I’d say most Amazon FBA profit margin calculations are wrong in the same direction, and by roughly the same amount. Sellers subtract the referral fee, the fulfillment fee and the unit cost, get a healthy-looking number, and never account for the five costs that arrive later: returns, storage, advertising, removals, and the unsellable inventory that eventually gets written off.
Those five are the difference between a 30% margin on a spreadsheet and a 12% margin in the bank.
The full cost stack
Work down this list. I’d not skip a row. Anything you skip appears later anyway, just without warning.
| Cost | When it hits | Typical size |
|---|---|---|
| Landed unit cost | At purchase | Product plus freight plus duty |
| Referral fee | Every sale | Around 15% in most categories |
| FBA fulfillment fee | Every sale | Set by size and weight band |
| Monthly storage | Every month held | Per cubic foot, much higher in Q4 |
| Long-term storage surcharge | On aged inventory | Punishes slow movers |
| Returns | 2 to 15% of orders | Fulfillment fee plus processing plus often the unit |
| 广告 | Ongoing | 8 to 15% of revenue on a mature listing |
| Removals and disposal | Occasionally | Per unit, on stock you give up on |
| Write-offs | Annually | Inventory that never sells |
The row I’d flag hardest is landed unit cost, because people use the factory quote. The factory quote is not your cost. Your cost is the quote plus freight plus duty plus any inspection plus the portion of a sample round attributable to the unit. On small first orders the difference between quote and landed is substantial.
A worked calculation
Take a product selling at $34.99, costing $7.20 from the factory, in a mid-size FBA band. Here is how I’d run it.
Start at revenue: $34.99.
Referral fee at 15%: minus $5.25.
FBA fulfillment fee for the band: call it minus $5.50. Look up your own rather than using this.
Landed unit cost, with the factory quote at $7.20 plus $1.60 freight and $0.50 duty: minus $9.30.
That leaves $14.94 contribution per unit, which is where most calculations stop and declare a 43% margin.
Now the five that get missed.
Storage, for a product turning over in roughly two months: minus $0.45.
Returns at a 6% rate, where a return costs the fulfillment fee plus processing and half the units are not resellable: minus $0.95 spread across all units sold.
Advertising at 12% of revenue: minus $4.20.
Removals and write-offs, amortized: minus $0.30.
Real contribution: $9.04, or 26% of the sale price. Still a decent product. But it is 26%, not 43%, and the difference decides whether you reorder.
The five costs people forget, in order of damage
广告 is the largest by far and the one most often modeled as zero. It is not a launch expense that goes away. Most mature listings still run 8 to 15% of revenue through ads permanently, because that is what it costs to hold position.
Returns hurt more than the rate suggests. A 6% return rate does not cost you 6% of revenue. It costs you the original fulfillment fee, the return processing, often the unit itself, and sometimes a negative review. Apparel and electronics run far higher rates than the average.
Storage compounds on slow movers. A product turning over in six weeks barely notices storage. A product sitting for eight months pays repeatedly and then attracts long-term surcharges.
Removals and disposal are the cost of admitting a mistake, and people defer them, which makes them worse by adding storage to the bill.
Write-offs are the honest annual accounting of inventory that will never sell. I’ve found sellers carry this on the balance sheet for years rather than recognizing it, which makes every margin figure optimistic.
Where the calculation goes wrong in practice
Beyond forgetting costs, three structural errors turn a correct method into a wrong answer, and I’ve seen all three repeatedly.
Averaging across the catalog. A blended margin across twelve products hides the two that lose money. Those two are usually funded by the winners and nobody notices, because the aggregate looks acceptable. I’d calculate per ASIN, always, and specifically look for the negative ones rather than at the total.
Using last year’s fee schedule. Amazon adjusts fees at least annually and the changes are not always small. A margin model built on an old schedule drifts quietly until something forces a recalculation, usually a cash flow problem.
Counting revenue before returns settle. A month’s sales figure includes orders that will be refunded weeks later. Looking at margin on gross orders rather than net of returns flatters every product, and the flattery is largest in exactly the categories where returns are highest.
The fix for all three is the same and it is not sophisticated: one row per ASIN, current fees, net of returns, recalculated every quarter. That is a spreadsheet, not a system, and in our experience it is the single most valuable spreadsheet an Amazon seller maintains.
One further thing worth doing once a year. Compare what the model says you should have earned against what actually landed in the account. The gap between those two numbers is real and it is informative, because it contains everything you are not tracking. A small gap means the model is sound. A large one means something on the cost stack is missing, and it is worth finding out what before you scale anything.
Contribution against net margin
Worth separating, because in our experience they answer different questions and get used interchangeably.
Contribution per unit is revenue minus everything that varies with the unit. It tells you whether selling one more unit makes money, which is the right question for pricing and for deciding whether to advertise harder.
Net margin subtracts your fixed costs too: the $39.99 seller account, software subscriptions, photography amortized across the product’s life, your own time if you pay yourself. It tells you whether the business makes money, which is the right question for deciding whether to continue.
A product can have healthy contribution and negative net margin if volume is too low to cover the fixed base. In our experience that is the most common shape for a first product, and it is why the second and third products are where profitability usually arrives: they share the same fixed costs.
The number that decides a reorder
Margin tells you whether a product is good. A different number tells you whether to buy more of it, and I’d keep them separate.
That number is return on the cash you tie up, across the time you tie it up for. A product returning 26% on a 90-day cycle turns your money four times a year. The same 26% on a product that takes eight months to sell through turns it once and a half. Those are not the same business, and the margin figure alone cannot tell them apart.
So when a reorder decision comes up I’d look at three things together. Contribution per unit, which says whether each sale is worth making. Sell-through speed, which says how often the money comes back. And whether the contribution is stable or drifting, which says whether the picture you are looking at is still true.
The third one catches the most people. A product with a slowly rising advertising ratio is losing margin every month while its historical figure looks fine, and in our experience nobody notices until the quarter closes badly.
What margin is enough
I’d want at least 20% after everything, and a minimum of $5 contribution per unit in absolute terms.
The percentage matters because it is your buffer against the things that move: fee increases, freight rate changes, a competitor cutting price, a return rate that turns out higher than modeled. Thinner than 20% and ordinary variance wipes out the profit.
The absolute figure matters separately. A 25% margin on a $12 product is $3 a unit, and $3 does not survive a single unexpected cost. High percentage on low price is a trap, which is why the FBA product research checklist screens on both.
Getting the numbers right
Three habits that make the calculation trustworthy, and I’d treat all three as standing practice.
Look up your actual fees rather than estimating. The FBA fulfillment fee depends on the size band, and the bands step sharply. Our guide on oversized items covers how much a boundary crossing costs.
Use your own return rate, not a category average. It is in your reports. Category averages hide enormous variation.
Recalculate quarterly. Amazon adjusts fees at least annually, freight rates move, and your advertising ratio drifts. A margin calculated eighteen months ago describes a product that no longer exists.
The last one is the habit I’d push hardest. Most sellers calculate margin once, at the point of deciding to buy, and then never again. In my experience that is how a product goes from profitable to loss-making without anyone noticing for two quarters.
FAQ
How do I calculate Amazon FBA profit margin?
Subtract the referral fee, FBA fulfillment fee and landed unit cost from the sale price to get contribution, then subtract storage, returns, advertising, removals and write-offs. Most calculations stop after the first three, which overstates margin substantially.
What costs do sellers forget when calculating FBA margin?
Advertising, which is ongoing rather than launch-only and typically runs 8 to 15% of revenue. Returns, which cost the fulfillment fee plus processing plus often the unit. Monthly and long-term storage. Removals and disposal. And annual write-offs of inventory that never sells.
What is a good profit margin for Amazon FBA?
At least 20% after every cost, and at least $5 contribution per unit in absolute terms. The percentage is your buffer against fee changes and competition; the absolute figure matters because a high percentage on a cheap product leaves too little to absorb any surprise.
What is the difference between contribution and net margin?
Contribution is revenue minus everything that varies with the unit, answering whether one more sale makes money. Net margin also subtracts fixed costs such as the seller account, software and photography, answering whether the business makes money.
Is the factory quote my unit cost?
No. Your landed cost is the quote plus freight, duty, inspection and a share of the sample round. On small first orders the gap between quote and landed cost is large enough to change whether a product is viable.
How often should I recalculate my FBA margins?
Quarterly. Amazon adjusts fees at least annually, freight rates move, and advertising ratios drift. Most sellers calculate once at the buying decision and never again, which is how a product becomes loss-making without anyone noticing.
Last updated: September 12, 2026. Amazon referral rates, FBA fulfillment fees, size tiers and storage charges change at least annually and vary by category and marketplace, so use current figures from Seller Central rather than the illustrative numbers here. ZonHack is an Amazon Ads verified partner and an Amazon SPN Verified Partner.
