. Most businesses can do the first two this week without hiring anybody.
- Rewrite the voicemail greeting. Free, ten minutes, and almost universally neglected. Most greetings state the business name and nothing else, which gives the caller no reason to wait. Give a specific callback window and ask for the one piece of information that lets you come back with something useful — the address, or what broke. A caller who has been told when to expect a call is measurably more patient than one who has been told nothing.
- Turn on a missed-call auto-text. Many business phone systems already include this and it is not switched on. Check what you are paying for before you buy anything new. The message should do one job: tell them a human saw the call and when they will hear back.
- Get the business off a personal cell. A personal number cannot be routed, cannot be answered by someone else when you are under a house, cannot be transferred, and does not belong to the company. A business line is usually the cost of a couple of coffees a month and it unlocks everything above and below it on this list.
- Put something on the phone that always answers. An answering service or an AI receptionist that captures the job and texts you immediately. This is the first step on the list with real monthly cost and real setup, so it is the first one that needs the arithmetic to justify it. Quotes vary enormously — get three, and make sure you are comparing what happens to the message after it is taken, not just the price per minute.
The order matters more than the list. Buying step four while step one is still broken is how businesses end up paying a monthly fee to route calls into the same hole.