Products & InventoryGlossary
What Is ABC Analysis? How to Use It in Inventory
Last checked against regulations: October 28, 2026
In short
ABC analysis sorts the products in your inventory into classes A (highest), B (medium) and C (lowest) based on their contribution to revenue. Count frequency, order tracking and safety stock are set according to this class.
Contents
ABC analysis is an inventory management method that sorts the products in your stock into classes A (highest), B (medium) and C (lowest) based on their contribution to total revenue. Its goal is to direct your limited time and capital toward the products that earn you the most money.
The Pareto principle behind ABC analysis
The method is built on the Pareto principle. According to this principle, most results come from a small number of causes. In inventory terms, a small share of your products brings in most of your revenue, while most of your products contribute little to total sales.
This distribution varies from business to business. For example, roughly one-fifth of products might bring in three-quarters of revenue; in your data, the ratio may turn out differently. ABC analysis makes this concentration visible, and you then give each class its own level of tracking.
For a marketplace seller, this has a practical meaning. The product that runs out of stock during a campaign is usually in class A. The products sitting in the warehouse and tying up capital are usually in class C.
How do you perform an ABC analysis?
You perform the analysis separately for each stock item. That is why your products need a unique stock code. If you don't have a consistent coding system yet, first read What is a SKU and how do you create a stock code.
- Pull sales data by SKU for a specific period, for example the last 12 months. Take data from all marketplaces together.
- Calculate each SKU's net revenue for this period: deduct returns and take amounts excluding VAT (KDV).
- Sort products by revenue from highest to lowest.
- Find each product's share of total revenue.
- Calculate the cumulative share by adding up the shares from top to bottom.
- Mark products as A, B and C according to the thresholds you set.
You find each product's share with the formula below:
Revenue share (%)=Product revenue / Total revenue × 100
- Product revenue
- the net sales amount of that SKU in the selected period, after deducting returns and excluding VAT (on the same basis across all channels)
- Total revenue
- the sum of the net sales amounts of all SKUs in the same period
- Cumulative share
- the sum of the shares up to that product in the ranking
Net revenue prevents two mistakes. A product with a high return rate can appear in class A based on gross revenue, even though the sales you actually keep are much lower. If one channel's report comes in including VAT and another's excluding VAT, products will also be ranked incorrectly. To convert a VAT-inclusive amount into a VAT-exclusive one, you can use the VAT calculation article.
Sample table
For example, suppose a store selling phone accessories has 10 products and annual net revenue of 1,000,000 TL. In this example, the thresholds were chosen as follows: A up to a cumulative share of about 75%, B up to 93%, and the remaining products C.
| Product (SKU) | Annual net revenue (TL) | Revenue share (%) | Cumulative (%) | Class |
|---|---|---|---|---|
| Wireless earbuds | 450,000 | 45 | 45 | A |
| Smartwatch | 300,000 | 30 | 75 | A |
| Power bank | 80,000 | 8 | 83 | B |
| Phone case | 60,000 | 6 | 89 | B |
| Charging cable | 40,000 | 4 | 93 | B |
| Screen protector | 25,000 | 2.5 | 95.5 | C |
| Car mount | 15,000 | 1.5 | 97 | C |
| Cable organizer | 12,000 | 1.2 | 98.2 | C |
| Cleaning kit | 10,000 | 1 | 99.2 | C |
| Earbud case | 8,000 | 0.8 | 100 | C |
When you total the same example by class, the Pareto distribution becomes clear:
| Class | Number of products | Share of products (%) | Revenue share (%) |
|---|---|---|---|
| A | 2 | 20 | 75 |
| B | 3 | 30 | 18 |
| C | 5 | 50 | 7 |
In this example, 20% of products bring in 75% of revenue. Class C, which makes up half of the products, accounts for only 7% in total. The two items that really need attention are the earbuds and the smartwatch.
Stock policy and count frequency by class
The value of classification shows when you apply a different rule to each class. The frequencies in the comparison below are examples; adapt them to your own volume.
| Criterion | Class A product | Class C product |
|---|---|---|
| Count frequency | frequent, e.g., once a month | infrequent, e.g., once or twice a year |
| Stock tracking | daily, per channel | periodic batch check |
| Safety stock | always defined | kept low or not held at all |
| Ordering | in small batches, frequently | in bulk, infrequently |
| Supplier | a backup supplier is recommended | a single supplier is usually enough |
| Campaigns | stock availability is checked in advance | can be used to clear out stock |
B items fall between these two extremes. For example, a count every three months and a standard reorder point are enough for most sellers.
For A products, running out of stock means direct revenue loss and can also affect listing performance. Pin down the reorder point for these products; the article How to set a critical stock level walks you through this calculation step by step.
If you sell on more than one marketplace, the risk of overselling shows up most with A products. That is why syncing stock across channels in real time matters; details are in the article Inventory management in multichannel selling.
Common mistakes
- Analyzing at the model level: The M size of a T-shirt model can fall into class A while the XXL size falls into class C. Run the analysis separately for each variant; for the method, see Stock tracking for products with variants.
- Evaluating channels separately: A product may look weak on a single marketplace but be in class A overall. Look at the total across all channels for stock decisions.
- Classifying by gross revenue: An analysis done without deducting returns makes products that sell a lot and get returned a lot look stronger than they are. Base it on net revenue.
- Not filtering out campaign periods: A single big discount week can temporarily move a C product into class A. Use data from a longer period instead of a short one.
- Looking only at revenue: Products with high revenue but low profit after commission and shipping are misleading. Repeat the analysis using profit contribution as well; for the calculation, use the marketplace profit calculation article.
- Putting a new product straight into class C: A product with a short sales history shows low revenue. Track these products in a separate group for a while.
1/2How should each SKU's revenue be calculated in ABC analysis?
Show answer
Correct answer: B) Net amount after deducting returns, excluding VAT
The article says to calculate each SKU's net revenue by deducting returns and taking amounts excluding VAT.
2/2At what level should ABC analysis be done for products with variants?
Show answer
Correct answer: C) Separately for each size and color, at SKU level
The article notes the M size of a model can be class A while the XXL size is class C, so the analysis should be run separately for each variant.
Frequently asked questions
What percentages should the A, B and C thresholds be in ABC analysis?
There is no universal standard. You set the thresholds yourself based on your own product range. What matters is keeping the thresholds you choose the same every period; that way, you interpret products that change class correctly.
How often should an ABC analysis be updated?
For products whose sales change with the season, classes become outdated quickly. Refreshing the analysis periodically, for example every quarter using the last 12 months of data, is a balanced approach.
Does ABC analysis replace the year-end inventory count?
No. Cycle counts based on ABC are an operational control. Statutory inventory and record-keeping obligations are fulfilled separately; confirm the details with your certified public accountant (mali müşavir).
How do you perform ABC analysis for products with variants?
Run the analysis not at the model level but at the level of a separate SKU for each size and color. One size of the same model can fall into class A while another size falls into class C.
#abc analysis#inventory management#pareto#inventory count#marketplace seller
Kafatura Editorial Team
We answer the everyday questions of marketplace sellers using a verified regulatory knowledge base. If you spot an error or something out of date, please let us know.
This article is translated from the Turkish original; regulatory references apply to Türkiye. Read the Turkish original →
