ABC analysis of the range
Splits styles or groups into A, B and C by contribution to revenue or margin: what feeds the business, the middle, and what can be cut.
At a glance
- What question does it answer?
- Which items give most revenue or margin and which can be cut.
- What data is needed?
- A list of items or groups and revenue (or margin) for each over one period.
- How to read the result?
- Class A is the core and needs full depth and availability control. Class C are candidates for cutting, merging or a small test run. Look at how much stock sits in class C. Example on the default values: 10 positions worth RUB 63m: class A (up to 80% of the cumulative share) has 5 positions, class B (up to 95%) 2, class C 3. Five positions give 84% of the measure and class C only 5%. If class C holds a noticeable share of stock with a small share of revenue, cut its variants and buying. Next: “Days of inventory and excess capital” by class and “Width and depth: budget, models and pieces”.
Why and how to reason
ABC analysis starts from the observation that the contribution of items is uneven: a small part gives a large part of revenue. To see it, items are sorted from large to small and the cumulative share is counted. While the cumulative share before a row is below the A boundary (80% by default), the item is class A; up to 95%, B; the rest, C.
Why these boundaries? It is a convention, not a law of nature. It is needed to split attention: class A needs personal control of availability and sizes, B ordinary control, C a decision to keep, merge or drop.
Example. Total RUB 63 m. Jackets 18.5 (29%), with coats 30.8 (49%), with jeans 40.6 (64%), with dresses 48.0 (76%), with shirts 53.1 (84%). Shirts start at 76%, below 80, so they are still A. So A is 5 items of 10 but 84% of revenue; B is knitwear and skirts; C is shorts, belts and scarves.
When to use it
Use it to decide where to hold deep stock and control, and where to cut. It works on styles, categories, suppliers, customers. It does not answer “why”: a new item with small revenue can be a success, and an old A item may simply be habitual. Assumption: the analysis period is typical. For seasonal groups analyze within the season, otherwise winter goods fall into C in summer.
How to read the result
Look at how many items are in class A and what share of revenue they give. If A is 5 of 10 items, concentration is moderate. If 5 of 100, the business depends on a narrow core and a shortage of those styles hurts most. Class C is not always a candidate for removal: check the margin and the role in a look. Decision: for A hold safety stock and a full size run, for B standard rules, for C cut depth or merge. Then cross-check ABC by margin, not revenue alone.
Common mistakes
Typical trap: analyzing revenue when profit matters; mixing categories of very different scale in one table; treating C as unnecessary. Another trap is too few rows. With 10 items the 80% boundary gives a coarse split, so add detail. Another trap: assuming class A stays class A forever, although the composition changes every season.
What data is needed?
- A list of items or groups and revenue (or margin) for each over one period.
- Class thresholds: 80% and 95% by default.
- Where to take it: revenue or margin by SKU and category come from the sales report in 1C or the ERP (or the tills) for a season or a year; margin from cost accounting. Export a “position — amount” table, paste the rows and check that the column holds one metric.
Inputs
| Field | Unit | Value in the example | Meaning |
|---|---|---|---|
| Class A: up to cumulative share | % | 80 | Styles that together give this share of the metric. |
| Class B: up to cumulative share | % | 95 | The cumulative share up to which rows are class B; the rest is class C. It must be above the class A threshold. |
How it is calculated
| Step | Formula |
|---|---|
| Share | Value ÷ total |
| Class | A while the cumulative share before the row is below the A threshold; then B and C |
Worked numbers
Default values from the app; in the interactive version you enter your own.
- Class A
- 5 of 10 · 84 % of the metricThe core: these styles need full depth and availability control.
- Class B
- 2 · 11 %
- Class C
- 3 · 5 %Candidates for cutting, merging or a small test run.
- Total
- 63 m ₽
Data as a table
| Measure | Value |
|---|---|
| Jackets · A | 18.5m |
| Coats · A | 12.3m |
| Jeans · A | 9.8m |
| Dresses · A | 7.4m |
| Shirts · A | 5.1m |
| Knitwear · B | 4.2m |
| Skirts · B | 2.6m |
| Shorts · C | 1.5m |
| Belts · C | 900k |
| Scarves · C | 700k |
Data as a table
| Items by descending contribution | Cumulative share |
|---|---|
| Jackets | 29.4 |
| Coats | 48.9 |
| Jeans | 64.4 |
| Dresses | 76.2 |
| Shirts | 84.3 |
| Knitwear | 91 |
| Skirts | 95.1 |
| Shorts | 97.5 |
| Belts | 98.9 |
| Scarves | 100 |
Classes in descending order
| Name | Value | Share | Cumulative | Class |
|---|---|---|---|---|
| Jackets | 18.5 m ₽ | 29.4 % | 29.4 % | A |
| Coats | 12.3 m ₽ | 19.5 % | 48.9 % | A |
| Jeans | 9.8 m ₽ | 15.6 % | 64.4 % | A |
| Dresses | 7.4 m ₽ | 11.7 % | 76.2 % | A |
| Shirts | 5.1 m ₽ | 8.1 % | 84.3 % | A |
| Knitwear | 4.2 m ₽ | 6.7 % | 91 % | B |
| Skirts | 2.6 m ₽ | 4.1 % | 95.1 % | B |
| Shorts | 1.5 m ₽ | 2.4 % | 97.5 % | C |
| Belts | 900,000 ₽ | 1.4 % | 98.9 % | C |
| Scarves | 700,000 ₽ | 1.1 % | 100 % | C |
5 of 10 items give 84 % of the metric. Class C accounts for 5 %: check how much stock and how many variants sit in it.
Nuances and limits
- Use one metric in all rows: mixing revenue and margin breaks the classes.
- A new style with low revenue may be succeeding: ABC sees contribution, not cause.
- One set of cut-offs for all categories: categories with different turnover give incomparable classes, calculate within a category.