LTV:CAC and CAC payback calculator

LTV:CAC and CAC payback get reported together and read as one number, which is a mistake: they answer different questions. The ratio asks whether a customer is worth having. The payback period asks how long you finance that customer before you find out. A business can pass the first test and be strangled by the second, which is why the tool above prints both.

Two formulas, two questions

Four inputs, and arithmetic short enough to check by hand.

LTV = (ARPA × gross margin) ÷ monthly churn

CAC payback = CAC ÷ (ARPA × gross margin)

On the inputs as loaded — $420 of ARPA at 78% margin, 2.8% monthly churn and $3,600 of CAC — each account contributes $328 of gross profit a month. Divide that by churn and lifetime value is $11.7K, giving a ratio of 3.3x. Divide CAC by the same monthly gross profit and payback is 11 mo.

Notice the shared term. Both sit on monthly gross profit, so anything moving margin or ARPA moves both. Churn appears in only one and CAC in the other, which is why they can disagree: the ratio is a verdict on the whole relationship, the payback a statement about your bank balance in the meantime.

Why gross margin belongs in LTV

A great many published LTV figures are revenue multiplied by expected lifetime, with no margin term at all. Revenue handed straight to your hosting provider, payment processor and support team was never yours. Counting it as lifetime value compares a gross number against a net cost.

Leaving margin out is arithmetically identical to asserting a margin of 100%. On the default inputs that turns $11.7K of lifetime value into $15K and the ratio from 3.3x to 4.2x — an overstatement of 28%, for free, with no change to the business. At 60% margin the same omission takes the ratio from 2.5x to 4.2x.

The same term belongs in payback: recovering $3,600 takes 11 mo at $328 of monthly gross profit, and considerably longer if you keep half of each dollar. What counts as cost of revenue therefore has consequences well beyond the margin line — SaaS gross margin explained works it through.

Churn is the whole denominator

Churn sits underneath LTV on its own, which makes lifetime value inversely proportional to it. This is not a sensitivity in the usual sense, where a change in an input produces some smaller change in the output. Halving churn doubles LTV exactly; doubling it halves LTV exactly. Nothing else on the panel behaves like that.

Read the extremes. At the bottom an account lasts a working lifetime; at the top the business replaces its whole customer base every year and the ratio is barely above one. Same product, same price, same acquisition cost. Moving churn from 2.8% to 1.4% takes this company from 3.3x to 6.5x without touching a marketing budget, which is why retention work is usually the cheapest improvement available.

It also explains why the ratio is nearly meaningless for young companies: churn is measured over a few months of a customer base that has not had time to churn, so the denominator is small and badly estimated. An early-stage LTV:CAC of 8x usually means the cohort is young, not that the economics are exceptional. See SaaS churn, compounded.

The 3:1 bar, and its ceiling

The conventional benchmark is 3:1, and the logic is rough but sound: you need enough margin above acquisition cost to fund the overhead around the customer and the ones who never work out. Below 1:1 every sale destroys value; between 1:1 and 3:1 the model survives only if overheads are unusually thin.

The part that gets ignored is the top of the band. A ratio of 6.5x is not twice as good as 3.3x; it usually signals that you are not buying enough customers. It comes from the same $11.7K of lifetime value against a CAC of $1,800 rather than $3,600, payback falling to 5 mo. Channels get more expensive as you scale them, so a very high ratio means you are sitting on the cheapest slice of demand.

The right response to a ratio well above 5:1 is rarely celebration. It is to spend more per customer, accept a lower ratio and grow faster.

Good ratio, ruinous payback

The combination that does real damage is a ratio clearing the benchmark comfortably while payback runs for years. The two companies below have an identical LTV:CAC.

The enterprise business retains far better — 9.2% of annual churn against 28.9% — and holds more than twenty times the lifetime value per account. It is also the one that runs out of money. At 3y 2mo of payback, every ten accounts it signs commit $750K that will not return in full until past the three-year mark, and growing faster deepens the hole.

The self-serve business recovers its acquisition cost in 11 mo and funds the next cohort out of the last one. That is the difference between growth you can self-finance and growth that needs a balance sheet, and the ratio cannot see it. Below twelve months of payback is best-in-class for business software; twelve to eighteen ties up cash; beyond eighteen, growth is debt in everything but name. CAC payback vs. LTV:CAC goes further.

None of this makes long payback unacceptable. It makes it a financing decision rather than a marketing one: know it before you hire the sales team, and hold the cash to cover the trough it creates. That is what the cash runway calculator is for.

Blended CAC and the missing payroll

The largest source of variation in reported CAC is the choice of definition, and almost every available choice flatters the number. Take one company acquiring 100 accounts a month. Roughly 40 arrive through word of mouth, organic search and existing customers; the other 60 come from paid channels. Media spend is $216,000 a month, and the sales and marketing salaries behind it cost another $120,000. Four defensible numbers come out, and four different verdicts.

The range is 5.4x down to 2.1x for one business in one month, and payback moves from 7 mo to 1y 5mo. The top row is the one that appears in decks: it spreads paid acquisition cost over customers who cost nothing, and excludes the salaries of the people running acquisition.

• Use paid CAC to decide whether to spend more. Marginal decisions need marginal costs. Organic accounts arrive regardless; including them tells you nothing about the next dollar of budget.
• Use blended CAC to sanity-check the whole engine, and only if it includes fully-loaded sales and marketing payroll, tooling and agency fees. A blended figure that omits payroll is a different metric wearing the same name.
• Say which one you are using, every time. Most disagreements about whether the economics work are two people using different denominators.

The CAC input above is labelled blended and expects sales payroll in it. If yours excludes payroll, both outputs flatter you by roughly the share payroll takes of acquisition cost.

What this does not model

The arithmetic is exact. The assumptions underneath are simplifications, and three matter enough to name.

• CAC is constant. The next customer costs what the last one did. Real channels saturate: the cheapest keywords, warmest referral lists and most obvious segments run out, marginal CAC climbs, and the average lags behind it. A business bidding against its own ceiling can hold a respectable average for a year while every incremental customer is unprofitable.
• There is no expansion revenue. ARPA is fixed for the life of the account, so one that doubles its seats over two years is modelled as flat. Where net revenue retention runs above 100% this understates lifetime value substantially, which is the best argument for treating the output as a floor rather than an estimate. See NRR vs. GRR.
• There is no segment mix. One ARPA, one churn rate and one CAC describe an average customer who may not exist. If a fifth of your revenue is enterprise accounts churning at 0.5% and the rest self-serve churning at 6.0%, the blended figure describes neither. Run this once per segment; the answers are often far enough apart to change where the sales team is pointed.

Discounting is absent too: a dollar of gross profit in month forty counts the same as one today. That is conventional and generous, particularly at long payback periods. For a stricter reading, take both numbers and give the payback the deciding vote.

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.