How to calculate stock loss in a supermarket and set a loss target
Who it is forGrocery chains, area managers, store managers, loss prevention, finance and internal audit
Questions such as "supermarket count shortfall", "what is stock loss" and "stock-loss bonus" are among the most searched topics for staff and managers at Turkish grocery chains, where the shortfall is known as envanter açığı. The question looks simple but carries two jobs at once: calculating the loss correctly, and making sure the target attached to it steers the store towards the right behaviour. This guide covers how the stock that should be in the system is calculated, why the loss is expressed against sales, what the sources say about loss causes and how to set the target. For the count itself and closing individual variances, see the stock count and count variance guides.
How is supermarket stock loss calculated?
Stock loss is the difference between two stock figures. The first is the stock that should be in the system: every receipt during the period is added to the opening count and every recorded issue is deducted. The second is the stock physically counted at period end. Both are valued with the same cost method; if one side is valued at weighted average cost and the other at last purchase cost, the loss grows or shrinks artificially.
Recorded shrink is not part of the loss. Product entered as shrink in the department has already reduced system stock; stock loss carries only the unrecorded part. That is why the loss is never read on its own: a low figure can mean good loss control or over-recording of shrink.
Net sales is used as the denominator. The ratio to inventory value measures count accuracy and is covered in the count variance guide; the ratio to sales gives the effect on profit and makes stores of different sizes comparable. The loss rates reported by chains and indices are also expressed against sales.
- Stock that should be in the system
Opening count + Goods received + Transfers in − Cost of sales − Recorded shrink − Transfers out and returns- Stock loss
Stock that should be in the system − Counted stock (both at cost)- Stock loss rate
Stock loss ÷ Net sales for the period × 100- Total loss rate
(Recorded shrink + Stock loss) ÷ Net sales for the period × 100- Unknown loss share
Stock loss ÷ (Recorded shrink + Stock loss) × 100
Stock loss is only as accurate as the count. A loss figure from a count taken without a freeze, or closed after a single count, cannot separate real loss from counting error.
How large is stock loss relative to sales, and what do the sources say?
No recent, public store loss measurement for Türkiye could be found; the figures available are dated, and the indices' definition of loss may not match a business's own stock loss calculation exactly. They still give a sense of scale.
According to the Sensormatic loss index cited in TPF's article "On store losses" of 20 August 2018, loss as a share of sales stood at 1.79 per cent for regional chains and 1.73 per cent for hypermarkets. According to the Global Retail Theft Barometer reported by Retail Türkiye on 15 January 2015, the loss rate in Türkiye for 2013-2014 was 0.99 per cent against a global average of 1.29 per cent.
Large chains manage loss as a separate function. ŞOK Marketler's interim report for the period to 30 September 2025 lists an "Operational Control and Loss Prevention Director" on its executive committee. In its 2025 integrated annual report, Migros identified a 30 basis point improvement in its loss rate as the main driver of the improvement in gross profit.
According to the index cited by TPF, 34.34 per cent of losses come from external theft, 24.28 per cent from suppliers, 22.95 per cent from employees and 18.43 per cent from administrative error; two thirds of the loss arises from causes other than external theft.
How you read the distribution matters: two thirds of the loss comes from causes other than shopper theft. Supplier shortfalls and administrative errors are closed by goods receipt and recording discipline, not by security measures.
How do you break stock loss down by cause?
Stock loss is read by breakdown, not as a total: by product, department, supplier and cashier. The four main causes leave different traces in the data; which one dominates is read from those traces and the intervention is chosen accordingly.
Supplier-related loss is the easiest to close. With suppliers delivering direct to store, goods accepted without scanning the delivery note are entered in full even when short, and the gap surfaces at the count. Calculating stock loss by supplier shows the pattern at the first count.
Employee-related loss leaves its trace at the till: voids and refunds, and movements after closing. A cashier's void rate standing clearly apart from the store average is where the review starts; a high rate can point to a training gap or a till layout problem as much as to theft.
- Stock loss rate by supplier
Stock loss on the supplier's products ÷ Net sales of the supplier's products × 100- Cashier void rate
Value of lines voided by the cashier ÷ Cashier's sales value × 100Example rule: a cashier more than three standard deviations above the store average is reviewed.
| Cause | How it shows in the data | Control point |
|---|---|---|
| External theft | Recurring loss on small, high-value items and in the same shelf area | High-risk item list, shelf position, camera event log |
| Supplier-related | Loss concentrated on the products of direct-delivery suppliers | Scanning and weighing at receipt, stock loss by supplier |
| Employee-related | Cashier void and refund rates standing apart, movements after closing | Till void-refund report, matching till events to sales records |
| Administrative error | Opposite variance pairs, unmatched transfer and return lines | Barcode scheme, transfer approval, record reconciliation |
How do you set a stock loss target and a stock-loss bonus?
A single percentage target does not fix the process; it only changes which line the loss is written to. Tie the bonus to stock loss alone and unrecorded loss gets written up as shrink, inflating shrink. Tie it to shrink alone and shrink goes unrecorded while stock loss grows. The two measures are twins and can only be targeted together.
The right scorecard carries three measures: the total loss rate (recorded shrink plus stock loss), the unknown loss share and the share of stock loss with an identified cause. The first shows the size of the loss, the second recording discipline and the third whether the store owns its loss.
The numeric value of the target comes from the store's own history and from the best quartile of stores in the same cluster. The cluster matters: a large, fresh-heavy store and a small, ambient-heavy store do not carry the same target.
| Target design | What it encourages | Side effect |
|---|---|---|
| Stock loss rate only | Making the loss look small | Unrecorded loss is written up as shrink; shrink inflates |
| Shrink rate only | Making shrink look small | Shrink goes unrecorded; stock loss grows |
| Total loss rate | Reducing real loss | Does not show whether the cause is being recorded |
| Total loss + unknown share + cause identification rate | Reducing loss and recording it | More measurement effort; needs a single data model |
Whether a count shortfall can be deducted from staff pay is a question of employment law; the answer depends on the employment contract and the circumstances, and it is outside the scope of this guide. Take legal advice before any such decision.
How is stock loss calculated and followed up in Mindpera?
In Mindpera a count is opened in the Mind Stock module; the moment it opens, the balance is frozen and movements during the count are held in a separate queue. Stock loss is calculated in one place by comparing the frozen balance, the interim movement queue and the count result, and broken down by product, location, count team and date, with value and units reported separately. Because shrink is entered with a reason code in the same module, recorded shrink and stock loss sit side by side in the same report, and the approved variance is posted as an accounting entry.
In Mind Ops, unmatched lines between the POS and the ERP are listed with a reason code and value difference; events at the till, entrance and shelf are logged with a timestamp and linked to the sales record, and a till event that does not match a transaction raises an alert. In Mind Cost, average cost and last purchase cost are held side by side, so it is clear which cost the loss has been valued at.
- Mind Stock: immutable count snapshot, variance analysis by breakdown, shrink and stock loss read in one report
- Mind Ops: unmatched POS lines, camera event log and till anomaly alerts
- Mind Cost: both cost methods side by side, valuation difference on the loss
- Agents find the location or team where the loss concentrates, open a recount task for its owner and follow it until the work is closed
If count, sales and shrink records are scattered between the ERP and store systems, installation brings them into a single data model.
Frequently asked questions
- What is supermarket stock loss?
- It is the shortfall between the stock that should be in the system at period end and the stock physically counted. It carries unrecorded shrink, theft, supplier shortfalls and recording errors together. Recorded shrink is not part of it, because it has already reduced system stock; that is why stock loss is read together with recorded shrink.
- How is stock loss calculated?
- Add goods received and transfers in to the opening count; deduct cost of sales, recorded shrink, transfers out and returns. The result is the stock that should be in the system. Subtract the counted stock and divide the result by net sales for the period. For example, ₺105,000 of stock loss on ₺16,800,000 of net sales gives a stock loss rate of 0.63 per cent.
- What level of stock loss relative to sales is normal?
- There is no recent public benchmark for Türkiye. According to the Sensormatic index cited by TPF in 2018, loss stood at 1.79 per cent of sales for regional chains and 1.73 per cent for hypermarkets. The Global Retail Theft Barometer gave 0.99 per cent for Türkiye in 2013-2014. The figures are dated and defined differently; set the target from your own history and the best stores in the same cluster.
- What should a stock-loss bonus be tied to?
- To three measures rather than one percentage: the total loss rate (recorded shrink plus stock loss), the unknown loss share and the share of stock loss with an identified cause. A bonus tied to stock loss alone inflates shrink records; one tied to shrink alone pushes shrink off the record and grows stock loss. Unless both are targeted together, the loss simply moves between lines.
- How do you find the cause of stock loss?
- Read it by product, department, supplier and cashier. Loss concentrated on small, high-value items points to theft; loss concentrated on a direct-delivery supplier points to goods receipt; a cashier's void rate standing apart points to the till process; opposite variance pairs point to recording errors. According to TPF 2018, two thirds of losses come from causes other than external theft.
Sources
- TPF (Turkish Federation of Retailers), Mağaza kayıpları üzerine, citing the Sensormatic loss index (Turkish) — 2018-08-20
- Retail Türkiye, report on the Global Retail Theft Barometer 2013-2014 (Turkish) — 2015-01-15
- ŞOK Marketler, interim annual report 01.01-30.09.2025 (Turkish) — 2025-09-30
- Migros, 2025 Integrated Annual Report (KGK archive, Turkish) — 2025