If your customers buy again, the ACoS you can profitably run on the first order is much higher than first-order math suggests — because later orders subsidize the cost of acquiring the customer. A one-time-purchase product has to make its money on order one. A repeat-purchase product is buying a relationship, and that changes everything about what you can afford to bid.
This is the edge most Amazon sellers leave on the table. They set allowable ACoS off a single transaction, get outbid by competitors who understand lifetime value, and never know why they lost the placement. Here's the math, and how to use it without kidding yourself.
Quick note: I run a fractional Amazon team, and cohort economics is one of the few genuine advantages a smaller brand can build over a bigger, lazier competitor. It rewards the operator who does the homework.
Why First-Order Math Misleads
Most sellers evaluate an ad this way: did this click, on this order, make money after fees and ad spend? That's the right question for a product nobody rebuys. It's the wrong question for a product they do.
When a customer repurchases, the first order isn't the whole transaction — it's the down payment on a stream of orders. Judging it on order-one economics alone systematically underspends on acquisition, cedes the top placements to whoever's willing to lose a little on the first sale, and caps your growth below what your unit economics could actually support.
On a repeat-purchase product, you're not buying a sale. You're buying a customer — and pricing the bid as if it's a sale is how you lose the customer to someone who knew the difference.
What Customer LTV Really Means on Amazon
Customer lifetime value (LTV) is the total contribution margin a customer generates across every order they place — not just the first. If a customer's first order contributes $12 and they reorder three more times at $14 each, their LTV contribution is $54, not $12. That's the number that should anchor acquisition spend.
The honest caveat: LTV is harder to measure on Amazon than in DTC, because you don't own the customer data. But you have enough to estimate it well — Subscribe & Save enrollment, Brand Analytics repeat-purchase and customer-loyalty metrics, and cohort analysis of your own order data. You don't need perfect; you need directionally right, tracked over time.
It also helps to be deliberately conservative with the estimate. Because LTV is uncertain, the safe move is to count only the repeat behavior you can actually see in the data — not the loyalty you hope to build. Discount future orders slightly, assume some churn, and set your acquisition budget against the lower end of the range. That way an aggressive first-order ACoS is still protected if a cohort underperforms. The brands that get burned by LTV-based bidding are almost always the ones that bid against optimistic projections; the ones that win bid against measured, slightly-discounted reality and let pleasant surprises be upside rather than the plan.
Reading Your Repeat-Purchase Cohorts
A cohort is a group of customers who first bought in the same period. Track what share of each cohort comes back, and how quickly, and you can put a number on LTV instead of guessing.
| What to measure | What it tells you |
|---|---|
| Repeat rate within 90 days | How fast the relationship pays back acquisition cost |
| Orders per customer, 12 months | The multiplier on first-order contribution margin |
| Subscribe & Save share | Locked-in, predictable repeat revenue |
| Cohort trend over time | Whether retention is improving or quietly eroding |
Consumables — supplements, coffee, grocery, pet, beauty refills — often see 30–50%+ of customers repurchase within 90 days. Durable one-time purchases may see almost none. There's no universal benchmark; the point is to measure your rate by cohort and watch the trend. Subscribe & Save is the clearest signal, which is why it deserves its own strategy.
How LTV Changes Your Allowable ACoS
Here's the payoff. Your normal allowable ACoS is set off a single order's contribution margin. An LTV-adjusted allowable ACoS is set off the customer's lifetime contribution — so it's higher, sometimes dramatically.
| First-order view | LTV-adjusted view | |
|---|---|---|
| Contribution counted | Order 1 only ($12) | Lifetime ($54) |
| What you'll spend to acquire | A fraction of $12 | A fraction of $54 |
| First-order ACoS you can run | Tight | Much higher — even at or above first-order break-even |
| Who wins the placement | Whoever bids highest | You, profitably |
Running above first-order break-even sounds reckless until you remember you're not trying to profit on order one — you're trying to acquire a customer whose lifetime margin more than covers it. That's a deliberate, measured trade, and it's exactly the kind of intentional overspend the break-even vs allowable ACoS framework is built to make on purpose rather than by accident.
Which Products Justify Outbidding
This edge is real but conditional. It only works where the repeat behavior is real.
How to Actually Use This
Turn it into a repeatable decision, not a vibe:
Done right, this is how a disciplined brand profitably outbids a bigger, sloppier one for the same customer — and it compounds, because every retained cohort raises what you can afford to spend on the next. It's the same margin-first thinking behind the Amazon profitability playbook, and it's the kind of edge a fractional Amazon team is built to find and run.
Treat it as a compounding advantage, not a one-off tactic. Every cohort you retain raises the lifetime value you can bid against for the next one, so the gap between you and a competitor who only looks at first-order ACoS widens over time rather than closing. That's the quiet reason disciplined brands pull ahead in categories everyone assumes are commoditized.
The one-line version: measure lifetime contribution margin, set your allowable ACoS off the customer instead of the order, and spend to acquire the customers who come back.
FAQ
If a customer buys again, the profit from later orders subsidizes the cost of acquiring them on the first one. A product with strong repeat purchase can profitably run a higher first-order ACoS — sometimes at or above first-order break-even — because you are buying a customer, not just a sale. A one-time-purchase product has to make its money on order one, so its allowable ACoS is much tighter.
Customer lifetime value is the total contribution margin a customer generates across all their orders, not just the first. On Amazon it is harder to measure than in DTC because you do not own the customer data, but Subscribe & Save, Brand Analytics repeat-purchase metrics, and cohort analysis of your own order data give usable estimates. LTV is what justifies spending more to acquire a repeat-purchase customer.
It varies enormously by category. Consumables like supplements, coffee, and grocery can see 30-50%+ of customers repurchase within 90 days; durable one-time purchases may see almost none. What matters is measuring your own rate by cohort and tracking whether it is rising or falling, not hitting a universal benchmark.
Often yes. Subscribe & Save locks in repeat orders, which raises the lifetime contribution margin of each acquired customer and therefore the first-order ACoS you can justify. The key is to base the decision on measured repeat rate and retained margin, not hope — an aggressive acquisition ACoS only pays off if customers actually stick.
Estimate the total contribution margin an average customer delivers across their expected orders, then decide how much of that lifetime margin you are willing to spend to acquire them. That target acquisition cost, expressed against first-order revenue, is your LTV-adjusted allowable ACoS — which is typically higher than a first-order-only allowable ACoS for repeat-purchase products.