Auditing slow-moving and obsolete inventory

Auditing slow-moving and obsolete inventory means building an ageing analysis of stock by SKU based on last movement date, comparing the client's provision against a policy that is both reasonable and consistently applied, and testing whether items provided for (or not provided for) in a prior year were actually realized at or near their carrying value afterward. A written provisioning policy that nobody re-ran this year is the single most common finding in this area.

Building the ageing analysis

The ageing analysis is the foundation, and it needs to be built at SKU level, not at a category or stock-group level, because obsolescence is an item-specific condition. A category can look fine on average while individual SKUs inside it have not moved in two years.

1. 01Extract last movement date per SKULast date of any inward or outward transaction, not last date of a stock adjustment entry — a client that nets off a zero-movement item with a cosmetic adjustment each year will otherwise show recent activity that is not real activity.
2. 02Bucket by age since last movementCommon buckets: 0–6 months, 6–12 months, 1–2 years, 2–3 years, over 3 years. Tally this to the closing stock value so every rupee of inventory sits in exactly one bucket.
3. 03Cross-check against consumption rateAn item with no outward movement in 18 months but sitting at six months of average consumption may simply be a large single receipt, not slow-moving stock. Pair the ageing bucket with months-of-cover to avoid flagging normal lumpy procurement as obsolescence.
4. 04Flag items with movement but a declining trendA SKU moving steadily less each quarter is heading toward obsolescence even if its last transaction was recent. Ageing by last movement date alone will miss this; a trend view catches it earlier.

Testing the provisioning policy against reality

A policy that says '50% provision between one and two years, 100% beyond two years' is only useful if it was actually run against this year's ageing data and the resulting figure was booked. The test is not whether the policy sounds reasonable on paper — it is whether this year's provision, recomputed from the ageing analysis, matches it.

1. 01Recompute the provision from the ageing bucketsApply the client's own stated percentages to this year's ageing analysis, independently, and compare the result to the booked provision.
2. 02Investigate every item where policy and booking disagreeA manual override below the policy percentage needs a specific, documented reason — a confirmed forward order, a committed buyback arrangement, anything beyond 'management believes it will sell'.
3. 03Test subsequent realization of prior-year provided itemsTake last year's provided-for items and trace what actually happened to them this year: sold at a discount consistent with the provision, scrapped, or still sitting unmoved. If a large share of provided items simply sat unmoved for another year with no further action, the provision is being booked but not followed by any disposal decision.
4. 04Test the reverse: items not provided for that should have beenPick a sample of items in the oldest ageing bucket carried at full value and ask the client directly why no provision was taken. 'We expect to sell it' without a specific buyer or plan is not sufficient support.

Red flags that the policy is cosmetic

• The same provision percentage, or the same absolute rupee figure, booked two years running regardless of how the ageing profile has changed.
• A provision policy stated in the notes that has never once, across several years of audit files, produced a figure that differs from a round number chosen by management.
• Items that stay in the oldest ageing bucket for three, four, five years, provided at 100%, but never actually written off, scrapped, or disposed of — the carrying value sits at a notional zero or near-zero but the stock physically remains on the floor indefinitely.
• A sudden, large write-down in a single year with no corresponding build-up in prior years' ageing data, suggesting the issue was known and deferred rather than provided for as it arose.
• Management overriding the computed provision downward for specific items shortly before year end, timed to coincide with a results push.

None of these individually proves the provision is wrong. Together, they tell you whether the obsolescence policy is a live control or a disclosure exercise, and that assessment should shape how much the audit team is willing to rely on management's own judgment for the items not tested in detail.