Margin by channel is hard to calculate because its five inputs usually live in five different systems, and four of them arrive later than the sale does.
The five inputs
Revenue, from the channel. Discounts, which reduce it, recorded at the order. Channel fees, charged on the discounted price and usually revealed at payout. Shipping, which may be partly promised to the customer. Cost of goods, which depends on what the units actually cost to buy or build. Returns arrive later still and change all of the above.
Where the calculation usually goes wrong
The fee is estimated from the channel’s published rate rather than the actual charge. The cost is taken from an average rather than the specific units sold. The discount is netted into revenue and lost. Shipping is booked as an operating expense instead of against the orders that incurred it. Each shortcut is defensible on its own, and together they produce a number precise enough to be trusted and wrong enough to misrank two channels.
Why misranking is the real risk
The decision this number informs is where to put money next. If channel A appears to earn two points more than channel B and actually earns three points less, the brand spends a quarter scaling the wrong one. That outcome is worse than not having the number, because nobody questions a report.
What it takes to get right
Fees, discounts, shipping and returns attached to the orders they belong to, rather than summarized monthly. Cost from the actual units, which means cost layers and landed cost rather than an average. And all of it in one place, because a calculation assembled from four exports gets rebuilt monthly at best.
Focal carries fees, discounts, shipping and returns on the orders they came from, costs units at what they actually cost, and reports profit and loss by marketplace, fees by channel, channel contribution and cost of goods by SKU from the same records.
