Glossary - Products and inventory
ABC analysis
ABC analysis is an inventory classification method that ranks products by their share of a value measure, usually revenue or contribution margin, and splits the range into three classes: A, B and C, from the vital few to the long tail. Ecommerce operators use it to focus buying, stock control and reporting effort where the value sits.
| Class | Common convention (cumulative value share) |
|---|---|
| A | The SKUs that together produce roughly the first 80% of value, usually a small minority of the range |
| B | The SKUs producing roughly the next 15% of value |
| C | The SKUs producing roughly the last 5% of value, usually the majority of the range |
The 80/15/5 cuts are a convention borrowed from the Pareto principle, not a law. Pick cuts that produce classes you will actually treat differently.
Worked example
Sixty SKUs, a clear majority of the range, produce $25,000 between them. Those 60 SKUs still need photography, stock counts, storage and reorder decisions. That imbalance is the entire point of the exercise.
What is a good ABC analysis?
There is no benchmark shape, because the split reflects how concentrated your catalog already is, and that varies with category, range strategy and margin structure. A good ABC analysis is judged by its inputs instead: it ranks on the value measure you actually optimise (contribution margin is the better default, since revenue flatters discounted, low-margin heroes), it runs at the grain where decisions happen (variant, not just product), and it is re-run often enough that classes track seasonality and lifecycle. How concentrated your range is, on its own, is a separate question - see value concentration.
ABC analysis vs related metrics
| Metric | What it tells you | How it differs |
|---|---|---|
| GMROI | Gross margin earned per dollar of inventory investment | A per-SKU return score; ABC is a ranking of the whole range |
| Inventory turnover | How many times stock sells through in a period | Speed of stock, blind to how value concentrates across SKUs |
| Sell-through rate | Share of received units sold in a window | Grades a buy or a season; ABC grades the standing range |
| Value concentration | How much of total value the top slice holds | The concentration curve itself; ABC cuts that curve into working classes |
Common mistakes
- Ranking on revenue alone. A heavily discounted, low-margin hero can outrank the products that actually fund the business. Run the ranking on contribution margin as well and investigate every SKU where the two disagree.
- Set and forget. Classes drift with seasonality and product lifecycle. An analysis from last year quietly misclassifies this year's range.
- One dimension only. ABC measures value, not predictability. Pair it with an XYZ axis for demand variability before setting stock policies, or a stable A item and a spiky A item get the same treatment.
- Treating C as a delete list. Some C items win first orders or attach to A items in baskets. Check what a C SKU does for acquisition and repeat behaviour before culling it.
- Classifying at the wrong grain. Product-level classes hide variant-level dead weight: one size or colour can be class A while its siblings sit in C.
FAQ
A is the small group of SKUs producing most of the value, B is the middle band, and C is the long tail producing the least. The cuts are conventions you choose, commonly around 80/15/5 of cumulative value.
Contribution margin is the better default, because revenue ranks a discounted, low-margin hero above the products that actually fund the business. Running both and comparing the two rankings is even more useful: the disagreements are the finding.
Often enough to catch seasonality and lifecycle shifts: quarterly is a common cadence, monthly for fast-moving ranges. The honest test is whether a re-run would change any buying or stocking decision; if it would, it is overdue.
Updated July 2026