Relying on a client's cycle counts
When an auditor can rely on a client's cycle counting instead of a single year-end count, what to test first, and when a full count is still needed.
By Achekam · · 5 min read · For Accountants & audit firms in South Africa, Nigeria, Kenya
The client's operations director is proud of the new warehouse system. Stock is counted a section at a time throughout the year, every location gets covered, and the records are corrected as the counts come in. She would very much like to skip the year-end shutdown count, which means weekend overtime and a pause in dispatches. Can the auditor accept that?
Sometimes, yes. Cycle counting can produce inventory records accurate enough to support the year-end figure without a full count. But the auditor has to test that the system really works before relying on it, and the test is more demanding than watching a single count. This article sets out what to look at.
What cycle counting is
AccountingTools defines cycle counting as counting a small amount of inventory each day with the aim of counting the entire inventory over a period of time. Its guide to cycle counting explains how a well-run programme works, and it reads as a useful checklist of what an auditor should expect to find.
The idea is that, instead of one big count that sets the records right once a year, many small counts keep them right all year. The year-end quantity then comes from the perpetual records rather than from a count taken on the last day.
AccountingTools also notes the benefit from the auditor's side: if outside auditors can rely on inventory records kept accurate by cycle counting, they may scale back their audit procedures and reduce their fees. That "if" carries the weight of everything that follows.
When records are accurate enough to rely on
Reliance on cycle counts is reliance on a control: the counting programme and the correction of the records that follows it. The auditor therefore needs evidence that the programme operated effectively throughout the year. Questions to answer:
- Does the programme cover every location and every item over the year, with higher-value or faster-moving items counted more often?
- Are counts done by people independent of the custody of that stock, or at least reviewed by someone who is?
- Are counts done blind, without the counter seeing the recorded quantity first?
- Are transactions processed up to date before each count?
- How large and how frequent are the variances, and is the trend improving or worsening?
That fourth question matters more than it looks. AccountingTools warns that if outstanding inventory transactions are not entered before a cycle count, a counter's adjustment followed by the late transaction can leave the record less accurate than before. It adds that this is particularly common when the same item is stored in several locations. A cycle count programme with poor transaction discipline can make the records worse, not better.
Testing the client's variance investigations
AccountingTools explains that errors found in cycle counts should lead to an adjustment of the inventory records and an investigation into the reason for each error, and it describes investigating variances, and discussing them with the warehouse manager, as the most important cycle counting activity.
For the auditor, that is where to look hardest. Select a sample of count variances from across the year and, for each:
- Confirm the variance was recorded and the records adjusted.
- Check the adjustment was approved by someone other than the counter.
- Read the investigation: was a cause found, such as a late receipt, a picking error or theft?
- See whether the cause was fixed, so that the same error stopped recurring.
A programme that adjusts records without investigating causes is just a series of small write-offs.
Shelf to record and record to shelf
AccountingTools explains that cycle counters should trace items they see on the shelf back to the counting report, as well as checking the report against the shelf, in case some items are not recorded in the database at all.
The auditor's own test counts should run both ways for the same reason. Record to shelf, picking items from the inventory listing and finding them, gives evidence of existence. Shelf to record, picking items in the warehouse and finding them in the listing, gives evidence of completeness. A client whose counters only work from the list will never find stock that is missing from it.
Observing counts during the year and recording them in Achekam
Attending some of the client's cycle counts during the year, rather than one full count at year end, gives the auditor direct evidence of how the programme operates. Choose counts at different times and locations, observe the procedure, perform test counts in both directions, and follow the variances through to adjustment.
In Achekam, the Inventory Observation module saves its work on the server for the chosen engagement, so observations from several visits can sit on the same file. The ISA numbers on its badge are labels, not evidence of compliance. For the variance testing, staff can load the year's count adjustments as a population in Audit Intelligence; Achekam risk-scores it and proposes a sample, a reviewer who did not prepare the plan approves it before it is drawn, and each draw can be reproduced from its stored seed. The programme itself, as a control, belongs with the rest of the inventory walkthrough; see walkthroughs and controls.
When a full count is still needed
Reliance is not the right answer when:
- variances are large or frequent, or are growing through the year;
- investigations are missing, superficial or never lead to fixes;
- transactions are routinely processed late;
- parts of the warehouse or some item categories were never counted;
- the counting is done by the people responsible for the stock, without review;
- the year-end figure is so material, or the risk of misstatement so high, that direct evidence at year end is needed.
In those cases, the honest advice to the client is that the year-end count stays, and why. Attending the inventory count covers that work. A well-run cycle count programme is one of the best controls a stock-holding client can have, and when the evidence shows it working, relying on it is good auditing rather than a shortcut.
General guidance only, not advice on your own circumstances. Tax and payroll rules change: check the current position with the authority or your adviser.
