The Retention Math Every Founder Should Know: LTV, Churn, and Repeat Rate

The short version: Retention math comes down to four numbers: repeat purchase rate, cohort retention, LTV on contribution margin and the payback window. A loyalty program pays back only if the contribution from the incremental orders it creates exceeds the rewards it hands to customers who would have returned anyway, so estimate that incrementality against a randomized holdout.

Loyalty conversations go wrong when they run on vibes — "engagement," "delight," "community." Getting retention right comes down to four numbers. None of them are complicated; all of them are easy to compute wrong.

Key takeaways

• Count repeat purchase rate only on cohorts old enough to have finished the window, and watch your own trend rather than a borrowed benchmark.
• Read retention by acquisition cohort; a blended rate can hide cohort losses.
• Compute LTV on contribution margin and per acquired customer, not per repeat buyer.
• Set your payback ceiling from your own runway; no universal band tells you what is acceptable.
• Judge a loyalty program by the incremental orders it creates against a randomized holdout, and charge its rewards against contribution, not revenue, in a cash and liability budget.

1. Repeat purchase rate (RPR)

Repeat purchase rate is the share of first-time buyers who place a second order inside a stated window, counted only on cohorts old enough to have finished that window. It varies by category, so watch the direction of your own number, not a borrowed benchmark. No incentive can rescue a product nobody wants twice.

2. Customer retention rate — measured on a cohort, not a blend

Customer retention rate is the share of a starting group of customers still buying at the end of a period, and cohorts help separate acquisition timing from continued buying. A blended rate can hide cohort losses. If you acquired heavily last month, your "average" churn looks great while every cohort is quietly leaking.

The period version is retention rate = (customers at end − new customers in period) / customers at start. A store without subscriptions has to define who counts at the end; this article counts anyone with at least one order in the period. Whatever definition you pick, keep it fixed. The version to act on is the cohort: of customers acquired in month X, how many were still active in month X+n?

To build the cohort grid, export orders with three columns: customer ID, order date and net order total. Use enough prior order history to identify true first orders, and record the observation date. Give each customer the month of their first order; that is their row. Columns are months since first order, and each cell is the share of the row with at least one order in that month. Keep the original cohort size as the denominator and leave not-yet-observed months blank; do not turn missing follow-up into zero activity. These cells count that month alone, not cumulative activity. Plot three cohorts and you will learn more than from a year of blended dashboards.

3. LTV — with margin, not revenue

The first LTV inflation to check: using revenue instead of contribution margin. A $300 revenue LTV at 25% margin is a $75 customer. If acquisition costs $60, you are running a very tight boat while your dashboard celebrates. Loyalty rewards come out of that margin too — a 2% earn rate on a 25%-margin business consumes 8% of your profit pool.

4. Payback window

Payback window is how long until a cohort's cumulative contribution covers what you paid to acquire it, counted per acquired customer rather than per repeat buyer. No universal band tells you what is acceptable: the ceiling is the number of months of cash you are prepared to fund before the money comes back, so set it from your own runway. Retention is a direct lever on this number, because every extra order lands inside an already-paid-for relationship.

The uncomfortable conclusion

A loyalty program is a margin reallocation: you tax every transaction to change future behavior. It pays back only if the contribution from the incremental orders it creates exceeds the discounts it hands to customers who would have returned anyway. Estimating that incrementality — not the point balance, not the signup count — is the entire game, and it is why I recommend a properly randomized holdout group from launch: customers who were not offered it provide a comparison for estimating extra orders. Size the split to detect the break-even lift; if volume is too low, extend the test and treat an inconclusive result as directional. For the full incremental calculation, see how to calculate loyalty program ROI.

If you want to run these numbers against your own figures, the retention calculator does the arithmetic above, including what one percentage point of retention is actually worth to you.

The retention math on one business

Take a store with 1,000 customers acquired in January, an average order value of $80, and a 25% contribution margin after payment fees, fulfillment and returns, so each order contributes $20. These inputs are illustrative; replace them with your own export. Every figure below is read 12 months after first order, so the January cohort is read the following January.

Repeat purchase rate: 240 of them place a second order within those 12 months, so RPR is 24%. Whether that is good is a question for your own trend, not for a benchmark.

Cohort churn: in month six alone, 40 of the original 1,000 place an order, so that cell of the grid reads 4%. That fits the rest of the cohort: its 240 repeaters place 240 × 2.2 = 528 repeat orders in the year, about 44 a month (528 ÷ 12), and one customer can order more than once in a month. Read the cell as 4% of the cohort active in month six, not as "96% churn", which sounds like a crisis and describes nothing you can act on.

LTV: the 240 repeat customers average 3.2 orders in their first 12 months, and the other 760 bought once. Multiplying 3.2 orders by $20 gives $64, but that is the LTV of a repeat buyer, and three in four customers never become one. The number to set against acquisition cost is LTV per acquired customer:

LTV per acquired customer = (one-time share × 1 + repeat share × orders per repeater) × contribution per order

Here that is (0.76 × 1 + 0.24 × 3.2) × $20 = 1.528 orders × $20 = $30.56 of contribution, or $122.24 of revenue. If acquisition cost $45, the cohort loses $14.44 per customer, $14,440 across the 1,000, in its first year. The business does not work at these inputs.

Any one of four changes reaches break-even with the others held fixed: acquisition cost at or below $30.56; a repeat share of at least 56.9% instead of 24% (1 + 2.2 × share = 2.25 orders); repeaters averaging at least 6.21 orders instead of 3.2 (1 + 0.24 × (orders − 1) = 2.25); or at least $29.46 of contribution per order ($45 ÷ 1.528), which at a 25% margin means an average order near $118.

Payback: at $20 of contribution per order, a repeat customer clears $45 at the third order ($60 cumulative, against $40 after two). On an illustrative six-week buying interval the third order lands around week 12, about month three. The cohort as a whole does not pay back inside the year, because its average customer contributes $30.56 against $45. A payback figure quoted for the repeat buyer describes a quarter of the cohort.

Now add a loyalty program earning 2% of spend on every order. That is $1.60 per $80 order against $20 of contribution — 8% of the margin pool, charged on every order including the ones that needed no incentive at all. On this cohort's 1,528 orders the reward cost is 1,528 × $1.60 = $2,444.80, and LTV per acquired customer falls to 1.528 × $18.40 = $28.12.

To cover that cost, the program has to create $2,444.80 ÷ $18.40 = 133 orders that would not otherwise have happened, about 8.7% more orders from the same 1,000 customers. If points only count from the second order, there are 240 × (3.2 − 1) = 528 eligible orders, the cost is 528 × $1.60 = $844.80 and the bar is 46 extra orders, assuming the extra orders also earn rewards. Either way, a properly randomized holdout is the recommended comparison for estimating whether those orders were incremental; a small sample may not resolve the required lift.

Frequently asked questions

How many cohorts do I need before cohort churn means anything?

The constraint is cohort size, not cohort count. A cohort of forty customers moves several points of retention when two people leave, so the noise is larger than the signal you are looking for. Plot three consecutive cohorts and read the shape rather than the decimal. If your monthly volume is small enough that single-digit churn swings the line, widen the cohort to a quarter instead of reading noise as a trend.

Gross margin or contribution margin in the LTV calculation?

Contribution: gross margin minus the costs that scale with an order — payment fees, fulfillment, shipping subsidy, returns, and the reward itself. Gross margin flatters LTV precisely because it hides the costs that grow with every extra order a loyalty program buys you. If contribution margin is not available yet, run the number both ways and treat the gap between them as the size of your uncertainty rather than picking the friendlier one.

How large does the holdout need to be?

Large enough to detect the effect you would act on, which depends on your base rate and that effect — not on a fixed percentage. Work backwards: if repeat rate is 25% and you would only keep the program for a three-point lift, the holdout has to be big enough for three points to be distinguishable from ordinary variation. If that sample exceeds your monthly volume, run the holdout for longer rather than shrinking it, and accept that you are measuring a quarter rather than a month.