Most calculators in this category are built to produce a large number. They ask for your missed calls and your average sale, fill the rest in for you, and multiply. The result is arithmetic performed on somebody else’s guesses about your business, and it is usually wrong in the same four ways at once.
This one is built to produce a number you can defend in front of whoever controls the money. That means it is slower to fill in, it will tell you “unknown” where you have no answer, and the figure it gives you will be smaller.
Work out your own
Pick one and keep every figure on it. A missed-call count from one busy week beside a monthly cost is the most common way one of these estimates goes wrong.
The count off your phone system, as it comes: attempts, not people. The next box turns it into people.
One person ringing three times in an afternoon is one opportunity and three missed calls. At least 1. If you have never measured it, leave it blank rather than guessing: the answer will say unknown, which is true.
Wrong numbers, robocalls, suppliers and existing customers ringing about a finished job are all in that count and none of them were a sale.
These are the numbers nobody can look up, so give each one a range you are willing to defend. The cautious end of all three at once is the low estimate; the optimistic end of all three is the high one. Leave any of them blank and everything downstream of it reads as unknown.
Use whatever currency you bill in. Nothing here converts anything.
What is left after the cost of delivering the work. Revenue is not profit, and a calculator that stops at revenue is answering an easier question than the one you asked.
Software, an answering service, the extra hours, whatever it is. Zero is a valid answer if the change costs you nothing.
An invented example, so you can watch the chain move. Not a benchmark and not taken from anybody's research. Replace every number with your own.
Your numbers stay in your browser. Nothing is uploaded, and nothing is stored.
What the numbers say
| Cautious | Likely | Optimistic | |
|---|---|---|---|
| Already come back on their own | unknown | unknown | unknown |
| Of the rest, reached by the change | unknown | unknown | unknown |
| Of those, become a sale | unknown | unknown | unknown |
| Unique qualified opportunities | unknown | unknown | unknown |
| Still unrecovered today | unknown | unknown | unknown |
| Recovered by the change | unknown | unknown | unknown |
| Extra sales | unknown | unknown | unknown |
| Extra revenue | unknown | unknown | unknown |
| Contribution before the cost | unknown | unknown | unknown |
| Contribution after the cost | unknown | unknown | unknown |
| The same, over a year of these periods | unknown | unknown | unknown |
What it has to win to pay for itself
- Extra sales needed per period
- unknown
- Recovered conversations needed per period
- unknown
Measured in contribution, not revenue: the cost comes out of what is left after delivering the work.
Left blank, so everything downstream of it reads as unknown:
- Missed calls in that period
- Times the same person rings before giving up
- Share of those callers who were a real opportunity
- Share who already come back on their own
- Share of the rest a change in call handling would reach
- Share of recovered conversations that become a sale
- Average value of one sale
- Contribution margin on that sale
- Extra cost of the change, over the same period
Worth saying out loud
- A revenue figure needs a conversion rate and an average sale. Without both it stays unknown.
- A contribution figure needs a margin and the cost of the change as well.
One file: the three scenarios, every input behind them, and what the model does not know.
per month (12 in a year)
The four ways this goes wrong
Rings counted as people. A caller who tries you three times in an afternoon and gives up is one lost job and three missed calls. Your phone system counts the second thing. Feeding that straight into an opportunity count multiplies the answer by your repeat rate, and nothing downstream corrects it.
Everyone counted as a prospect. Wrong numbers, robocalls, suppliers, couriers and existing customers ringing about a job already paid for are all in that count, and none of them were ever going to buy anything.
Credit taken for the people who come back anyway. Some share of missed callers ring again the next morning, or you already call them back off your voicemail. That revenue is in your accounts today. A model that does not subtract it first is claiming to recover money you already have, and this is the single largest source of inflation in the category.
Revenue presented as profit. A recovered job brings in the sale value and then costs you materials, labour and time. The cost of whatever produced the recovery has to come out of what is left, not out of the sale value. On the worked example below, the cautious case has 330.75 of extra revenue and still loses money.
A worked example, invented on purpose
Every number in the table below is made up. It is not a benchmark, it is not a survey result, and it is not taken from any customer of ours. It exists so you can watch the chain shrink, and so the page, the calculator and npm run check:missedcall all print the same worked example and cannot drift apart. Replace all of it with your own figures.
| Cautious | Likely | Optimistic | |
|---|---|---|---|
| Already come back on their own | 50% | 35% | 25% |
| Of the rest, reached by the change | 20% | 35% | 50% |
| Of those, become a sale | 15% | 25% | 35% |
| Unique qualified opportunities | 52.5 | 52.5 | 52.5 |
| Still unrecovered today | 26.3 | 34.1 | 39.4 |
| Recovered by the change | 5.3 | 11.9 | 19.7 |
| Extra sales | 0.8 | 3 | 6.9 |
| Extra revenue | 330.75 | 1,254.09 | 2,894.06 |
| Contribution before the cost | 148.84 | 564.34 | 1,302.33 |
| Contribution after the cost | -31.16 | 384.34 | 1,122.33 |
Read the first and last rows together. 120 missed calls became 52.5 real opportunities, then 11.9 recovered conversations, then 3 extra sales in the likely case. The same inputs at the cautious end of all three ranges lose 31.16 over the month. That spread is the honest output, and the middle column on its own is not.
How it works underneath
The whole model is one chain of multiplications, and printing it is most of the point. Missed calls divided by the number of times the same person rings gives unique callers. Times the qualified share gives unique qualified opportunities. Times one minus the natural recovery share gives the ones still unrecovered today. Times the additional recovery share gives the ones a change would newly reach. Times conversion gives extra sales, times the average sale gives extra revenue, times the margin gives contribution, and the cost of the change comes off the end.
Two of those steps are the guards. Dividing by the repeat rate is what stops the same prospect being counted three times, and applying the additional recovery only to what is left after natural recovery is what stops the model taking credit twice for the same caller. Both are missing from a calculator that asks for four numbers.
Three of the inputs are forecasts rather than facts: how many callers already come back, how many of the rest a change reaches, and how many of those buy. Each is given as a range, and the cautious column takes the least flattering end of all three at once. Note that the cautious end of natural recovery is its high end, because more people coming back on their own leaves less to recover. Average sale, margin and cost are not given ranges, because those are read off your books or quoted by a supplier.
Anything you leave blank stays blank all the way down. A missing conversion rate means no revenue figure at all, rather than a revenue figure computed from a number nobody supplied. The tool also names the edges rather than computing through them: a hundred percent recovery rate, a hundred percent conversion, a period with no missed calls and a range entered backwards each produce a sentence instead of a plausible figure.
All of it runs in your browser, in about a dozen multiplications. There is no endpoint behind this page. Your missed-call counts, your close rate, your average sale and your margin are among the most private numbers your business has, and they are none of ours.
What this is not
It is not a measurement. Nothing here was established by an experiment, and the two numbers that move the answer most are forecasts you made about your own business. It is not a forecast either, because a forecast implies somebody is predicting the outcome. It is a worksheet: it makes your assumptions explicit, multiplies them consistently, and shows you which one the whole case rests on.
It is not a promise about any product, ours included. Buying a phone system does not produce the number in the last row. What a phone system can change is where a call goes when nobody picks it up, and by how much is the assumption you typed in yourself.
It is not a comparison of vendors. Other calculators exist in this category, some tied to a product; Nextiva publishes one alongside its voice-agent offer. The difference worth knowing is not whose arithmetic is better, it is which assumptions each one makes on your behalf.
The incremental-recovery chain used here is this site’s own model, written up in what missed calls are actually worth. It is not a measured finding from any vendor.
Questions
How do I work out what a missed call costs my business?
You cannot look it up, and anybody who hands you a figure has guessed at your business. What you can do is build it from numbers you already have: how many calls you actually missed in a period you choose, how many of those were the same person ringing twice, how many were a real opportunity rather than a supplier or a wrong number, how many of those come back on their own anyway, and what one job is worth to you after the cost of doing it. The calculator above multiplies exactly those together and shows the chain, so you can argue with any step of it.
Why does this calculator not fill anything in for me?
Because a pre-filled conversion rate is somebody else's guess about your business, printed in your currency. The moment a calculator ships with 30% in a box, the number it shows a first-time visitor is fiction, and the visitor has no way to know which parts are theirs. Every box here starts empty and every figure that depends on an empty box reads as unknown rather than as a number. If you have never measured your close rate on missed enquiries, unknown is the true answer and it is more useful than a plausible one.
What is the difference between a missed call and a missed opportunity?
A missed call is one unanswered ring. A missed opportunity is one person. Somebody who rings you three times in an afternoon and gives up is a single lost job and three missed calls, and a phone bill counts the second thing. Reading a missed-call count straight into an opportunity count therefore multiplies the answer by whatever your repeat rate is, which for a small service business is commonly somewhere between one and a half and two. The second box on this page is the only correction for it, and if you have never measured it, leaving it blank is honest.
Why does the tool subtract the people who call back anyway?
Because they were never lost, so no change in your call handling can recover them a second time. A share of missed callers ring again the next morning, or you already ring them back off your voicemail. That share is already in your revenue. Counting it as recovery is the single largest reason these calculators produce numbers that sound enormous and never show up in the accounts. The additional recovery figure on this page applies only to what is left after the ones who come back on their own are taken out.
Why does it ask for a margin instead of just showing revenue?
Because revenue is not profit, and a business case decided on revenue is decided on the wrong number. A recovered job brings in the sale value and then costs you materials, labour and whatever else it takes to deliver. What is left is contribution, and that is what has to cover the cost of whatever produced the recovery. The tool prints both, in that order, and then subtracts the cost. A large revenue figure sitting above a negative contribution figure is a real and common result.
Does using Ringfully produce the number this calculator shows?
No. Nothing on this page is a forecast of what any product will do for you, ours included. The output is your own assumptions multiplied together, and the two that matter most, how many missed callers a change would reach and how many of those would buy, are forecasts you supplied about your own operation. Treat the result as a way to test whether a change is worth investigating, and to see which assumption the whole case rests on. The useful next step is usually to look at where calls are actually going now, not to buy anything.
A next step that costs nothing
Before deciding anything, map what happens to a call today when nobody picks up: how long it rings, where it goes after that, whether the caller is told anything, and whether a message reaches a person the same day. Most of the recovery a small business finds is in that map rather than in any purchase, and you cannot forecast a change you have not described.