When does a customer pay back what they cost?
CM payback is how long a new customer takes to earn back their acquisition cost in contribution margin - the money actually left after discounts, landed COGS, shipping and fees. Divide CAC by contribution margin per order for payback in orders, then use purchase frequency to convert it to days. Revenue payback flatters everyone; margin payback is the number cash flow actually feels.
Blended or paid, per new customer. Need to build it? Use the blended CAC calculator.
What an order leaves after discounts, landed COGS, shipping and fulfilment, and payment and platform fees. The contribution margin calculator steps it down.
Trailing 12 months of orders divided by the unique customers who placed them, from your store's order export. Decimals are normal: 1.3 means the average customer orders 1.3 times a year.
Your average customer pays back their acquisition cost on order 2, roughly 365 days in.
That is an average. Cohorts and channels pay back at different speeds - treat it as the centre of the distribution, not a promise.
Cumulative contribution margin after each order, measured against your CAC. The bar fills as the customer earns their cost back; the teal row is where it lands. Computed from your inputs, nothing assumed.
Three inputs, and where to find them
Every number here already exists in your stack. Pull all three from the same trailing period so the answer is internally consistent.
Pull your CAC
Total acquisition spend for the period - ad platforms, agency fees, affiliate and promo, from your ad accounts and marketing P&L lines - divided by the new customers that period brought in, counted from your order data. If you need to build it properly, the blended CAC calculator does the division.
Pull your CM per order
AOV minus discounts, landed COGS (cost plus freight-in plus duty, from supplier invoices and freight bills), shipping and fulfilment (your 3PL invoice), and payment and platform fees (your processor statement). The contribution margin calculator runs the step-down.
Pull your purchase frequency
Orders in the trailing 12 months divided by the unique customers who placed them, straight from your store's order export. Use the trailing year, not launch-month cadence, or the days number will read faster than your cash ever will.
The whole method is two divisions
CAC is paid once, up front. Contribution margin comes back one order at a time. Payback is simply how many orders that takes, and how long those orders take to arrive.
Same margin definition as The Math: contribution margin per order is what is left before marketing, after discounts, landed COGS (cost plus freight-in plus duty), shipping and fulfilment, and payment and platform fees. The Math writes the same payback on a monthly clock - CAC ÷ monthly contribution margin per customer - and its verdict holds here: margin tells you if, payback tells you when.
A new customer costs $60.00 to acquire. Each order they place leaves $30.00 of contribution margin after discounts, landed COGS, shipping and fulfilment, and payment and platform fees.
Orders to pay it back: $60.00 ÷ $30.00 = 2.0 orders. At 2 orders per customer per year, that is 365 × 2.0 ÷ 2 = 365 days from acquisition: the cost comes home on order 2, roughly 365 days in. Until that point the customer is a use of cash, not a source of it.
These are your live inputs from the calculator above, not a canned example. Change a number up there and this paragraph follows.
Margin tells you if. Payback tells you when.
A customer paid back quickly funds the next customer. The same unit economics paid back slowly lock your cash away in the meantime. Both can be profitable on paper; only one of them feels profitable at the bank.
How hard you can push
Payback days set how much acquisition the operating account can carry. Short payback and the same dollars recycle into new customers several times a year; long payback and every extra customer widens the cash gap before it closes it.
Who is funding growth
If payback runs past what your cash cushion can wait out, growth is being funded by working capital, not by customers. That can be a deliberate choice. It should never be a surprise you discover in the bank balance.
Catching drift early
Recompute it monthly and watch the direction. A lengthening payback is CAC inflation or margin decay showing up in a single number, months before the P&L makes it obvious.
One average, honestly labelled
Because revenue payback flatters everyone
Counted in revenue, most customers look recovered on their first order, because AOV usually dwarfs CAC. But most of that revenue is already owed to product, freight and fees before it ever touches the acquisition bill.
Counting payback in contribution margin - only what the order genuinely leaves behind - is the honest version, and it is the one your bank balance agrees with. It turns a feel-good metric into a cash planning tool you can run on three numbers.
Never for channel decisions
A single blended payback cannot allocate budget. The average mixes fast and slow cohorts, so it can read fine while one channel never pays back at all. Purchase frequency is measured over customers who stayed, so the newest cohort usually runs slower than the average says. And it assumes every future order carries the same margin, which discount-led reorders do not.
What we do instead: when the decision is where the next dollar goes, we stop using the blended average and read payback per cohort and per channel, on reconciled data.
Payback, per cohort and channel
This page runs one blended average. Blufire S2 Unit Economics computes CM payback and NCAC per cohort and channel, on reconciled data, and its time-gated payback windows unlock as calendar days accrue: 30, 60, 90, 180 and 365 days. No cohort gets judged before its window has actually elapsed.
The protocol
Recompute on the trailing 12 months. CAC, margin and frequency all drift with mix, so last quarter's payback is not this quarter's. File the number: the trend is the point.
Re-run it with the CAC you are about to pay, not the trailing average - the cost of the next cohort, not the average of the last. Payback at the margin is what the extra spend actually buys.
Model the launch offer's real margin. A discount-led launch lowers contribution per order and stretches payback even when CAC holds still.
Payback lengthens two months running, or crosses 12 months - the line The Math judges LTV against. Past it, acquisition is funded by working capital, not by the customer: fix CAC, fix margin, or fund the gap deliberately.
Questions operators ask
There is no universal good payback: the right window is set by your margin structure and how long your cash can wait, not by a league table. The Math judges LTV alongside sub-12-month payback as a working line, because past a year the customer is being carried by working capital rather than carrying you. Inside that, shorter is simply better - the same dollars recycle into the next customer sooner.
Because only margin can pay the acquisition bill: revenue arrives already owed to product, freight and fees. Revenue payback marks customers recovered on order one and flatters every channel equally. Margin payback counts only the dollars that are genuinely left, which is why it lands later, reads harsher, and matches what your bank balance does.
Use blended CAC for the whole-business cash question this tool answers, and paid CAC when you are judging the paid program specifically. Blended CAC is total sales and marketing spend divided by all new customers; paid CAC is paid media spend divided by customers from paid. The blended CAC calculator builds the first one properly.
No: it assumes the average customer keeps ordering at the average frequency. Customers who never place a second order stretch the true payback beyond what the average shows, because the frequency input is dominated by the customers who stayed. That gap is exactly why anything channel-shaped should be read as cohort payback on real order dates, not a blended estimate.
Then every new customer locks cash up for more than a year, and acquisition is being funded from working capital. Check the inputs first - a too-thin contribution margin, an inflated CAC or an overstated purchase frequency are the usual suspects. If the inputs hold, the choice is deliberate: fix margin or CAC, slow acquisition to what cash can carry, or fund the gap knowingly.