Profit per order vs profit per month: why sellers overestimate their margin by a third

Ask a marketplace seller their margin and you will usually get a confident number in the thirties. Ask to see the bank balance a year later and it tells a different story. The gap is not dishonesty and it is rarely bad arithmetic. It comes from looking at the business one month at a time, in totals, and from a handful of habits that feel sensible and systematically flatter the result. This guide shows the gap with numbers, then explains where it comes from. The "third" in the title is what the worked month below produces from ordinary omissions; it is an illustration, not a survey result, and your own gap may be smaller or larger.

The month as most sellers see it

A seller with one Amazon store takes the month's payouts, their purchase invoices and a feel for fees, and builds this:

Thirty-five per cent. It is the number they will quote, price against and plan around.

The same month, with everything in

The seller believed 35% and the month made 23%. Twelve points out of thirty-five is a third of the margin they thought they had, which is where the title comes from. Every one of those lines is routine; none is a mistake in the sense of a wrong number, and all of them are real cash leaving the business.

Why the monthly view hides it

The payout feels like profit. Money arriving from Amazon has already had fees deducted, so it reads as "clean". It is not: it is gross sales less some fees, before stock, before ads, before returns that have not yet come back. Treating the deposit as the headline figure is the single most common error in seller bookkeeping, and the payout reconciliation guide exists because of it.

Totals average winners and losers. A month is a sum. Four products can sum to a comfortable profit while one of them loses money on every order:

The month looks fine. The jar loses £132 and the spatulas earn almost nothing for 400 orders of work, storage and risk. Nobody sees it because nobody looks at the order. Price the jar up, or stop selling it, and the same month earns 7% more with fewer parcels.

Costs arrive late. Returns come back two to five weeks after the sale; retailers expected 19.3% of online sales to be returned in 2025. Storage is charged monthly and triples in the last quarter. The aged-inventory surcharge lands months after the purchasing decision that caused it. A month's profit statement is drawn up before most of that month's costs exist.

Some costs do not look like costs of sale. Advertising on a company card feels like marketing. Software feels like overhead. Currency margin feels like a bank thing. Each has a home in the accounts that is not "cost of goods" or "fees", and so each is left out of the margin sum even when it is in the accounts.

The best case becomes the plan. The margin calculator in Seller Central is a good tool, but it answers for a unit that sells, is not returned, was not advertised and did not sit in storage. It is a ceiling, and people remember ceilings as averages.

Memory keeps the winners. The products that went well are the ones you think about, reorder and talk about. The ones that lost money were quietly cleared and forgotten, and they are not in the number you quote.

What to do instead

• Attach every cost to an order where it belongs. Commission, fulfilment, the refund administration fee, the ad click that caused the sale, the storage the unit used. What cannot be attached (the subscription, the accountant) stays monthly, but it is then a known overhead, not a surprise.
• Look at profit per order and per product first, and the month second. The month is the sum of the orders; it cannot tell you anything the orders do not.
• Hold back for returns. Expect a percentage of each month's sales to come back next month and provision for it, rather than being surprised in arrears.
• Separate advertised and organic sales. A product can be profitable organically and loss-making when every sale needs a click. Both numbers are true; only the split tells you what to do.
• Keep the fee, ad and storage numbers from the statements, not from memory. Rate cards drift and so does recall.

Merchvio exists for the first of those: it attaches each fee, refund, ad charge and the unit's landed cost to the order they belong to, and shows the month as the sum of its orders, so the glass jar is visible in the list before it has cost a year of margin.

The fee, return and advertising figures used in the examples come from Amazon's published UK rates and the industry sources below, checked in October 2026. The monthly figures are an illustration built from typical omissions, not a study; the point is the method, and your own numbers will differ.