Expecting the auditor to count everything
Counting belongs to the business; the auditor judges the method.
Stock audit glossary · Counting methods
noun · an audit procedure at a stock count
A test count is the auditor’s own count of selected items during a physical count, compared with what the business’s counting team recorded for the same items. It runs in both directions, from the count sheets to the floor and from the floor to the sheets, and it tests whether the counting procedure can be relied on, not only whether the items picked are right.
Meaning
An auditor attending a physical count is not there to count the stock. Counting is the business’s job and its staff do it; the auditor is there to judge whether the way they are counting can produce a reliable figure. Counting selected items personally and comparing them with the team’s records is how that judgement is formed. Where inventory is material, SA 501 expects the auditor to attend the count, evaluate the procedures, inspect the inventory and perform test counts.
The test runs both ways on purpose. Items picked from the sheets and found on the floor show that recorded quantities exist; items picked from the floor and traced to the sheets show that everything present was counted. A test in one direction cannot find stock that was simply missed. Where differences appear, what matters is what they say about the process: a pattern of differences usually leads to the count being extended or repeated, not to individual lines being corrected.
At the count
The day’s core work for the attending auditor, deliberately limited.
In practice
How the exercise is run says much about how far its result can be trusted.
Do the recorded figures describe goods that are really there?
Was anything on the racks left off the sheets?
Each selected item against the team’s figure, with every difference recorded.
A clean result supports the wider count; a pattern leads to extending, repeating or reporting it as unreliable.
Worked example
Two groups, treated differently.
| Test group | Book / counted (INR) | Treatment |
|---|---|---|
| High-value items, 40 lines, every line checked | 3,10,00,000 / 3,08,40,000 | 0.52% short; a fact, not projected |
| Random sample, 60 lines | 90,00,000 / 88,20,000 | 2.0% short; projected |
| Untested balance | 4,20,00,000 / not counted | Projected shortfall 8,40,000 |
Every high-value line was checked, so its 0.52% difference describes that group and is not carried anywhere else. The random sample is a sample in the statistical sense, so its 2.0% shortfall is applied to the untested balance.
The projected 8,40,000 is what the auditor sets against materiality. A difference the business cannot explain usually leads to a recount of the affected area rather than to a larger sample.
Watch out
The purpose is misread more often than the procedure is done badly.
Counting belongs to the business; the auditor judges the method.
They are the least likely to be wrong.
Items never written onto a sheet stay invisible.
The difference is evidence about the process.
The count is finished on a method already shown to fail.
Those lines get extra care and stop being typical.
Patron Accounting LLP. The statement on SA 501 on this page signed off by CA Sundram Gupta: 23 October 2026
Sources: Standard on Auditing 501 (ICAI) and general stock-count practice. General information, not professional advice; check the current position before acting.
Need help with a test count? See our stock audit service. A stock audit attends the count and tests it both ways.