ABC analysis ranks items by how much value they contribute (A = the top ~70–80%, C = the long tail); XYZ analysis ranks them by how predictable demand is (X = stable, Z = erratic). Combined into a 9-box grid, they tell you which items deserve tight stock control and which can run on a simple rule.
Not every item in your inventory deserves the same attention. A shop with 500 SKUs usually earns most of its revenue from a few dozen of them — and those are the ones worth counting often, forecasting carefully, and never running out of. ABC/XYZ analysis is a simple way to find them.
ABC: ranked by value
ABC analysis groups items by how much they contribute to your total sales value (or consumption value) over a period, usually the last 12 months. It typically follows something close to the 80/20 rule.
- A items: a small share of SKUs (often 10–20%) that make up roughly 70–80% of value
- B items: a middle group contributing around 15–25% of value
- C items: the long tail — often half or more of SKUs — contributing around 5% of value
How to calculate it
- Export 12 months of sales per item: units sold × unit price = annual sales value.
- Sort items from highest to lowest annual sales value.
- Add a running cumulative percentage of total value.
- Mark items up to ~80% cumulative as A, the next ~15% as B, and the rest as C.
| Item | Annual sales value | Share | Cumulative | Class |
|---|---|---|---|---|
| Basmati rice 5 kg | ₹4,20,000 | 42% | 42% | A |
| Sunflower oil 1 L | ₹2,60,000 | 26% | 68% | A |
| Tea 500 g | ₹1,10,000 | 11% | 79% | A |
| Biscuits (assorted) | ₹80,000 | 8% | 87% | B |
| Detergent 1 kg | ₹60,000 | 6% | 93% | B |
| 40 other items | ₹70,000 | 7% | 100% | C |
XYZ: ranked by predictability
XYZ analysis is a second, independent cut — by how steady each item's demand is from period to period. A common way to measure it is the coefficient of variation (CV): the standard deviation of weekly or monthly sales divided by the average. Typical cut-offs:
- X items: CV below about 0.5 — steady, predictable demand
- Y items: CV between about 0.5 and 1.0 — some variability, seasonal or trend-driven
- Z items: CV above about 1.0 — erratic, hard-to-predict demand
Combining the two: the 9-box matrix
| X (stable) | Y (variable) | Z (erratic) | |
|---|---|---|---|
| A (high value) | Tight reorder points, frequent review, never stock out | Forecast with seasonality, moderate buffer | Order close to demand, review weekly |
| B (medium value) | Automatic reorder rule | Periodic review, moderate buffer | Order on demand where possible |
| C (low value) | Simple "reorder when low" rule, larger order quantities | Simple rule, review quarterly | Consider stocking less or dropping |
An AX item — high value, predictable — deserves a carefully set reorder point, since running out is costly and demand is easy to forecast. A CZ item — low value, erratic — usually isn't worth the effort; it's also a common source of dead stock.
What to actually do with it
Once items are classified, the output is different stock policies per group: tighter reorder points and more frequent counts for AX/AY items, and a simpler rule for C items. It also tells you where to start a stock audit — A items first. Doing this on paper for a large catalogue is tedious; it's really a by-product of having clean sales and stock data in one place.
StashBill keeps the per-item sales and stock history this analysis needs, and its inventory intelligence uses it to flag reorder points and slow movers.
Frequently asked
How often should I redo ABC/XYZ analysis?
Quarterly is a reasonable default for most small businesses — more often if your catalogue or demand patterns change quickly.
Should I use revenue or profit for ABC?
Revenue (sales value) is the standard and easiest to calculate. If margins vary a lot between items, running it on gross profit as well can reveal items that sell a lot but earn little.
Do I need special software to do this?
No, it can be done in a spreadsheet from sales data, but it's a manual, recurring exercise unless your inventory system tracks the underlying sales history for you.