Cycle counting: what it is and how to run a cycle count
Cycle counting means counting a small part of your stock on a regular schedule instead of counting everything once a year. Each count is compared with the system record, the differences are investigated and corrected, and over the cycle every item is counted at least once.
What is cycle counting?
A cycle count is a planned count of selected items or locations. It is done every day or every week as part of normal work. The aim is not only to correct the quantities. It is to find out why the record was wrong, so the same error stops happening.
A full physical count, also called a wall-to-wall count or stocktake, counts everything in one go. The two are compared below.
| Cycle count | Full physical count | |
|---|---|---|
| What is counted | A small part of the stock each time | Everything, in one go |
| When | Daily or weekly, all year | At year end or period end |
| Operations | Normally keep running | Usually stopped or slowed |
| When errors are found | Soon after they happen, while the cause can still be traced | Long after they happen, when the cause is hard to trace |
| Effort | Small and steady | Large and concentrated |
Cycle counting methods
The methods differ in how the items for each count are chosen. Many warehouses combine two of them.
| Method | How items are chosen | Good for |
|---|---|---|
| ABC | Items are grouped by value or usage. A items are counted most often, C items least often. | Putting the counting effort where an error costs the most |
| By location | Locations are counted in sequence, aisle by aisle, until the whole warehouse has been covered. | Finding stock in the wrong place and making sure nothing is missed |
| Random sample | A random set of items or locations is picked for each count. | A fair measure of accuracy across the whole stock |
| Control group | The same small group of items is counted again and again over a short period. | Testing the counting process itself before a wider programme starts |
| Event based | An item is counted when something happens: the bin is empty, stock is low, or a picker reports a problem. | Quick counts when there is little to count, and following up on known problems |
How often to count: a worked example
With the ABC method you decide how many times a year each group is counted. The figures below are an example, not a rule. Set your own from the value of the items and how often their records are wrong. To sort your own items into A, B and C groups by annual usage value, paste your item list into the free ABC Analysis tool.
| Group | Items | Counts per year | Counts needed |
|---|---|---|---|
| A | 100 | 12 | 1,200 |
| B | 300 | 4 | 1,200 |
| C | 600 | 1 | 600 |
| Total | 1,000 | 3,000 |
Counts per day = total counts per year ÷ working days per year
With 250 working days, 3,000 ÷ 250 = 12 item counts a day. If that is more than your team can do well, lower the frequency of the B and C groups before you cut the A group.
How to run a cycle count
- Pick the items or locations for the day from the schedule.
- Clear the paperwork first. Post all receipts, issues and transfers for those items, or hold movements until the count is done. Unposted transactions are a common cause of false differences.
- Count blind. Give the counter the item and the location, not the system quantity.
- Compare the count with the system.
- Recount every difference outside your tolerance, if possible by a second person.
- Find the cause before you adjust. Check recent transactions, nearby locations and the unit of measure.
- Adjust with approval and record the reason.
- Record the accuracy for the cycle so you can see the trend.
Measuring the result
The usual measure is stock record accuracy: the share of counted items whose record was right, within a written tolerance. The guide to stock record accuracy gives the formula and a worked example. Track the figure cycle by cycle, and keep a simple list of the causes you find. The causes that come up most often show you which process to fix.
Common mistakes
- Counting while stock is moving, with no cut-off for receipts and issues.
- Showing the system quantity to the counter, who then tends to confirm it.
- Adjusting the record without finding the cause, so the same error returns.
- Skipping empty locations. An empty bin that the system says holds stock is a difference too.
- Counting only the easy locations and leaving high racks or bulk areas for later.
- Deciding the tolerance during the count instead of writing it down beforehand.
Common questions about cycle counting
What is the difference between a cycle count and a physical inventory?
A cycle count covers a small part of the stock and is repeated all year. A physical inventory counts everything at one time, usually with operations stopped.
How many items should be counted each day?
Add up the counts needed in a year for every group and divide by the number of working days. In the example above that is 12 item counts a day.
Should the counter see the system quantity?
No. A blind count, where the counter does not know the expected figure, gives a more honest result.
What is ABC cycle counting?
It is cycle counting where items are grouped by value or usage, and the most important group is counted most often. The ABC Analysis tool works out the groups from your own item list.
Does cycle counting replace the annual stocktake?
It can, when the records are shown to be accurate. Whether a full count is still required is a decision for your finance team and auditors.
Written and checked by a warehouse supervisor with about 15 years of hands-on experience. Last updated: October 2026. All guides