Amazon FBA vs FBM: the comparison most sellers get wrong
The usual version of this comparison puts the FBA fulfilment fee next to what a parcel costs you to post, and concludes that FBM is cheaper. On that arithmetic it usually is. The arithmetic is just missing about half the cost on each side.
What the FBA fee actually buys
The fulfilment fee is not postage. It covers picking, packing, the shipping itself, customer service in the local language, and — the expensive part — return processing. Amazon handles the refund conversation, the inbound label and the inspection.
It also buys the Prime badge, which is not a cost item at all but a conversion multiplier. On competitive listings the difference between Prime and non-Prime is routinely the difference between selling and not selling. That effect does not appear in any fee table, and it is often larger than the fee.
What FBM actually costs
Sellers comparing the two rarely price their own side properly. The honest list is longer than the postage:
- Packaging — box, filler, tape, label. Small per unit, real at volume.
- Your time. Ten minutes per order at any sensible hourly rate is usually more than the FBA fee for a small item.
- Customer service. Where is my parcel, in a language you may not speak, within the response window Amazon measures you on.
- Returns. Return postage, inspection, repackaging, and the units that come back unsellable.
- Late shipment and defect metrics. Missing them costs the Buy Box, which costs sales.
Put those in and the gap narrows sharply for small, light, fast-moving products.
Where each model wins
FBA wins on small, light items with steady turnover. The fee is low relative to the price, storage is cheap because stock moves, and the Prime badge does the most work in exactly the categories where buyers compare on delivery speed.
FBM wins on large, heavy or slow-moving goods, on anything fragile or made to order, and on products with a wide range of variants. Bulky items attract high fulfilment fees and punishing storage costs; slow movers accumulate long-term storage surcharges while sitting still.
The dividing line is roughly: does this product turn over fast enough that storage stays trivial, and is it small enough that the fee stays a modest share of the price? Two yeses point to FBA.
The storage trap
Storage is charged per cubic metre per month, and it rises sharply for inventory that ages past the long-term threshold. A product selling three units a month while holding six months of cover is quietly paying rent the whole time.
This is where FBA turns from cheap to expensive without anyone noticing, because the storage line arrives monthly and separately from the per-unit fee people modelled. If you are choosing between models, model storage over the real turnover period, not over one month.
How to actually run the numbers
Do it per product, never for the account as a whole. Run the Amazon FBA calculator twice with the same selling price:
- FBA: enter the fulfilment fee from your fee schedule, plus a monthly storage figure divided by expected monthly units.
- FBM: put your real shipping cost in the fulfilment field, and add packaging and a per-order allowance for your time into product cost.
Compare the profit per unit and the break-even price. The break-even is the more useful number: it tells you which model survives a price war, and that is usually what the decision comes down to. The break-even calculator is built for that second half of the reasoning.
Then adjust the FBM side downward for lost conversion. If you do not know your number, testing the same listing both ways for a fortnight will tell you more than any benchmark.
The answer is usually both
There is no rule requiring one model per account. Most sellers past a certain size run their fast small products through FBA and keep the bulky, slow or fragile ones on FBM.
Some also keep a small FBM listing alongside FBA as a fallback when stock runs out at the fulfilment centre, so the listing stays live rather than going dark during a restock. That is worth more than it looks: a listing that disappears loses ranking that takes weeks to rebuild.
One thing to keep in mind before splitting across countries: if you let a programme move your stock into other member states, you acquire VAT registration obligations there. That cost belongs in the FBA column too — see OSS or local registration.
Frequently asked questions
It can. Buy Box eligibility weighs delivery speed and seller performance, both of which FBA effectively guarantees. A well-run FBM operation with fast dispatch competes fine; a slow one does not.
Often not. The fulfilment fee is close to flat while the price is not, so on cheap products it eats a large share of the margin. Bundling into multipacks is the usual answer: it raises the price without raising the fee proportionally.
FBA absorbs the process but you still lose the unit value and the fulfilment fee on returned orders. FBM gives you control over inspection and resale, at the cost of doing the work. In high-return categories such as apparel, that control is worth real money.
Yes, and it is reversible. Test rather than theorise: run a fortnight each way and compare units sold as well as profit per unit, because the conversion effect is the part no spreadsheet can predict for you.