Product-Level Contribution Margin

Gross margin ranks a catalogue by how well each product was bought. Contribution margin ranks it by how well each product can be sold. Those are different orderings, and the second one is the one merchandising decisions should follow.

What separates the two rankings

Gross margin knows one thing about a product: the gap between what it cost and what it sold for.

Contribution margin knows what it costs to actually deliver it — and that depends on properties gross margin cannot see:

Weight, which drives base freight. Volume, which drives dimensional weight charges and often exceeds actual weight as the binding constraint. Fragility, which drives protective packaging and breakage replacement. Whether it ships alone, which decides if it carries a whole parcel or a share of one. Return rate, which decides how often the margin gets reversed.

In the worked example, two products at identical price and identical gross margin sit twenty-nine points apart on contribution margin. Nothing about how they were purchased explains it. Everything about how they ship does.

Allocating shared costs

Most of the costs below gross profit are incurred per order, not per unit, so getting them onto products requires an allocation. Three bases, each right for something different:

By weight or volume share — correct for shipping and packaging. A unit that is sixty percent of a parcel’s billable weight carries sixty percent of the label.

By revenue share — correct for payment fees, which genuinely scale with order value.

Evenly per unit — almost always wrong for anything, and specifically wrong in a direction that flatters heavy products.

The choice matters less than the consistency. These figures are used comparatively, so a stated method applied uniformly produces a valid ranking even if any single number carries an error bar.

Do it at variant level

Parent products hide the interesting variation.

Sizes differ in cost to make and weight to ship. Colours can differ in supplier price and in return rate. A parent product showing a comfortable blended margin frequently contains one variant losing money on every unit and another carrying the line.

Apparel is the clearest case: the largest sizes often cost more, weigh more and return more, and all three effects vanish when averaged into a single product row.

Two numbers, not one

Contribution margin per unit tells you whether an item earns its place. It does not tell you how much of the business depends on it.

Multiply by units sold and you get total margin contribution — this product’s share of the store’s contribution margin. The pair drive different decisions:

High rate, low total. A good product few people know about. Promotion and merchandising problem.

Low rate, high total. The product carrying your revenue at a rate you cannot really afford. The most dangerous position on any catalogue, because volume disguises it and any correction touches your biggest line.

High on both. Protect the stock position; running out is the expensive failure.

Low on both. Discontinue, unless basket effects say otherwise.

Basket effects, and the cut you should not make

The single thing per-product analysis cannot see is what an item does for the orders it appears in.

A thin-margin accessory appearing in a third of orders may be lifting average order value across the store. Removing it improves the product table and reduces total margin — a decision that looks right in the spreadsheet that prompted it and is wrong in the accounts a quarter later.

Two checks before cutting anything. What proportion of its units ship alone? And what is the average order containing it worth against the average order without it? If it rarely ships alone and the orders containing it are larger, the product is doing a job the per-unit number was never going to show.

Using it to price

The most direct application of a product-level figure is pricing, and it is a more precise instrument than a percentage uplift across the catalogue.

A blanket five percent price rise takes the same amount from every product regardless of whether it needed it. Product-level contribution margin says which items are actually below the rate the business needs, and by how much — so the increase can land where the shortfall is and leave your competitively priced lines alone.

The arithmetic is worth stating plainly, because it argues against across-the-board discounting more strongly than most people expect. On a product at 41% contribution margin, a five percent price rise adds roughly twelve percent to what that unit contributes. On a product at 12%, the same five percent adds over forty percent to its contribution. The thinnest-margin items are where price changes do the most work, and they are usually the last place anyone looks.

The same logic runs in reverse for discounts, which is why a sitewide percentage is nearly always the wrong instrument.

What usually breaks the calculation

Uncosted variants, counted as pure margin and appearing implausibly near the top of the ranking.

Bundles that do not explode into components, reporting at one hundred percent margin while being the most promoted items you sell.

List price used instead of realised price, which overstates margin most on the products that are discounted hardest.

Missing landed costs — freight and duty on imported goods, recorded nowhere and understating cost unevenly across the catalogue.

Returns left out, which matters most in apparel and moves the ranking most on the products that look best before the adjustment.

Check those five before acting on any ranking. Each of them shifts products in the same direction — upward, wrongly — and a catalogue decision made on an unchecked table usually removes the wrong item.