D2C Unit Economics and Contribution Margin 2026

Gross merchandise value is the value of orders placed on the platform. Net revenue is what survives returns and cancellations. Contribution margin is what remains after every variable cost of making that sale.

The gap between the first and third is wide, and it widens in exactly the categories that look most attractive on top-line growth. A brand reporting strong GMV growth while contribution margin is flat or negative is buying revenue, and the books should make that visible long before the board does.

Blended contribution margin is a comfortable number and a largely useless one. The value comes from the distribution behind it.

In most catalogues we see, a minority of SKUs generates the large majority of contribution, a substantial middle group roughly breaks even, and a tail contributes negatively once storage, returns and advertising are attributed properly. That tail is usually invisible in aggregate reporting, and it is often the part of the catalogue receiving the most operational attention.

Contribution margin computed once a quarter for a board pack is a reporting exercise. Computed monthly as part of the close, it becomes an operating tool, because the corrective actions all have lead times: repricing, delisting, repackaging to move a weight slab, or clearing ageing stock before storage fees escalate.

It also has a direct funding consequence. Investors testing unit economics in diligence will rebuild this from the raw settlement files themselves. A brand that already reports it consistently, and can tie it to filed GST returns and to the bank, is in a materially stronger position than one presenting a number assembled for the occasion.

No. Gross margin stops at cost of goods sold, while contribution margin also absorbs the variable costs of making the sale through a given channel, including marketplace fees, logistics and attributed advertising. For a marketplace business those costs are large enough that gross margin alone is misleading.

For D2C and marketplace selling, yes, because advertising functions as a direct cost of acquiring the sale rather than as a general overhead. Attribute it by channel, and by SKU wherever the platform's reporting makes that possible, since blended attribution hides which products actually require paid support.

Platform storage reports generally give volume and ageing by SKU, which is enough to attribute cost proportionately rather than spreading it evenly. Even attribution flatters slow-moving stock at the expense of fast-moving lines, which inverts exactly the signal the analysis is meant to produce.

There is no single benchmark, since it varies widely by category, price point and return behaviour. The more useful discipline is tracking the direction of your own margin by channel and by SKU over time, and knowing which cohort of products funds the business rather than aiming at a borrowed industry figure.

Not usually on its own, because the settlement rebuild has to happen before the ledger can carry the detail needed. The practical approach is to reconstruct the settlement data in a controlled worksheet, post it as a structured journal with channel tagging, and report from the ledger once the detail is in place.