AQ Consultancy

Stock and Fixed Asset Verification

A stock audit verifies that inventory recorded in the accounts actually exists, in the quantity and condition stated. For trading, retail, manufacturing and food and beverage businesses, stock is frequently the largest balance sheet item and the least verified — and a difference between recorded and physical stock is a profit adjustment, a control failure, or both. AQ Consultancy conducts stock counts and inventory verification for businesses in Dubai and Abu Dhabi.

Why stock is where the differences accumulate

Cash is counted daily, banks are reconciled monthly, and receivables are chased because customers are chased. Stock is counted once a year, sometimes by the people responsible for it, and the difference is often written off without anybody asking where it went.

That is how a business ends up carrying inventory in the ledger that has been sold, damaged, taken or was never received in the first place. In a business where stock is thirty or forty per cent of the balance sheet, an unverified stock figure means an unverified balance sheet, an unverified gross margin, and a profit figure that has not actually been tested.

And because stock differences build gradually, the year they are finally addressed produces an adjustment that belongs to several years rather than one.

Which businesses this applies to

Trading and distribution businesses. Retailers, particularly multi-location ones. Manufacturers, where work in progress and raw materials add valuation questions to existence questions. Food and beverage businesses, where waste, portion control and short shelf life make the reconciliation genuinely difficult.

Also businesses being audited, where the auditor will attend or require a count. Businesses being sold or valued, where stock is a material component of the price. And businesses that have found a difference and want to understand its cause rather than write it off again.

The work, step by step

What this looks like in practice:

  1. Plan the count. Locations, cut-off, whether trading stops, how movements during the count are handled, and who counts what — not the people ordinarily responsible for that stock.
  2. Establish the cut-off. Goods received not invoiced, goods invoiced not despatched, and stock in transit. Cut-off errors account for a large share of apparent differences.
  3. Count physically, with test counts and a second count on high-value or high-variance lines.
  4. Assess condition. Damaged, obsolete and slow-moving stock identified rather than counted as good, which is where valuation rather than existence goes wrong.
  5. Reconcile to the records, line by line for significant items rather than in total.
  6. Investigate the differences. A difference is a symptom; the cause is either process, valuation, cut-off or loss, and they are dealt with differently.
  7. Test the valuation basis — cost or net realisable value, whichever is lower, applied consistently.
  8. Report with the adjustment required and the control findings behind it.

Where the differences actually come from

Before assuming loss, the ordinary explanations have to be eliminated — and in most counts, they account for the majority of the variance:

  • Cut-off — goods received but not booked, or despatched and not relieved. The largest single cause of apparent differences
  • Unrecorded returns, from customers or to suppliers
  • Internal consumption — samples, staff use, marketing stock, never journalised
  • Unit of measure errors — cases counted as units, or the reverse, which produce large and confusing variances
  • Damage and wastage not written off as it occurred
  • Assembly and kitting not reflected in the records, in businesses that combine or break down products
  • Actual loss — theft or misappropriation, which is what remains once the above are excluded

The order matters. Concluding loss before eliminating cut-off and unit-of-measure errors leads to an accusation the evidence does not support, which is worse than no conclusion at all.

Counting properly

A count is only evidence if it is conducted in a way that could have detected a difference. Several things determine that, and they are easy to get wrong:

Counters should not be the people ordinarily responsible for the stock they are counting. Count sheets should not show the expected quantity, because a counter who can see the number will find it. Movements during the count must be controlled or halted. High-value lines should be counted twice, independently. And the count should be reconciled to the records afterwards rather than adjusted to them during it.

That last point is the one that quietly destroys the value of most internal counts. A count that is corrected to agree with the system has verified nothing at all.

Common mistakes

The expensive mistakes in this area are consistent:

  • Counting with the expected quantity printed on the sheet, which guarantees the count agrees.
  • Stock counted by the people responsible for it, which removes any independence from the exercise.
  • No cut-off control, so movements during the count create differences that are not differences.
  • Writing off the variance without establishing which of the ordinary causes explains it.
  • Counting existence and ignoring condition, so obsolete stock is carried at full cost.
  • Annual counts only, in a business where the difference took a year to build and cannot now be traced.
  • Adjusting the count to the system rather than reconciling one to the other.

Deadlines that apply

At the financial year end at minimum, and attended by the auditor where an audit is required — a count the auditor did not attend and could not verify may not be acceptable evidence.

Beyond that, cycle counting through the year is considerably more useful than a single annual count: high-value and high-movement lines counted monthly or quarterly, everything else on a rotation. Differences found within weeks can be traced. Differences found after a year usually cannot.

What lands on your desk

  • A physical count reconciled to the records, line by line for significant items
  • Differences analysed by cause rather than reported as a single variance
  • Condition assessment identifying damaged, obsolete and slow-moving stock
  • A valuation review against cost and net realisable value
  • The adjustment required, quantified
  • Control findings and recommendations to reduce future variance

What to have ready

The list is short and you will have most of it already:

  • Stock listing from the system as at the count date
  • Details of all storage locations, including third-party and in-transit stock
  • Goods received and despatch records around the cut-off
  • The valuation basis and costing method used
  • Details of consignment or customer-owned stock held
  • Prior count results and any adjustments made
  • Staffing available to support the count

How this is priced

Quoted per count, on the number of locations, the number of stock lines and whether trading has to be interrupted. Multi-location retail is the largest of these; a single-warehouse trading business is the smallest.

Cycle counting programmes are quoted annually and are usually better value than one large annual count, because they spread the cost and, more importantly, because a difference found in March can still be explained.

Related

Frequently Asked Questions

How often should we count stock?

At the financial year end at minimum. Beyond that, cycle counting — high-value and high-movement lines monthly or quarterly, everything else on rotation — is considerably more useful, because a difference found within weeks can be traced and one found after a year usually cannot.

Why does our stock never agree to the system?

Usually cut-off errors first, then unrecorded returns, internal consumption, unit-of-measure mistakes and unwritten-off damage. Actual loss is what remains after those are eliminated, and concluding loss before eliminating the others leads to an accusation the evidence does not support.

Should our own staff count the stock?

Not the staff ordinarily responsible for the stock being counted. And count sheets should not show the expected quantity — a counter who can see the number will find it. Independence is what makes a count evidence rather than a formality.

Does the auditor need to attend?

Where stock is material to the financial statements, generally yes. A count the auditor did not attend and cannot otherwise verify may not be acceptable audit evidence, which can result in a qualified opinion on an otherwise sound set of accounts.

What about obsolete stock?

It has to be identified during the count and valued at the lower of cost and net realisable value. Counting existence and ignoring condition is how a business ends up carrying unsellable stock at full cost, which overstates both assets and profit.

We have stock in several locations. Does that complicate it?

It affects planning rather than method — counts should be simultaneous or the transfers between locations controlled, otherwise stock in transit can be counted twice or not at all. Third-party and consignment stock needs separate confirmation.

What do we do with the difference?

Analyse it by cause first, then adjust. The accounting adjustment is the easy part; the useful output is the control findings, because a variance written off without a cause is a variance that will recur.

When did you last verify your stock?
Tell us how many locations and roughly how many stock lines. Cycle counting is usually better value than one annual count, and considerably more useful.
Check my compliance status 058 101 9570

Last reviewed 27 July 2026. Rates, thresholds and deadlines change — the e-invoicing provider deadline has already moved once. Confirm current requirements with the Federal Tax Authority before acting, or ask us to check your position.