# Inventory Record Accuracy (IRA)

> Source: https://netray.co/glossary/inventory-record-accuracy
> Publisher: Netray (netray.co), AI and integration engineering for ERP and manufacturing systems.
> Section: Glossary

## Short answer

Inventory record accuracy (IRA) is the percentage of item-location records whose system quantity matches the physical count within an agreed tolerance. It shows how far MRP and order promising can trust the on-hand balances in the ERP.

## Aliases

- IRA
- inventory accuracy
- stock accuracy
- inventory accuracy percentage

## Category

erp-core

## Full Definition

IRA is measured by counting a sample of items or locations and comparing counted quantity to system quantity. A record is accurate if it falls within tolerance, for example exact match for high-value items and plus or minus a few percent for bulk items. IRA is accurate records divided by records counted.

The metric matters more than it looks because MRP, available-to-promise and replenishment all assume on-hand balance is correct. A record that shows 20 when 0 exist will suppress a purchase order, and a plant discovers the shortage at the line. Planning systems commonly aim for 95 percent or better, and many lean programs aim higher.

The usual way to measure and improve IRA is cycle counting. Infor SyteLine has cycle count and physical inventory functions, NetSuite has inventory counts, and Infor LN has inventory counting in warehousing. Counting by ABC class, on transaction triggers (zero balance, negative balance, receipt) and by problem-area frequency finds causes, not just variances.

Root causes are usually process, not people: unrecorded scrap, unit of measure conversion errors, backflush assumptions that differ from reality, mis-picks, receiving to the wrong location, and transactions posted late. Fixing the cause is what moves IRA; recounting alone just resets the error.

## Why It Matters

- MRP and available-to-promise outputs are only as good as the on-hand quantities beneath them.
- Low IRA drives hidden costs: expediting, line stoppages, excess safety stock held to compensate.
- It can justify replacing annual wall-to-wall counts with continuous cycle counting.
- Auditors accept strong IRA as support for inventory balances without a full shutdown count.

## In Practice

A stockroom counts 400 item-locations in a month. 372 match within tolerance, so IRA is 93 percent. Analysis of the 28 misses shows 14 trace to backflush quantities that ignore scrap, 8 to wrong-location receipts and 6 to unit of measure errors. Correcting the backflush scrap factor and adding a receipt location scan lifts IRA to 97 percent the following quarter.

## Faq

### How do you calculate inventory record accuracy?

**Answer:** Divide the number of counted item-locations whose system quantity matches the physical count within tolerance by the total number counted, then multiply by 100.

### What is a good IRA target?

**Answer:** MRP-driven plants usually aim for 95 percent or better, and mature lean or high-value operations target 98 percent. The target should be set by class, tighter for A items.

### Is value accuracy the same as record accuracy?

**Answer:** No. Value accuracy compares total dollars, where over and under counts can cancel out. Record accuracy compares each item-location, which is what planning actually depends on.

## Search terms

inventory record accuracy, IRA formula, inventory accuracy percentage, how to improve inventory accuracy, inventory accuracy ERP, cycle count accuracy, inventory accuracy target, stock record accuracy

---
Talk to an engineer: https://netray.co/contact?intent=erp_ai | Machine catalog: https://netray.co/agent/catalog.json