What ABC Analysis Is
Walk through almost any warehouse and you will find that items are not equally important. A spare parts distributor may stock three thousand SKUs, yet a few dozen of them account for most of what it spends on stock each year, while thousands of screws, tapes and O-rings together make up a small slice. Many companies still treat every item the same way: same count frequency, same follow-up, same buying effort.
ABC analysis sorts items into three classes by annual consumption value, so management effort follows financial impact. Class A is a small number of high-value items that deserve the tightest control. Class C is a long tail of low-value items that can be handled with simple rules. Class B sits between them.
The idea comes from the Pareto principle, the observation that a small share of causes often drives a large share of results, popularly called the 80/20 rule. Inventory rarely splits at exactly 80/20, but some concentration shows up in most businesses. This guide works through the calculation with a full example, sets out a practical policy for each class, and covers the method's limits and how to complement it.
The Three Classes at a Glance
| Class | Typical share of SKUs | Typical share of consumption value | Control level |
|---|---|---|---|
| A | 10% to 20% | 70% to 80% | Tight and continuous |
| B | 20% to 30% | 15% to 25% | Regular |
| C | 50% to 70% | 5% to 10% | Simple |
These are common rules of thumb, not a standard. Choose cut-offs that fit your own distribution, then keep them fixed and documented so the classification is applied the same way at every review.
How to Calculate It
- Pick the period. Usually the last twelve months, so every season is included. A shorter window can inflate or bury a seasonal item.
- Pull annual consumption per item. Quantity sold or issued to production, taken from system transactions rather than estimates.
- Set the unit cost. Weighted average cost or last purchase cost, as long as the same basis is used for every item. Use cost, not selling price, because the point is capital invested.
- Compute annual consumption value: annual quantity × unit cost.
- Sort descending by consumption value.
- Compute the cumulative percentage of total value down the list.
- Apply the cut-offs, for example up to 80% cumulative is A, 80% to 95% is B, the rest is C.
Worked Example: An HVAC Parts Distributor in Jeddah
To keep it readable, take a sample of ten items. All figures are hypothetical:
| Item | Unit cost (SAR) | Annual qty | Consumption value (SAR) | Share | Cumulative | Class |
|---|---|---|---|---|---|---|
| AC compressor | 1,500 | 300 | 450,000 | 45% | 45% | A |
| Control board | 2,300 | 100 | 230,000 | 23% | 68% | A |
| Fan motor | 550 | 200 | 110,000 | 11% | 79% | A |
| Expansion valve | 175 | 400 | 70,000 | 7% | 86% | B |
| Temperature sensor | 50 | 1,000 | 50,000 | 5% | 91% | B |
| Run capacitor | 35 | 1,000 | 35,000 | 3.5% | 94.5% | B |
| Copper pipe (metre) | 25 | 1,000 | 25,000 | 2.5% | 97% | C |
| Air filter | 15 | 1,000 | 15,000 | 1.5% | 98.5% | C |
| Insulation tape | 5 | 2,000 | 10,000 | 1% | 99.5% | C |
| Mounting screws (box) | 2.5 | 2,000 | 5,000 | 0.5% | 100% | C |
Total consumption value is SAR 1 million. Three items, 30% of the sample, carry 79% of the value; four items, 40% of the sample, carry just 5.5%. In a real catalogue of thousands of SKUs the concentration is usually sharper, so A items are a smaller share of the count.
Notice that ranking by quantity would produce the opposite order. Screws and tape move in the largest volumes but sit at the bottom by value. A team that allocates attention by transaction count or carton volume ends up busy with the items that matter least.
A Practical Policy for Each Class
Classification changes nothing unless each class is actually managed differently:
| Aspect | Class A | Class B | Class C |
|---|---|---|---|
| Cycle count | Monthly or more | Quarterly | Twice a year or annually |
| Reorder point review | Monthly | Quarterly | Annually |
| Forecasting | Item-level with human review | Automatic with periodic review | Average usage is enough |
| Safety stock | Calculated tightly to service level | Standard formula | Generous, because it is cheap |
| Order size | Smaller, more frequent | Economic order quantity | Large, infrequent |
| Supplier negotiation | Intensive, annual contracts | Periodic price reviews | Simple agreements, consolidate suppliers |
| Purchase approval | Higher authority | Standard | Wider buyer discretion |
| Storage | Secured, monitored area | Standard locations | Bulk storage |
The safety stock row looks backwards at first. Why hold more buffer on cheaper items? Because running out of a box of screws can stop the installation of a unit worth thousands, while two extra boxes cost almost nothing to hold. Every extra compressor, on the other hand, ties up SAR 1,500, so A-class buffers are calculated carefully. See our guides to safety stock and economic order quantity.
The Effect on Working Capital (Hypothetical)
Back to the Jeddah distributor. Before the analysis it held two months of cover on every item. On SAR 1 million of annual consumption, that is about SAR 166,700 of stock, around SAR 131,700 of it in A items.
After classifying, it cut A cover to three weeks with weekly or fortnightly orders and close monitoring, kept B at six weeks, and raised C to three months to reduce the number of small orders:
- Class A: from about SAR 131,700 to about SAR 45,600.
- Class B: from about SAR 25,800 to about SAR 17,900.
- Class C: from about SAR 9,200 to about SAR 13,800.
Total stock falls from about SAR 166,700 to about SAR 77,200, even though C-class stock went up, and C-class orders became quarterly. The numbers are illustrative; real results depend on lead times, supplier reliability and demand variability. The logic holds though: tighten where the money is, relax where follow-up costs more than the item is worth.
Using ABC to Schedule Cycle Counts
One of the most useful applications is replacing a single annual wall-to-wall count, which can shut a warehouse for days, with continuous cycle counting. Take 2,000 SKUs: 200 A items counted monthly, 500 B items quarterly, 1,300 C items once a year.
- Class A: 200 × 12 = 2,400 counts
- Class B: 500 × 4 = 2,000 counts
- Class C: 1,300 × 1 = 1,300 counts
That is 5,700 counts a year, roughly 23 items per working day over 250 days. A small team can do that in an hour or two daily without stopping operations, and the highest-value items get checked twelve times a year instead of once. Our stocktaking guide covers counting methods in detail.
A year-end count still matters for the financial statements and legal requirements; Saudi commercial books rules require year-end inventory to be recorded. Regular cycle counts simply make that year-end count faster and less surprising.
ABC and Inventory Valuation
Under IAS 2, as applied in Saudi Arabia, inventory is measured at the lower of cost and net realisable value. Items whose expected selling price has dropped, that have become obsolete, or that are partly damaged must be written down. Reviewing NRV across thousands of items is heavy work, and the classes help prioritise:
- A items: review individually, since a small drop in the selling price of a compressor or control board moves total stock value noticeably.
- Moving B and C items: a group-level test is usually sufficient.
- Slow movers in any class: a report of items with no movement for six or twelve months flags write-down candidates.
IAS 2 does not permit LIFO, so the unit cost used in the analysis should be FIFO or weighted average, consistent with your books.
Where ABC Falls Short
- It ignores operational criticality. A cheap spare for a key production machine may land in C even though running out stops the plant.
- It ignores demand variability. Two A items may need very different policies if one is steady and the other erratic.
- New items have no history and fall into C, even if they are strategic products meant to grow.
- Seasonality skews results if the period is too short.
- It goes stale. Prices and demand change, so reclassify every quarter or half-year.
Complementing ABC
XYZ for demand variability
XYZ classifies items by how predictable demand is: X steady, Y moderately variable or seasonal, Z sporadic. It is usually measured by the coefficient of variation, the standard deviation of monthly consumption divided by its mean. Combined with ABC it gives nine cells:
| X steady | Y variable | Z sporadic | |
|---|---|---|---|
| A | Tight automatic replenishment, frequent orders | Detailed forecast, monthly review | Buy to order where possible, case-by-case review |
| B | Standard automatic replenishment | Automatic forecast, quarterly review | Higher buffer or buy to order |
| C | Large orders or supplier-managed refill | Generous buffer | Question whether to stock it at all |
The CZ cell is the classic source of slow-moving stock: cheap, rarely demanded items that sit on shelves for years. Our article on dead stock covers how to deal with them.
Criticality
Maintenance and spares teams often classify items as vital, essential or desirable. A vital item is handled cautiously even in class C, because the cost of a stockout far exceeds its value. In practice, add a "critical" flag to the item record that overrides the ABC result when setting safety stock.
Velocity for slotting
Warehouse slotting should follow pick frequency, not value. A cheap item picked a hundred times a day deserves a spot near packing more than an expensive compressor picked weekly. Use a separate velocity classification alongside ABC, and keep high-value A items in a secured area regardless of speed.
The Same Logic Beyond Inventory
Rank suppliers by annual spend and you will usually find a handful taking most of the budget. They deserve annual contracts, scorecards and review meetings; small suppliers can go through a lightweight buying process. Rank customers by revenue or margin and you see who merits a dedicated account manager and priority allocation when stock is short.
A Two-Week Rollout
- Days 1 to 3: extract twelve months of consumption and unit cost per item from the system; exclude discontinued items.
- Days 4 to 5: compute values, ranking and cumulative shares; agree cut-offs with finance and purchasing.
- Days 6 to 7: review results with warehouse leads to catch critical items that landed in C.
- Days 8 to 10: write the policy per class: count frequency, reorder point reviews, approval limits, stock cover.
- Days 11 to 14: record the class on item master data, build the cycle count calendar, and set the next reclassification date.
Common Mistakes
- Using selling price instead of cost, which inflates high-margin items and misstates capital invested.
- Using on-hand balance instead of consumption. A dead item with a big balance lands in A despite never moving. That is a useful slow-stock report, but it is not ABC.
- Classifying once and forgetting. After a year the classes no longer reflect reality.
- Classifying without changing procedures. If counts, approvals and forecasting stay identical for all classes, the exercise is wasted.
- Ignoring branches. An item can be A in one branch and C in another; multi-site companies should classify at company and warehouse level.
Where the ERP Fits
A spreadsheet can do the analysis once. Repeating it every quarter for thousands of items and wiring the result into daily procedures takes a system that provides consumption and cost from actual transactions, stores the class on each item record for use in count schedules, reorder points and approval rules, and reports stock by warehouse and class.
In Tranquil, the inventory management module handles central inventory, multiple warehouses and multiple units of measure, linked to purchasing, sales and accounting, so the data the analysis needs comes from one source. A classification report built around your own criteria can be set up through customization services.
Frequently Asked Questions
Do the classes have to split 80/15/5?
No. Those are common guidelines. Choose cut-offs that fit your distribution and keep them stable between reviews so results are comparable.
Should I use cost or selling price?
Cost in most cases, because the aim is to measure capital invested and the effort to manage it. A separate ranking by revenue or margin is useful for sales decisions.
How often should I reclassify?
Quarterly or half-yearly for most companies, using the last twelve months of data so all seasons are included.
What about a cheap item that is critical to operations?
Flag it as critical so the flag overrides the ABC class when setting safety stock and purchase priority, or run a parallel criticality classification.
Does Tranquil support ABC analysis?
Tranquil keeps stock, consumption and cost data in one system, and an ABC report to your criteria can be built through customization services. Contact the Tranquil team to discuss what you need.