Why your stock count is always wrong, and what to do about it
Stock drifts for a handful of ordinary reasons. Finding which one is yours is more useful than counting harder.
7 min read
In short
- Theft is real but is usually the smallest cause of stock drift. Miscoded sales and unrecorded deliveries are bigger.
- Count a few items every week rather than everything once a year — a days-old discrepancy can still be explained.
- Correcting the number without finding the cause guarantees the same gap next month.
- Decide a tolerance per category. Perfect accuracy costs more than it saves on cheap fast-moving lines.
You count the shelf, you check the app, and the numbers disagree. The instinct is to assume theft. Usually it is not. Stock drift almost always comes from a small number of ordinary causes, and identifying which one is yours is far more useful than counting more often.
The usual causes, roughly in order
- Sales recorded against the wrong item — two similar products, one barcode scanned by habit.
- Goods received but never entered, because the delivery arrived mid-rush.
- Partial deliveries recorded as complete, so the system believes you have stock the supplier still owes you.
- Breakage, expiry and samples given away, none of which are sales and none of which get recorded.
- Own use — items taken for the household and never rung through.
- Unit confusion, where an item is bought by box and sold by piece with no conversion.
- Theft, which is real but is usually the smallest of these.
| If the gap looks like this | Suspect this first |
|---|---|
| One item short, a similar one long | Miscoded sales between lookalike products |
| App higher than shelf, consistently | Deliveries recorded but short-shipped |
| Shelf higher than app | Goods received and never entered |
| Only fragile or perishable lines | Breakage and expiry, never adjusted |
| A whole category off by a factor | Box-versus-piece unit confusion |
Count a few items often, not everything rarely
A full stocktake once a year tells you that you are wrong, twelve months after it started going wrong. By then the trail is cold and there is nothing to learn.
Counting a handful of items every week is more work in total and far more useful, because a discrepancy is days old rather than months. You can still remember the delivery that came in wrong. Start with your fastest-moving and highest-value lines — the ones where drift costs the most.
Fix the cause, not just the number
When you find a discrepancy, resist correcting the count and moving on. The corrected number will be wrong again next month, because the reason is still there.
If two similar items keep getting confused, give them clearly different names or separate barcodes. If deliveries go unrecorded, record them against a purchase order so partial arrivals are visible. If breakage is the cause, record it as an adjustment so it appears in your costs rather than vanishing.
Low-stock alerts are a symptom check
Setting a reorder threshold per item is usually framed as an ordering tool, and it is. But it doubles as a drift detector: an item that hits its threshold much sooner than your sales rate suggests is telling you that stock is leaving without being recorded.
Accept a tolerance
No shop has perfect stock accuracy, and chasing it costs more than it saves. Decide what level of drift is acceptable for each category — tighter for high-value goods, looser for cheap fast-moving ones — and spend your attention where the money actually is.
Common questions
Why does my stock count never match the app?
Usually because of ordinary recording gaps rather than theft: sales rung up against a similar-looking product, deliveries that arrived mid-rush and were never entered, partial deliveries recorded as complete, breakage and samples that were never adjusted, and items bought by the box but sold by the piece.
How often should a small shop count stock?
A few lines every week beats a full count once a year. A weekly discrepancy is days old, so you can still remember the delivery that came in short. A yearly one tells you that you were wrong at some point in the last twelve months, which is not something you can act on.
Which items should I count first?
Your fastest-moving and highest-value lines. That is where drift costs the most money and where the cause is most likely to still be traceable.
Is a low-stock alert useful for finding stock loss?
Yes, as a side effect. An item that hits its reorder threshold much sooner than your sales rate would explain is telling you that stock is leaving without being recorded.
Khata is free to use
Billing, khata, stock and reports on your phone. No card, nothing to cancel.
See what you getinventorystockreportsstock count