What a customer is worth, on stated assumptions
Lifetime value is a model, not a measurement: it stretches today’s revenue per customer across an expected lifetime set by retention. This page declares its model (a steady-state cohort, constant revenue and retention), keeps revenue value and contribution value separate, shows a simple and a discounted version with the timing assumption of each, and tables the sensitivity, because a single LTV figure without its assumptions is a number waiting to mislead a budget.
Disclaimer: This calculator gives estimates for general information only. It is not financial, medical or other professional advice. Lifetime value is a projection under stated assumptions about retention, revenue and margin holding steady; real cohorts behave differently, and acquisition decisions should not rest on this figure alone.
What this calculator does
It builds lifetime value from a period (a year of orders, or a subscription month) and a retention rate, in a steady-state model where revenue per customer, margin and retention are assumed constant. Two versions are always shown. The simple model multiplies contribution per period by the expected lifetime (1 / churn), counting the current period in full. The discounted model uses the standard steady-state formula, contribution x retention / (1 + discount rate – retention), which treats each period’s contribution as arriving at the period’s end, weighted by the chance the customer is still there, and discounts it back. The two differ, honestly and by construction, and the page says why, because teams have torn themselves apart over two LTV figures that turned out to be two models. Revenue LTV (before margin) is shown separately and labelled, and the sensitivity table varies retention and margin together over the discounted model so the figure’s fragility is quantified rather than suspected.
How to use it
- Use contribution for decisions, revenue for context. CAC is spent in real money, so it may only be compared with contribution LTV (after gross margin). Revenue LTV will always flatter the ratio by the inverse of the margin; at a 25% margin it quadruples it. The ratio rows on this page are built on contribution for exactly that reason.
- Retention is the load-bearing wall. Lifetime is 1 / (1 – retention): at 80% retention the average customer lasts 5 periods, at 90% ten, at 95% twenty. Small retention errors near the top of the range explode the lifetime, which is why the sensitivity table spans retention in both directions and why a lifetime beyond a few decades is flagged on the page as an extrapolation to treat as a ceiling.
- Measure retention on a cohort, not a mood. The right input is the share of a real group of customers still active one period later, measured at the same point in their lives as your model period. Blending new-customer churn (usually brutal) with mature-customer churn (usually gentle) into one rate builds a model of nobody.
- Set the discount rate like a grown-up. Money later is worth less than money now; the discount rate should reflect your cost of capital or your hurdle rate, in the same period as the model (per year for orders, per month for subscriptions, where 1% a month compounds to about 12.7% a year). Leaving it blank runs the discounted model at 0%, and the results say so.
- LTV to CAC is a ratio with two moving parts. The payback row (CAC divided by contribution per period) often matters more than the ratio: it says how long each new customer’s contribution takes to repay their acquisition cost, which is a cash-flow fact a young business feels in its bank account long before lifetime value arrives.
How it is calculated
- Revenue per period: average order value x orders per year, or the monthly subscription revenue. Contribution per period = revenue per period x gross margin.
- Churn = 1 – retention. Expected lifetime = 1 / churn, in periods.
- Simple LTV = contribution per period x expected lifetime. Simple revenue LTV = the same with revenue in place of contribution.
- Discounted LTV = contribution x retention / (1 + discount rate – retention), the standard steady-state discounted form, with contribution treated as arriving at each period’s end.
- LTV to CAC = LTV / CAC, shown for both models. Payback = CAC / contribution per period, in periods.
- Sensitivity: the discounted model recomputed at retention 10 and 5 points either side of yours and your margin 10 points either side, all other inputs held.
Worked examples
The repeat-purchase shop. An $80 average order, 4 orders a year, is $320 of revenue per customer per year; at a 60% margin the contribution is $192. Retention of 80% a year means a 5 year expected lifetime: simple contribution LTV is $960, revenue LTV $1,600. Discount at 10% a year and the discounted model gives 192 x 0.8 / (1.1 – 0.8) = $512.00. Against a CAC of $150 the ratios are 3.41 on the discounted model and 6.40 on the simple one, and the payback is 0.78 years. Same business, same inputs: the model choice alone halves the headline, which is why the model is always named on this page.
The subscription. $50 a month at a 70% margin contributes $35 a month. Monthly retention of 95% gives a 20 month expected lifetime (about 1.67 years): simple contribution LTV is $700, revenue LTV $1,000, and at a 1% monthly discount rate the discounted model gives $554.17. Note how much harder retention works monthly: a single point of monthly churn (95% to 94%) cuts the expected lifetime from 20 months to under 17.
The sensitivity tells the strategy. On the shop’s figures, moving retention from 80% to 85% lifts the discounted LTV from $512 to $652.80, while five extra points of margin (60% to 65%) would add proportionally less. And at 90% retention the same table shows $864: retention is the lever with the gearing, which the table proves on your numbers instead of asserting in general.
Reading your results
Quote an LTV only with its model attached (“$512, discounted at 10%, contribution basis”) or do not quote it at all; a bare figure invites comparison against someone else’s different model and a budget built on the gap. Then look at the spread between the simple and discounted figures: a wide spread means the value sits far in the future, where retention assumptions go to die, and decisions should weight the payback row more heavily. Finally, read down the sensitivity table’s retention column for your margin: if the business case only works in the top rows, the case is really a retention programme, and it should be funded and judged as one.
Limitations
- Steady state is a strong assumption. Real retention improves as cohorts age (the customers likeliest to leave do so early), revenue per customer drifts, and margins move. A full cohort model with period-by-period rates is the next step up; this page’s model is the honest first approximation, and its extrapolation warning fires when the implied lifetime leaves the range any plan should lean on.
- Retention at 100% is refused, not capped. It implies customers who never leave and an infinite lifetime value, which is not a big number but a broken input; the page stops and explains rather than printing a fantasy.
- CAC is taken as given. Whether your CAC is fully loaded (salaries, tools, agency fees) or media-only changes the ratio as much as the LTV model does; the page cannot audit your numerator, only insist it be compared with contribution.
- Service costs beyond gross margin are excluded. Support, success and retention spending sit below gross margin in most accounts and are not deducted here; a business with heavy service costs should treat contribution LTV as an upper bound.
Frequently asked questions
How is customer lifetime value calculated?
In the steady-state model used here: work out the contribution (revenue after gross margin) a customer produces per period, and how long they are expected to stay, which is 1 divided by the churn rate. The simple model multiplies the two. The discounted model applies contribution x retention / (1 + discount rate – retention), valuing each period’s contribution at its chance of happening and its distance in time. Both are shown, with the assumption of each stated, because the phrase “the LTV formula” hides at least these two legitimate answers.
What is a good LTV to CAC ratio?
The common rule of thumb wants contribution LTV at least 3 times CAC, with payback inside about a year for subscription businesses, but the rule presumes the discounted, contribution version of LTV. On the simple revenue version almost any business can manufacture a passing ratio, which is why this page computes the ratio on contribution and shows the model beside it. Treat 3:1 as a screening heuristic with known provenance, not a law of nature.
Why is revenue LTV shown at all if it misleads?
Because it circulates: vendors quote it, decks repeat it, and a reader who has only seen the contribution figure cannot tell which one they are being shown elsewhere. Displayed and labelled, revenue LTV is context (the gross size of the relationship). Used against CAC, it is an error, and the page’s structure makes that error hard to commit here.
What retention rate should I enter if I only know churn?
Retention is 100 minus churn, in the same period. A 5% monthly churn is 95% monthly retention and a 20 month expected lifetime. Keep the periods consistent: an annual retention rate in the monthly subscription mode (or the reverse) is the quietest way to be wrong by an order of magnitude, which is why the field labels on this page name their period and change with the mode.
Related calculators
Method: declared steady-state cohort model; simple LTV as contribution times 1 / churn; discounted LTV as contribution x retention / (1 + discount rate – retention); ratios on the contribution basis, exactly as defined on this page. Last reviewed: 13 October 2026.