Landed Cost: Why Your Store Knows Its Revenue but Not Its Profit

Here is a store's last quarter: revenue up two and a half times on a few dozen orders, with the average order value up by even more.

Every one of those numbers is real, and not one of them tells you whether the business made money.

Revenue and orders are certain. Under them: the thousand-plus SKUs that sold nothing in the same window.

Revenue is the easy number

Marketplaces hand you revenue because it's the number they can compute without knowing anything about your business. It's also the number most flattering to look at, and the least useful for a decision.

The same quarter contains two facts that matter more. One sale produced most of that growth — take it out and the period is flat. And almost the entire catalog sold nothing at all: a thousand-plus listings occupying shelf space and telling the marketplace's ranking that this store is thinner than its listing count suggests.

Both facts sit inside the revenue number, invisible.

Profit needs landed cost, and nobody hands you that

Ask this store what it earned and the honest answer is: nobody knows.

"Profit: No cost data (none of these products are linked to a supplier), so true margin is unknown."

That's SellerClaw — a team of AI agents connected to your stores, suppliers and ad accounts — answering in a chat about the store it runs. It prints "unknown" rather than a zero or a plausible-looking estimate, because a made-up margin is worse than no margin: you'd price against it.

The missing input is landed cost — what the unit actually costs you delivered, not the supplier's headline price.

The distinction is the whole problem. The headline price is the number that is easy to find and the number people put in spreadsheets, and it is reliably too low. Getting the item from the supplier to the place it ships from costs something, and that cost does not scale the way the unit price does: it moves with order size, with the warehouse it comes from, and with the currency it was quoted in. Two units of the same product, bought the same week from two warehouses, do not have the same landed cost.

Which is why a per-product cost figure maintained by hand goes stale faster than anyone expects. It was correct for one order.

Connect the supplier the products come from and cost stops being a spreadsheet you update on Sundays. Every number after that — margin per product, which listings are worth advertising, whether a discount is survivable — becomes computable rather than estimated.

What actually goes into landed cost

There is no single formula that fits every business, which is why a landed cost calculator asked to produce one usually produces a wrong one. What there is, is a list of things that have to be in the number before it means anything.

The unit price the supplier quotes. The easy part, and the only part most people record.

Getting it to where it ships from. Inbound shipping, and whatever the supplier charges for handling or packing. This is the component that breaks the arithmetic, because it does not divide evenly: shipping quoted per order, spread across a mixed carton, gives every item in that carton a different share depending on what else travelled with it.

The currency it was bought in. If the supplier quotes in one currency and you sell in another, the cost of a unit bought in March is not the cost of the same unit bought in August. A cost figure recorded once, in the wrong currency, silently ages.

Which warehouse it came from. Suppliers with more than one location do not price them the same, and the cheaper unit is often the one with the slower or dearer route to you.

Notice what is not in that list: marketplace fees, shipping to the buyer, ad spend, returns. Those are real costs and they come out of the same money, but they belong after the sale, not in the cost of the goods. Mixing the two is how a seller ends up with a "margin" that changes depending on which channel the item happened to sell on, and no way to compare two products.

Keep the two groups apart and each answers a different question. Landed cost tells you whether the product is worth selling at all. Everything after it tells you whether the channel is worth selling it on.

Where the rest of the money goes

Revenue minus landed cost still isn't the end of it. Marketplace fees, shipping, ad spend and payout timing all sit between the sale and your bank account.

The last tile on the right: no ad spend in the window, so ROAS is a dash — not a zero.

That's what the finance reports are for: where the week's revenue went, fee by fee; the month's P&L; payouts received, expected and still on hold. They generate on a schedule, so the question "did this week make money" doesn't require an evening with a spreadsheet.

Payout timing deserves its own mention, because it is the one that surprises people who have done the margin arithmetic correctly. Money that is earned, real and yours can still be money you cannot spend this month — and a business that is profitable on paper and short of cash makes worse decisions than one that is neither.

Where a number can't be honestly derived, it isn't drawn. A report with no ad spend in the window shows a dash for ROAS, not a zero — the same rule as the margin above.

What changes for you

Today you have a revenue number you half-trust and a profit number you compute occasionally, by hand, for a subset of products. That's not a discipline problem — nothing in your stack knows what your goods cost.

Connect one store and one supplier — free credits cover it — then ask which of your products actually make money. The list is usually shorter than people expect, and it is the most valuable list in the business: it tells you what to restock, what to advertise, and what to stop shipping at a loss.