What is landed cost and what belongs in it?

Landed cost is the full cost of getting a unit into your warehouse and ready to sell: the supplier price, inbound freight, duty, customs brokerage, insurance in transit, and any inspection or repackaging. Excluding freight and duty is the most common omission, and because it applies to every unit it moves gross margin by a meaningful amount on physical products.

Why should COGS not be booked as a percentage of revenue?

Because it makes margin an assumption rather than a measurement, and the assumption is usually inherited from a supplier price list that has since changed. A percentage estimate cannot show that one SKU is unprofitable while another subsidises it, which is exactly the information a paid media budget needs. It also means the inventory balance on the balance sheet drifts from reality with every period.

How often should inventory be counted?

Cycle counting continuously, rather than a full count once a year, is the workable approach for most e-commerce brands. Counting a rotating subset each week or month finds discrepancies close to when they occurred, when the cause can still be identified. An annual full count tells you the number is wrong without telling you why, and by then twelve months of margin reporting has already been based on it.

How do returns affect e-commerce inventory accounting?

Returned stock that goes back to sellable condition returns to inventory at cost; stock that is damaged, opened or out of season generally does not, and should be written down to what it will realise. Treating all returns as fully sellable inflates both the inventory balance and reported margin. A documented write-down policy is what keeps the judgement consistent between periods.

Does per-SKU margin need separate software?

Not necessarily, but it does need landed cost recorded per SKU and a reliable link between the sales channel and the item sold. Many brands can get there within their existing ledger and inventory system once the cost basis is corrected. The constraint is almost always the quality of the cost data rather than the reporting tool sitting on top of it.

E-commerce · 🇺🇸 United StatesScenario

Why e-commerce margin is wrong when COGS is a percentage estimate

Cost of goods sold booked as a percentage estimate produces a margin figure that cannot be used to make ad spend decisions, because the estimate hides both the freight and duty that belong in landed cost and the difference between ledger stock and what is physically in the warehouse. This is an illustrative example of how Salt approaches e-commerce inventory accounting, using a US brand across three warehouses as the profile.

Ask how we'd handle this

Scenario — an illustrative example of how Salt approaches this problem. Not a description of a specific client engagement.

The business shape it describes: US e-commerce brand, 3 warehouses, around 5,600 transactions a month.

Scenario covers: Bookkeeping, inventory reconciliation and per-SKU margin reporting
Jurisdiction: United States
Sector: E-commerce

The setup

Take a US e-commerce brand holding inventory in three locations, with around 5,600 transactions a month and no reconciliation of physical stock to the ledger.
Cost of goods sold is booked as a percentage of revenue, and paid media decisions are made against the margin that estimate produces.

What usually turns out to be wrong

Ledger inventory diverging from physical counts across all three locations, with no cycle count process to catch the drift as it happens.
Landed cost excluding inbound freight, duty and customs brokerage, which understates cost of goods sold on every unit.
Returns and damaged stock not written down, so inventory on the balance sheet includes units that will never sell at full price.
No per-SKU margin at all, so a product line scaled on paid media could be unprofitable at real cost without anything in the reporting revealing it.

How this would be worked

Reconcile each warehouse to the ledger and establish a recurring cycle count, so the difference is found monthly rather than annually.
Move to landed cost, capitalising inbound freight, duty and brokerage into inventory rather than expensing them separately.
Restate the affected periods so the margin trend is comparable rather than stepping at the date of the change.
Write down returns and damaged stock on a documented policy rather than leaving them at full cost.
Rebuild margin reporting at SKU level on landed cost, and put it in front of whoever sets the paid media budget.

Why we'd be the right fit

Stock held across three locations and a cost of goods sold that omits freight and duty are the two things that make e-commerce margin unreliable, and neither is fixed by better reporting on the same numbers.
Reconciling physical counts to the ledger and restating the affected periods is slow, unglamorous work that has to finish before a single per-SKU figure can be trusted.

What the business would be able to do

The brand would be able to see margin per SKU on landed cost, which is the number a paid media decision actually requires.
Inventory on the balance sheet would reflect what is in the warehouse and what it is worth, rather than a rolling estimate.
The gap between ledger and physical stock would be found by a cycle count each month instead of at year end.

What we'd flag

Restated margin is usually worse across most of the catalogue, because landed cost is always higher than an estimate that omits freight and duty. The restatement changes the understanding of the cash position, not the cash position itself.

Answers

Why e-commerce margin is wrong when COGS is a percentage estimate

Where this applies

BookkeepingService ControllerService All services in United StatesUS GAAP More scenarios42 illustrative, labelled examples

Start with a number

Recognise this shape in your own books?

Tell us where you trade and what shape the books are in. You get scope, price and a start date in writing within one business day — no obligation.

Get a quote in 24 hoursBook a call instead