Cost per SKU Calculator

What a SKU costs beyond its stock

Every SKU carries costs that have nothing to do with how many units sell: photography, listing creation, description writing, a warehouse location, a line in every stock count, and a share of the systems that track it.

These are largely fixed per SKU. A product selling 2,000 units a year and one selling 20 carry roughly the same setup and maintenance burden, and the second spreads it across a hundredth of the volume.

Which is why range breadth is expensive in a way that does not appear in any margin calculation. A catalogue of 400 SKUs where 300 sell rarely is carrying 300 sets of these costs against very little revenue.

The setup costs

Photography is the largest for most sellers: a proper set of images for one product runs from perhaps £40 doing it yourself to several hundred with a studio.

Then copy, specification, category assignment, and listing on each channel. Multi-channel sellers pay this repeatedly, since each marketplace wants its own format.

Then compliance where applicable: testing, documentation, registration. For regulated categories this can dwarf everything else and it is per product rather than per unit, which makes low-volume SKUs in regulated categories particularly hard to justify.

The costs that recur

A warehouse location, occupied whether or not the SKU moves. A line in every cycle count. Space in the pick face, which for a slow mover is prime real estate holding nothing useful.

Systems cost, since many platforms and tools charge by SKU count or by variant count, and a catalogue that doubles can move a business into a higher tier.

Then attention: every SKU needs its stock level watched, its price reviewed, its reviews monitored and its content updated. That is management time, it scales with range breadth, and it is the constraint that actually binds in most small businesses.

Finding the tail

Ranking SKUs by contribution after allocating these costs produces a list, and the bottom of it is usually longer than expected.

The typical finding is that a substantial minority of SKUs contribute nothing or contribute negatively once their share of fixed costs is charged. In a 300 SKU catalogue that might be 80 or 100 products.

Before cutting, check for halo effects. Some slow SKUs bring customers who buy other things, complete a range that customers expect, or exist for a strategic reason. Those are legitimate exceptions and they should be identified deliberately rather than assumed for everything.

Rationalising a range

Cutting SKUs releases cash from stock, space in the warehouse, and attention. Those releases are the actual benefit and they are worth quantifying before the decision.

The process that works is staged: stop reordering the bottom tier, sell through what remains, and remove the listing when it is gone. That avoids write-offs and the cash comes back.

The discipline that prevents it recurring is a threshold for new SKUs: a minimum expected annual contribution that must exceed the setup and carrying cost by a stated margin. Most catalogues bloat because there is no such threshold and every new product seems worth trying.

Variants deserve the same scrutiny and rarely get it, because adding a colour feels smaller than adding a product. It is not: each variant needs its own photography, its own location, its own stock forecast and its own line in every count, and a range offered in seven colours where two sell is carrying five sets of those costs indefinitely. Launching with three variants and adding a fourth once the first three prove out is slower and considerably cheaper than the reverse.