Combined XYZ+ABC analysis
XYZ+ABC joins two questions in one code. The first letter is the ABC class of the total across all selected periods: how much the item brings in. The second is the XYZ class from the coefficient of variation: how evenly it sells. Nine cells, AX through CZ; AX is the backbone of the assortment, CZ the tail that rarely deserves attention.
An XLSX file: a row per item, columns as periods in chronological order, plus a column with the name. Two periods minimum, a year by month in practice. The total across periods is computed for you — no separate total column is needed. Both pairs of thresholds are adjustable: 80 / 95 for ABC and 10 % / 25 % for XYZ.
Important: merge duplicate rows before uploading. The service does not detect or combine duplicates — an item that appears in the file twice is counted twice, and the shares and classes come out wrong.
The table gains the total across periods, the coefficient of variation, the ABC and XYZ classes, and the combined cell code — AX, BY, CZ and so on. The matrix shows how many items and how much money fall into each of the nine cells, and the recommendations take them one by one: what to protect, what to move to order-only, what to retire. The report downloads as XLSX.
This is the fullest of the four: ABC and XYZ run on the same data and land in one code. With no series of periods, only a single snapshot, classic ABC analysis is enough. With two metrics at one date — complex ABC. If only the evenness of demand matters — XYZ.
cell = ABC class + XYZ class
Two computations run over one file: ABC on the revenue column, XYZ on the period columns. Each item takes a letter from each, and together they are its cell.
Nine cells are nine different decisions. AX items stay on the shelf permanently, AZ items make money and are unpredictable at the same time, CZ items usually do not earn their warehouse space.
| Item | Revenue | Share | ABC class | CV | XYZ class | Cell |
|---|---|---|---|---|---|---|
| SKU-101 | 404 | 8.8% | A | 2.2% | X | AX |
| SKU-102 | 800 | 17.4% | A | 4.0% | X | AX |
| SKU-103 | 1,220 | 26.6% | A | 3.7% | X | AX |
| SKU-104 | 500 | 10.9% | A | 13.6% | Y | AY |
| SKU-105 | 335 | 7.3% | A | 11.5% | Y | AY |
| SKU-106 | 250 | 5.4% | B | 18.3% | Y | BY |
| SKU-107 | 200 | 4.4% | B | 36.7% | Z | BZ |
| SKU-108 | 100 | 2.2% | C | 78.7% | Z | CZ |
| SKU-109 | 60 | 1.3% | C | 102.7% | Z | CZ |
| SKU-110 | 720 | 15.7% | A | 30.4% | Z | AZ |
Revenue gives the ABC letter, the swing in quarterly sales gives the XYZ letter; together they are the cell.
Two analyses run over one file: ABC on the revenue column and XYZ on the period columns. The first letter of the cell is the ABC class, the second the XYZ class.
The letter order is the industry one: importance first, stability second. The codes are never reordered and never translated.
AX is the backbone of the assortment: predictable and important, so the safety stock for it can be kept thin. AZ earns money and jumps around at the same time — it is held with a buffer and reviewed by hand.
CZ is small and unpredictable: usually a candidate for delisting or for order-on-demand. The report breaks every one of the nine cells down and gives recommendations for each.
One row per item. You need a column with a name or SKU, a numeric revenue column and several columns of sales by consecutive periods.
Merge duplicates before uploading: the service does not look for them and does not combine them, and an item that appears twice spoils both the shares and the classes.
