ABC Analysis for Inventory: How to Do It

ABC analysis inventory classification is a way of sorting items by importance so that management attention, forecasting effort, safety stock and service targets go where they matter most. It rests on a pattern found in almost every product range: a small share of items generates most of the revenue or margin, and a long tail of items contributes very little. Treating all items the same wastes effort on the tail and under-serves the head. This guide shows how to do an ABC analysis step by step, works through a ten-item example with checked cumulative percentages, adds an XYZ dimension for variability, shows how to turn the classes into policies for service, forecasting and replenishment, and explains how to keep the classification current. It is part of demand planning and inventory optimization, follows economic order quantity and reorder point formula, and leads to inventory carrying cost and excess and obsolete inventory. Segmenting inventory is a core tool for freeing cash, as described in cash flow management.

The basic idea

ABC analysis applies the Pareto principle to inventory. Items are ranked by an annual value measure, and classed as A (the few that account for most of the value), B (a middle group) and C (the many that account for little). The usual breakpoints are roughly the top 70 to 80 percent of value for A, the next 15 to 25 percent for B and the remainder for C, but the exact cutoffs should reflect your own range and how many items you can manage closely. The point is to differentiate, not to find a perfect partition.

Step 1: Choose the value measure

Common measures include annual revenue, annual cost of goods sold (usage value), gross margin or the dollar value of average inventory. Revenue is simple and favors high-selling items. Usage value (annual units times unit cost) reflects the investment needed to support the item. Margin favors profitable items. For inventory decisions, usage value or margin is often more meaningful than revenue. Whichever you choose, apply it consistently, and consider a second classification for strategic items that matter for reasons other than value, such as a hero product or an item required by a key customer.

Step 2: Rank and accumulate

List all active items with annual value, sort descending, and compute each item's share of the total and the running cumulative share. Define cutoffs and assign classes.

A worked example

Ten items, with annual revenue in thousands of dollars (illustrative):

Check: the values add to 400 + 250 + 120 + 80 + 50 + 40 + 25 + 15 + 12 + 8 = 1,000. The three A items are 30 percent of the items and account for 77 percent of the value. The three B items are another 30 percent of the items and 17 percent of the value (80 + 50 + 40 = 170). The four C items are 40 percent of the items and 6 percent of the value (25 + 15 + 12 + 8 = 60). In a real range of a thousand items, the proportions are often even more skewed, with the bottom half of items contributing a few percent of value.

Step 3: Add variability (XYZ)

Value alone ignores how predictable demand is. Add a second dimension based on the coefficient of variation of demand (standard deviation divided by mean) over a recent period. A common split is X for low variability (for example below 0.5), Y for medium (0.5 to 1.0) and Z for high (above 1.0). The combination yields nine segments. For example, an item with monthly demand of 200 units and a standard deviation of 60 has a coefficient of variation of 60 / 200 = 0.30, an X item. An item with a mean of 50 and a standard deviation of 70 has 70 / 50 = 1.4, a Z item. AX items are valuable and predictable and should run on automation with tight control. AZ items are valuable and volatile and need the most planner attention, driver-based adjustments and careful safety stock. CZ items are low value and erratic and may be candidates for delisting or make-to-order.

Step 4: Turn classes into policies

The specific targets belong to your business, and the table shows the logic: the effort and the service level rise with value and fall with predictability only where the cost of a miss is lower. See safety stock formula for translating a service target into stock.

Step 5: Use it to find opportunities

ABC analysis often surfaces surprises. Look for C items that hold a large share of inventory dollars. If C items are 6 percent of sales but 25 percent of stock, there is cash to release. Look for A items with frequent stockouts, where better forecasting and safety stock pay off most. Look for C items with high carrying costs and low margins that could be dropped. Compare the share of inventory by class with the share of sales. A healthy range holds inventory roughly in proportion to sales importance, with A items turning fastest. See inventory turnover.

Step 6: Keep it current

Product ranges change. Reclassify at least quarterly, and whenever there is a major launch or a change in sales mix. Avoid reclassifying so often that policies flip back and forth; add a rule that an item must be outside its band for two cycles before moving. Treat new items separately for their first months, since they have no history, and reclassify once they have enough data. Flag items that move from A to C, since they may be in decline and need a run-down plan.

In Excel

Sort the data by value descending. In one column compute each item's share (=value/SUM(values)). In the next, compute the cumulative share (=previous + share). Assign classes with a nested IF: =IF(cum<=0.8,"A",IF(cum<=0.95,"B","C")). For XYZ, compute the coefficient of variation from monthly history with =STDEV.S(range)/AVERAGE(range) and classify with another IF.

Common mistakes

1. Classifying by revenue only, ignoring margin and variability.
2. Treating the cutoffs as fixed rules.
3. Applying the same service level to every class.
4. Never reclassifying.
5. Ignoring strategic items that are low in value but important.
6. Using ABC without a policy, so nothing changes.

Where the inventory sits compared with the sales

The most useful follow-up to an ABC classification is to compare each class's share of sales with its share of inventory. Suppose a company holds $5.0 million of inventory (illustrative) and the A, B and C classes account for 77, 17 and 6 percent of sales. If inventory is split as A $1.2 million, B $1.3 million and C $2.5 million, the shares are 24, 26 and 50 percent. Half the stock sits in items that generate six percent of sales. That is the signature of a long tail that has been allowed to grow, and it is the first place to look for cash. If C-class inventory could be cut by 30 percent through clearance, tighter reorder rules and delisting, the cash released would be 2.5 × 0.30 = $0.75 million, which is 15 percent of total inventory. At the same time, A items hold less than a quarter of the stock but drive most of the sales, so a single stockout on an A item costs far more than the carrying cost of the buffer. The classification therefore argues for investing a little more in A-class availability and pruning the tail.

Multi-criteria classification

Value alone sometimes misleads. A low-revenue item may be critical to a key customer, have a long and unreliable lead time, or be hard to replace. A multi-criteria approach scores each item on several factors and combines them with weights. For example, score revenue, margin and strategic importance from 1 to 5, and weight them 50, 30 and 20 percent. An item scoring 5 on revenue, 3 on margin and 4 on importance has a weighted score of 0.5 × 5 + 0.3 × 3 + 0.2 × 4 = 2.5 + 0.9 + 0.8 = 4.2. Items above, say, 4.0 are A, those between 2.5 and 4.0 are B, and the rest are C. The weights are judgments, so write them down and test whether the resulting classes make sense to the sales and operations teams. Keep the method simple enough that people can explain why an item sits in its class.

Edge cases

New items have no history. Classify them provisionally using an analog or the launch plan, and review after a few months. Seasonal items can look like C items in the off-season and A items in season, so use annual value, or classify by season. Kits and bundles should be classified at the level where stock is held. Spare parts and accessories may be low in value but critical to customer satisfaction; classify them by criticality. Items on contract with committed volumes should be flagged so that they are not caught by delisting reviews. Write down the rule for each case, so that different planners classify the same way.

FAQs

What is ABC analysis in inventory?

A method of ranking items by annual value and grouping them into A (most valuable), B and C (least valuable) so that control and service effort match importance.

What percentages define A, B and C?

Commonly A is the top 70 to 80 percent of value, B the next 15 to 25 percent and C the rest, but adjust to your range.

What is XYZ analysis?

A classification by demand variability. X items are predictable, Y moderate and Z volatile. It is often combined with ABC.

How often should I update the classification?

At least quarterly, and after major changes in the range or sales mix.

Should I use revenue or margin?

For inventory decisions, usage value or margin is often more informative than revenue. Choose one and apply it consistently.