Churn calculator: annual retention and your revenue ceiling

Monthly churn is the one input in a subscription model that looks negligible on a monthly report and decides the maximum size of the company on a five-year view. At the 2.8% the calculator loads with, you keep 71.1% of a cohort’s revenue through a year and 18.2% of it through five — and you cannot grow past $535.7K of MRR however hard the sales team works.

What the four tiles mean

The calculator takes four inputs and returns four numbers that answer four different questions. Annual gross retention answers “how much of what I have will still be here in a year”. Annual net retention answers “does the base grow on its own”. Average account life answers “how long do I get to bill this customer”. The revenue ceiling answers the one nobody asks: “how big can this company get if nothing about retention or acquisition changes”.

Only the last of those is a hard constraint. The first three describe the shape of the business; the ceiling tells you where it stops. On the loaded inputs — 2.8% monthly churn, 1.5% monthly expansion and $15K of new MRR a month — the model reports gross retention of 71.1%, net retention of 85.5%, an average account life of 3y and a ceiling of $535.7K a month. Three of those are diagnostics. The fourth is a wall.

Annualising churn properly

Churn compounds against a shrinking base, so the annual figure is 1 − (1 − monthly)¹², not monthly × 12. Multiplying by twelve charges every month’s churn against the original base, which double-counts revenue that already left.

At 1% a month the shortcut is off by 0.6 points and nobody would notice. At 8% it understates what you keep by 32.8 points of your revenue base — the shortcut says 4.0% survives the year when 36.8% actually does. Past 8.33% a month the shortcut returns a negative retention rate, which is the clearest possible signal that it is not a formula.

The error runs in the direction that makes people give up too early. A team told it is losing 96% of its revenue a year concludes the product is unsalvageable; a team told it is losing 63.2% concludes it has a retention problem worth fixing. The second team is right.

Gross, net and the hollow base

Gross retention only ever falls. It counts cancellations and downgrades and stops there, so it is capped at 100% by construction. Net retention adds expansion — extra seats, higher usage, tier upgrades — back in, so it can exceed 100% and often does. The calculator reports both because they are different claims about the business.

On the loaded inputs the two diverge by 14.4 points over a year: 71.1% gross against 85.5% net. That gap is the whole of the expansion motion, and at 1.5% a month it is not enough to hold the base flat — expansion would have to match churn exactly, at 2.8% a month, to keep net retention at 100%.

The rule is simple: never report net retention without gross retention beside it. Net retention above 100% is only good news when gross retention is also high. Read alongside each other they say “we keep customers and they grow”. Read alone, net retention can mean “we are losing customers and squeezing the survivors harder”.

The revenue ceiling

This is the most useful output the calculator produces and the one almost no churn tool surfaces. If you add a roughly constant amount of new recurring revenue each month, MRR does not grow forever. It converges on:

steady-state MRR = new MRR per month ÷ monthly churn rate

At that level, the dollars churning out each month exactly equal the dollars sales brings in, and growth stops. Not slows — stops. On the loaded inputs, $15K of new MRR a month against 2.8% churn caps the business at $535.7K of MRR, or $6.4M of ARR. Every plan that projects past that number is projecting a retention improvement, whether or not it says so.

Read the last column with the second. High churn does not only lower the ceiling; it drags you into it faster. At 1% churn the ceiling is $1.5M and you are still climbing towards it nineteen years later. At 8% you hit 90% of a $187.5K ceiling inside 28 mo and then flatten, which is exactly what a stalled SaaS business feels like from the inside.

Now the part that decides where you spend. Doubling new MRR to $30K a month doubles the ceiling to $1.1M. Halving churn to 1.4% also doubles it, to $1.1M. The arithmetic is symmetric, but the two levers are not: acquisition spend has to be paid for again every month to hold the higher ceiling, while a retention improvement holds the new ceiling for free and raises average account life to 5y 11mo along the way. Raising acquisition raises the ceiling. Only retention removes it.

Cohort survival and half-life

Average account life is 1 ÷ monthly churn, and it is a mean, which means it is pulled upward by a thin tail of accounts that stay for years. The number that matches what a cohort actually feels like is the half-life: the month by which half of that cohort’s revenue is gone.

The half-life is consistently a little over two-thirds of the average life, so any planning done on the average is planning against a date by which most of the cohort has already left. On the loaded inputs the average account life is 3y, but half the revenue from any given month’s cohort is gone by month 24.

This matters for payback. If CAC payback sits near the half-life, half your customers never repay their acquisition cost — the average customer does, and the median one does not. Run the same churn rate through the LTV, CAC and payback calculator and compare the payback period against the half-life column above, not against the average-life column. It is the stricter and more honest test.

Benchmarks by segment

Acceptable churn depends almost entirely on who you sell to, and comparing across segments tells you nothing. These are the ranges the industry works to:

Two cautions. First, a self-serve business at 5% churn is normal and a mid-market business at 5% is in trouble, so the same input means opposite things. Second, blended churn across segments is close to meaningless if the mix is uneven — a company with an enterprise base and a self-serve long tail will report a blended rate that describes neither, and the calculator will faithfully compute a ceiling that no part of the business is actually approaching.

What the model assumes

The calculator is deliberately small. It is worth knowing exactly where it stops:

• A single blended churn rate. There is no cohort mix and no segmentation. If your first-month churn is far higher than your steady-state churn — which is normal — a single blended figure will overstate long-run losses and understate early ones.
• Expansion as a flat monthly percentage. Real expansion is lumpy and arrives at renewal, not evenly across twelve months. The annual net retention figure is right; the monthly path to it is smoother than yours will be.
• No seasonality and no ageing. Churn does not vary by month and does not fall as a cohort matures, though in most businesses it does both.
• Constant new MRR for the ceiling. The ceiling assumes acquisition holds flat. If new MRR grows, so does the ceiling — but it is still a ceiling, just a moving one, and it still binds.
• Revenue churn, not logo churn. Feed it account counts and the ceiling becomes an account ceiling, which is a different and less useful number.

The one number to take away

If you record nothing else from this page, record the ceiling. Compare it to your board plan. If the plan projects revenue above the ceiling your current churn rate implies, the plan contains an unstated retention assumption, and it is worth making that assumption explicit before the year it has to come true. The mechanics behind it are worked through in the churn guide.

Educational content, not financial advice. This tool models the assumptions you enter and excludes taxes, financing and one-off items. Check any figure that matters with a qualified accountant before you act on it.