Inventory record accuracy (IRA)
Inventory record accuracy is the share of counted items whose system record matches the physical count within a defined tolerance; it shows item by item how far stock records can be trusted.
Also calledIRA, stock record accuracy, inventory accuracy, stock accuracy rate
Formula
Items with records correct within tolerance ÷ Total items counted × 100- Item
- The unit of counting: a product-location combination (SKU and shelf or store).
- Tolerance
- The deviation within which a record counts as correct: zero units for high-value items, a small written percentage for low-value ones.
- Total items counted
- Product-location combinations counted in the period by full or cycle count.
Calculator
Enter your own figures and the result updates instantly. The defaults are the figures from the worked example below.
Result%96
Worked example
How to read it, and what good looks like
Record accuracy is tracked with two measures that say different things. Stock count variance rate measures value; when negative and positive variances cancel out, it can look small. Inventory record accuracy counts items and shows how far the record can be trusted operationally. Goods that appear in the record but are not on the shelf make the in-stock rate look optimistic, stall automatic ordering and hide lost sales.
Once branch stock is opened to online orders, record errors reach the customer directly. When online grocery orders are picked from the branch shelf, an item that shows in the system but is not on the shelf means a short delivery or a substitute; the customer sees the error at the door. The Turkish data available on the source of loss is old: according to the Sensormatic index cited by TPF (the Turkish Retailers Federation) in 2018, 18.43% of store losses came from administrative error, that is record and process error.
Tolerance and target are set by product group: tight tolerance and frequent counts for A items, wider tolerance for the long tail. The target comes from the quarterly band and trend of your own cycle count history. In Mindpera, Mind Stock freezes the balance the moment a count opens and breaks variance down by product, location, count team and date; when variance concentrates on a particular shelf, the agent opens a recount task for the warehouse supervisor.
Where it misleads
- Without a written tolerance, the same count gives two teams different accuracy rates; state the tolerance by product group in the report.
- If sales and dispatches continue during the count, correct records look wrong; freeze the balance when the count opens and hold interim movements separately.
- Counting only problem shelves makes the rate pessimistic, counting only easy ones makes it optimistic; plan the count sample by ABC class.
- Defining the item as a SKU makes a record with the right total but the wrong shelf location look correct; in a warehouse, the item is the product-location.
Data you need
- Count readings: product, location, quantity, count team and time
- System balance frozen at count start and the queue of movements during the count
- ABC class and tolerance table by product group
Mindpera products that calculate this KPI
Frequently asked questions
- How do you calculate inventory record accuracy?
- Divide the number of items whose system record matches the count within tolerance by the total number of items counted, and multiply by 100. If 12,000 of 12,500 items were correct, accuracy is 12,000 ÷ 12,500 × 100 = 96%. An item is a product-location combination; tolerance is written by product group, and the balance is frozen during the count.
- What is a good inventory record accuracy?
- There is no single universal value; set the target by product group and tolerance. Best sellers get tight tolerance and a high target, the long tail a wider tolerance. Take the target from the quarterly band of your own cycle count history; if the rate falls, find which location, team and product group it fell in.
- What is the difference between inventory record accuracy and stock count variance rate?
- Inventory record accuracy counts how many item records are correct and measures operational trust. Stock count variance rate divides the value of variance by stock value and measures financial impact. A value-based variance can look small because negatives and positives cancel out; item-based accuracy does not hide that. Report both in the same count report.