ABC analysis: ranking SKUs so you watch the ones that matter
ABC analysis is the practice of ranking every SKU you stock by its contribution to the business, then grouping them into bands, usually three, so the few lines that carry your revenue get checked more often than the long tail. It is a priority tool for inventory, not a maths formula, and on a 50 to 2,000 SKU store it is the difference between watching everything badly and watching the top of the l
What ABC analysis means
ABC analysis is a way of sorting your inventory into a small number of bands, usually three, by how much each SKU contributes to the business, so most of your attention goes to the few lines that carry the sales.
The thinking behind it is simple and old. In most catalogues, contribution is lopsided. A handful of products do the heavy lifting, a middle group is steady, and a long tail sells slowly. If you review every SKU with the same care, you spend the same effort on a line that sells two units a month as on the one that pays the rent. Ranking fixes the effort, not the stock.
Choosing what you rank by
The band is only as useful as the measure under it. Three common choices, and what each one does to your list:
- Revenue per SKU. The default. It puts your best sellers at the top and is easy to pull from Shopify reports.
- Gross margin per SKU. Slower to assemble because you need landed cost, but it stops a high-revenue, thin-margin line from crowding out a quieter one that actually makes money.
- Units sold. Useful when your prices are close together, and useful for warehouse work, because units drive picking and shelf space rather than value.
Pick one, write down which one you used, and keep it consistent. Mixing measures between reviews makes the bands move for no real reason.
A worked example, numbers invented to show the shape
| Band | SKUs in this example | Share of revenue | What you do with it |
|---|---|---|---|
| A | 32 of 400 | Most of the revenue | Check cover weekly, confirm lead times with the supplier, never let these go to zero |
| B | 88 of 400 | A meaningful middle | Check cover every week or two, reorder on a rule rather than on instinct |
| C | 280 of 400 | A thin tail | Review the band monthly, batch orders with suppliers, consider dropping the deadest lines |
Where ABC analysis stops being enough
ABC tells you which SKUs deserve attention. It does not tell you when to place an order. Those are two different questions, and the second one is the one that costs you money when you get it wrong.
Two failure modes are worth naming.
A C-item with a long lead time can still stop a sale. If a slow seller takes ninety days to arrive and it is a component of a bundle, its band says ignore it and its lead time says watch it closely. Banding by contribution alone hides supplier risk.
Bands go stale. A product that was a C last quarter can be an A after one good week of traffic. Any ranking you refresh twice a year is a photograph of a catalogue that has already moved on.
The practical answer is to use ABC for the frequency of your review and a reorder point for the trigger. The band decides how often you look. The reorder point, built from each SKU's sales velocity and your supplier's lead time, decides when you order. A ranking does not fire; a threshold does.
How to run one without a spreadsheet weekend
Export your sales by variant for a window that reflects normal trading, ninety days is a reasonable start for most stores. Sum revenue per SKU, sort descending, then look for a visible drop-off and cut the list there rather than forcing a tidy percentage.
Then do the only step that matters: write the review cadence next to each band and keep it. A ranking that does not change your calendar has changed nothing. Reread the bands once a quarter, and again after any launch or season that reshuffles demand.
Related terms
ABC analysis sits next to the numbers that do the actual triggering. Each of these has its own page in this glossary.
- Reorder point. The stock level at which you place the order. The formula, worked through on a real SKU.
- Sales velocity. Units per day, and why the length of the window you average over changes the answer.
- Lead time. The supplier wait, the one number no app can read out of Shopify for you.
- Safety stock. The buffer you hold against demand and supply varying, and how merchants size it.
- Days of inventory on hand. Current stock divided by daily sales, the cover figure behind most alerting.
- Stockout. What running out actually costs a small store.
- Inventory turnover and sell-through rate. Two ratios that measure how hard your stock is working, and are often used to build the A, B and C ranking in the first place.
If you want the sequence rather than the vocabulary, the guide on when to reorder turns velocity and lead time into an order date.
Questions
What do A, B and C actually stand for?
Nothing beyond their order. A is the band of SKUs that contribute most, B is the middle, C is the long tail. The letters are labels for priority, not grades for the products.
Do I have to use three bands?
No. Three is conventional because it is easy to act on. Some merchants add a D band for discontinued or dead lines so they are visible for clearance rather than sitting inside C.
Should I rank by revenue or by margin?
Margin is the better measure if you have reliable landed costs, because it reflects what each SKU keeps rather than what it brings in. Revenue is the pragmatic start, since Shopify reports give it to you directly. Choose one and stay consistent between reviews.
How often should I redo the ranking?
Quarterly suits most small stores, plus an extra pass after a launch, a seasonal peak or a price change. Any slower and your bands describe a catalogue you no longer sell.
Does ABC analysis replace a reorder point?
No. ABC decides how much attention a SKU gets; the reorder point decides when to place the order. A C-item with a long supplier lead time still needs a threshold, or it quietly runs out between monthly reviews.
Is ABC analysis worth it on 200 SKUs?
It is worth it the moment you cannot scan your whole catalogue in one sitting. At 200 SKUs the value is mostly in deciding which twenty you check every week. Below that, a reorder point per SKU and an alert may be all the prioritisation you need.
Can Shopify do ABC analysis for me?
Shopify's sales reports give you the raw ranking input, sales by product and variant over a window, which you can sort and band in a spreadsheet. The banding itself is a judgement you make, not a report you run.