Typed as a lifespanfour years, or 2.08 per cent a month
420 × 3 × 4 × 45% = $2,268
Dividing four years into 48 months feels like arithmetic. Losing one forty-eighth every month leaves 77.7 per cent after a year, which is 22.3 per cent annual churn rather than the 25 you started from.
Typed as retention75 per cent a year, converted properly
monthly churn = 1 − 0.751/12 = $1,995
The same fact about the same customers, 13.7 per cent lower. Neither input is wrong; they simply do not convert into each other the way the forms assume.
Seven of the nine free calculators Google ranks already multiply by margin, so the formula argument is over. What still moves the answer is the field you type the retention into, and the two doors below lead to the same customer.
What I checked
I opened the nine free CLV calculators Google returns on page one for two phrasings, on 1 September 2026, and classified each by the fields it asks you to fill, not by the prose around it. There is a lifetime value calculator on this site too, built for clinics, and it already asks for contribution margin.
That distinction matters more than it sounds, and I got it wrong first. WebFX prints "(Average annual revenue from a single customer) X (Number of years) – (Customer acquisition cost)" on its page, so I filed it under revenue. Its calculator asks for Cost per Sale. The formula lives in the FAQ accordion; the tool underneath it works on margin. Selzy reads the same way until you notice the third field is Average margin, with a note telling you that entering one per cent gets you revenue instead of profit.
The convention has won. Churnkey prints it as a fraction on the page, ARPU x Gross Margin / Churn Rate, and Wall Street Prep prints the same thing and then quotes the three to one rule beside it. If you read that a lifetime value should be three times acquisition cost, that is the lifetime value being talked about.
The one holdout is upGrowth, which multiplies annual revenue by years and subtracts acquisition cost. On the business below its answer is 2.09 times the margin answer. Its own worked example calls the result profit, which it is not.
The input is where the money actually moves
Take a business invented for this: an average order of 420 dollars, three orders a year, a 45 per cent gross margin, and 75 per cent of customers still there a year later. One business, one formula family, three doors into it.
- Type a four year lifespan, which is one over 25 per cent annual churn: $2,268.
- Type 2.08 per cent monthly churn, which is four years divided into months: $2,268.
- Type 75 per cent annual retention and let the tool convert it properly: $1,995.
The third is 13.7 per cent lower than the other two, and none of the three inputs is wrong. They are the same fact about the same customers, entered through different fields.
Why the first two agree, and why that is the problem
Here is the part I got wrong while writing this, which is how I found it.
Dividing four years into 48 months and calling it 2.08 per cent monthly churn feels like arithmetic. It is not. Losing one forty-eighth of your customers every month leaves you with 77.7 per cent after twelve, so that input describes 22.3 per cent annual churn, not the 25 per cent you started with. The two agree with each other because they are the same mistake written twice, and they agree loudly enough that I built a check that could not fail and shipped it as evidence.
Converting properly is one line: monthly churn is one minus the twelfth root of your retention rate, which for 75 per cent is 2.37 per cent, not 2.08. The gap between 2.08 and 2.37 does not look like much. It is 273 dollars a customer here, and it grows as retention falls.
So the calculators that ask for churn and the calculators that ask for a lifespan are not interchangeable, even when they belong to the same family and print the same formula. Whichever one you use, the number you get is a statement about the door you came in through.
The acquisition cost subtraction
Two things get confused here, and the smaller one is not a scandal. Subtracting acquisition cost inside lifetime value gives you LTV minus CAC, and dividing that by CAC gives exactly LTV over CAC minus one. Always exactly one less, by algebra. That is a real quantity, return on acquisition spend, and finance people compute it on purpose. It is just not the number the three to one rule is written in, and on our figures the two readings are 7.6 and 6.6, which straddles nothing but is worth not confusing.
The larger point is the one above: at 2.09 times, the revenue basis moves the answer far more than the subtraction does.
Three things to write down
Which door you came in through
Lifespan, monthly churn or annual retention. They are the same fact and they do not give the same answer, so the input belongs beside the output.
Whether your churn conversion compounds
If you divided a lifespan into months, you have understated churn and overstated lifetime value. One minus the twelfth root is the fix, and it takes a cell.
Whether the margin is gross, contribution or something a vendor invented
WebFX defines its margin net of hosting and support. Our clinic tool uses contribution margin. Gross margin is a reporting measure with its own rules. The word alone does not tell you which.
Where this sits in our own tool
KPI Vitals shows a lifetime value on its customers page, and it is in the minority camp: it counts revenue, not margin. Everything above about the 2.09 times gap applies to it, and you would have to apply your own margin before comparing that number with an acquisition cost.
What it does instead of a formula is refuse to project. There is no lifespan field and no churn field, because there is nothing to type: it adds up what each customer has been invoiced and divides. That also means an invoice you have not been paid for is inside the number, and on the cash page at the same time. It divides by customers with two or more orders, on the grounds that a single purchase is not a lifetime, which pushes the figure up against averaging everyone.
Two of those are deliberate. The revenue basis is the one I would change, and saying so here is cheaper than letting you find it.
Related reading
- Patient lifetime value: how to calculate PLV for a clinic uses contribution margin and shows the working
- Accounts receivable aging report: which date your columns count from
- Breakeven ROAS: the one field no calculator asks for is the other half of the acquisition ceiling
- Revenue cycle KPIs: the six numbers that decide if you get paid
Questions people actually ask
The free calculators have largely settled on one: seven of the nine on page one multiply revenue by a gross or profit margin. Churnkey prints it as ARPU times gross margin over churn rate, and Wall Street Prep prints the same and quotes the three to one rule beside it. One of the nine still works on revenue alone.
Margin, whenever the number is going to be compared with acquisition cost, because the comparison asks whether what you get back covers what you spent and revenue is not money you keep. On the example here the revenue basis is 2.09 times the margin one.
No, and this is the error that costs most. Losing one forty-eighth of your customers every month leaves 77.7 per cent after a year, which is 22.3 per cent annual churn, not 25. The conversion is one minus the twelfth root of your retention rate: for 75 per cent that is 2.37 per cent a month, not 2.08.
Usually the input rather than the formula. A lifespan, a monthly churn rate and an annual retention rate are the same fact stated three ways, and they do not convert into each other the way people assume. Here they produce $2,268, $2,268 and $1,995.
It is not wrong, it is a different number. Lifetime value minus acquisition, divided by acquisition, equals lifetime value over acquisition minus exactly one. That is return on acquisition spend, which finance people compute on purpose. It is not the ratio the three to one rule is written in.
Three to one is the figure everyone quotes, and it assumes the margin version. Apply it to a revenue based lifetime value and you will pass a test you have not taken.
WRITTEN BY
Olha · clinic data analyst
I build the reporting our managers open every morning at a multi-branch medical clinic — and package it so other practices don't have to start from scratch.
Published on 1 September 2026. Four limits worth stating outside the body text. The sample is two search phrasings, one market, one day, and it is a sample of free calculators rather than of the literature: no textbook and no professional body is in it. The business is invented, and a real one has a margin that moves. The word margin covers gross, contribution and whatever a vendor decided, and the calculators do not agree on that either. And this piece was rebuilt after a fact-check found three errors of my own, including a build check written so that it could only confirm the answer I had already reached; the corrections are the second and third sources below rather than a quiet edit. I sell a $39 file that shows a lifetime value, which is why the article says which kind it is and which part of it I would change.