How to measure and reduce fresh produce shrink in grocery retail
Who it is forGrocery chains, produce department managers, fresh category managers, store operations
The produce department is the area of a grocery store that both drives footfall and loses the most product. Shelf life is measured in days, goods are sold by weight and ripeness changes daily. The general shrink formula is not enough on its own here: the same percentage can come from over-ordering or from the wrong code keyed at the till. This guide splits produce shrink into three components, shows how to measure each and sets out the daily routine that brings the rate down. For the general shrink formula and category comparison, see the shrinkage rate guide.
How is fresh produce shrink calculated?
In produce, shrink is the cost of goods that leave stock without being sold, and it arises in three ways. Disposal is product thrown away entirely because it has spoiled. Trimming is what is removed while preparing goods for sale: rotten leaves, bruised fruit, dried stalks. Weight variance is the loss between the kilos weighed at goods receipt and the kilos sold at the till that disposal and trimming records do not explain; water loss and drying show up here.
The rate is calculated on value, at cost. A weight-based measure is useful operationally but inflates cheap, heavy produce: a crate of watermelon weighs far more than a crate of strawberries yet carries far less cost. Management reporting uses value; the department screen uses kilos.
The sum of the three components gives the department's shrink rate; each component's share of the total points to the cause. A high disposal share means over-ordering. A high trimming share means weak goods receipt quality or poor display. A high weight variance means drying, a cold chain gap or the wrong product code at the till.
- Produce shrink rate
(Disposal + Trimming + Weight variance) cost ÷ Net sales × 100- Weight variance (kg)
Kg received − Kg sold − Recorded disposal kg − Recorded trimming kg- Component share
Component cost ÷ Total shrink cost × 100
Chains report loss on different bases: BİM tracks food waste as a share of sales, Migros tracks food disposal tonnage against food supply tonnage. One is value-based, the other weight-based, and the two figures cannot be compared directly. Write the base you use at the top of your own report.
Why record weight loss, trimming and disposal separately?
The three components have different causes and different fixes. Written into a single "shrink" line, the department manager cannot tell what to do and the meeting is spent arguing over the number.
Trimming is the record most often skipped. Product removed while the display is set up in the morning goes into the bin without being weighed; the loss surfaces at the count as weight variance and is mistaken for drying. Yet a high trimming share usually points to a weak batch accepted at goods receipt without inspection.
Part of the weight variance is not shrink at all. When the wrong price look-up (PLU) code is keyed at the till, one product shows a shortfall and a similar product shows a surplus. If two products show weight variances of similar size in opposite directions, it is a code mix-up, not a loss.
| Component | How it is measured | Typical cause | Intervention |
|---|---|---|---|
| Disposal | Discarded product weighed and entered with a reason code | Over-ordering, late markdown | Daily ordering, stepped markdowns |
| Trimming | Material removed during preparation is weighed | Weak batch, rough handling, stacking too high | Quality check at receipt, display height |
| Weight variance (drying) | Kg received − kg sold − recorded loss | Water loss, time outside refrigeration | Cold room dwell time, misting |
| Weight variance (code error) | Opposite variances on similar products | Wrong PLU at the till, scale list differs | Align till and scale PLU lists |
| Receipt shortfall | Delivery note kg versus receipt weight | Goods accepted without weighing | Weigh every batch at receipt |
Disposal and trimming are recorded in the department on a handheld. Any record that requires going back to the office to fill in a form gets skipped, and the loss returns at the count as an unexplained variance.
Which products are most exposed, and how do you read differences between stores?
The department total is enough for management but not for action. Decisions are made on a product × store matrix: products in rows, stores in columns, shrink rate in each cell. The matrix shows two different patterns, and each has its own fix.
If one product is high in every store, the problem lies with the product or its supply: a batch arriving with short remaining life, fragile packaging, orders placed in large blocks. If one store is high across every product, the problem lies in that store's process: ordering habits, refrigeration, rotation, display.
Academic evidence supports the distinction. In a study of a grocery chain in Ankara published in the Turkish journal Verimlilik Dergisi on 30 April 2024, low demand ranked as the most important driver of waste, followed by storage capacity, shelf life, location and customer characteristics. The same study found the most effective reduction strategy to be markdowns made at the right time.
| Product | Store A | Store B | Store C | Reading |
|---|---|---|---|---|
| Tomatoes | 4.8% | 5.1% | 4.6% | Product pattern: check supply and shelf life |
| Bananas | 2.1% | 6.9% | 2.4% | Specific to B: ripening room or ordering |
| Parsley | 9.5% | 10.2% | 8.8% | Product pattern: bunch size and order frequency |
| Apples | 1.2% | 3.8% | 1.4% | Specific to B: display and rotation |
The figures in the matrix are illustrative. Store B stands out on two products at once; that says the problem is in the store's process rather than the products, and the audit starts there.
How much do automated ordering and timely markdowns reduce shrink?
The main source of produce shrink is the order. According to the food waste section of Migros's 2020 sustainability report, the company moved produce to automated ordering driven by a learning algorithm in 2018. In the first 114 stores where it was tried, produce disposal fell by 3.1 points, and the system was then rolled out to every store.
There is also a legal reason the order matters so much. Article 6 of Türkiye's Law No. 6585 on the Regulation of Retail Trade and Article 4 of the Regulation on the Principles and Rules for Retail Trade classify returning products that spoil within 30 days of production to the supplier as "unsold" after delivery as an unfair trading practice. The consequence is plain: the shrink on over-ordered fresh produce stays with the retailer, and order accuracy is a direct shrink cost.
The second lever is the markdown. In produce the trigger is ripeness, not a date: ripened product is separated during the morning trim and moved the same day into a reduced pack or a separate display. A decision left until closing time usually ends in disposal.
- Weigh and inspect at goods receiptEvery batch is weighed against the delivery note. A batch that fails the quality threshold is refused; a weak batch accepted today is tomorrow's trimming.
- Record the morning trimMaterial removed while setting up the display is weighed and entered under the trimming code, at the display, on a handheld.
- Separate ripening produce the same dayProduct past the ripeness threshold moves to a reduced pack or a separate display. The decision is not left until closing.
- Set end-of-day markdowns in stepsSteps and discount rates for the hours before closing are written in advance; neither the cashier nor the department manager decides on the spot.
- Feed shrink data back into orderingThe order factor for product-store pairs with a high disposal share is reduced. Until the shrink report feeds the order, the same shrink recurs every week.
The detail and exceptions of the unfair trading rule are set out in the regulation itself. Read the text with legal counsel when drafting supplier contracts.
How is produce shrink measured and tracked in Mindpera?
In Mindpera, shrink is entered with a reason code in the Mind Stock module and calculated as a shrink rate by category and store. When a threshold is crossed an alert is raised and a task drops to the department manager. Shrink and count variance records are posted to accounting entries through the chart-of-accounts mapping table, so the financial statements and the stock report do not diverge.
On the ordering side, tomorrow's expected sales come from the forecasting model. Mind Send reads that signal alongside recent sales velocity and store cluster coefficients, and produces a replenishment proposal for any store that falls below its ideal stock level. In Mind Ops, the morning trimming and rotation checklist is completed on the shop floor; an open item drops to its owner and escalates until it is closed.
- Mind Stock: reason-coded shrink records, shrink rate by category and store, alerts when thresholds are crossed
- Mind Send: ideal stock level and replenishment proposals from sales velocity; the planner approves or changes each line and every decision leaves a trail
- Mind Ops: trimming and rotation checklist, open item and escalation report
- Agents find the cell that stands out in the product × store matrix, open the task and follow it until the work is closed
If shrink, sales and goods receipt data are scattered between the ERP and scale systems, installation brings them into a single data model.
Frequently asked questions
- How is the fruit and vegetable shrink rate calculated?
- Add up the cost of produce disposed of, trimmed away and lost on the scales, then divide by net sales. For example, a department with ₺180,000 of net sales and ₺9,720 across the three components has a shrink rate of 5.4 per cent. The components are kept apart because disposal points to over-ordering, trimming to receipt quality and weight variance to drying or code errors.
- Which reasons should produce shrink be recorded under?
- Four codes are enough: disposal (spoiled product), trimming (material removed during preparation), drying and weight loss, and receipt shortfall. Variances caused by the wrong product code at the till are tracked separately; if two similar products show opposite weight variances, it is a code mix-up, not a loss. A short code list gets filled in; a long one gets skipped.
- How do you reduce shrink in the produce department?
- Three things run together: daily, product-level ordering; moving ripening product to a markdown or a separate display the same day; and recording the trim. Every batch is weighed and inspected at goods receipt. Until shrink data feeds back into the order factor, the same product keeps producing shrink in the same store every week.
- Can unsold fresh produce be returned to the supplier in Türkiye?
- As a general rule, no. Article 6 of Law No. 6585 and Article 4 of the Regulation on the Principles and Rules for Retail Trade classify returning products that spoil within 30 days of production as unsold after delivery as an unfair trading practice. The shrink on over-ordering therefore stays with the retailer. Check the exceptions in the regulation with legal counsel.
- How much does automated ordering reduce shrink?
- Results vary by business, but there is a public Turkish example. According to Migros's 2020 sustainability report, the company moved produce to automated ordering with a learning algorithm in 2018, and produce disposal fell by 3.1 points in the first 114 stores where it was tried. A lasting effect depends on feeding shrink data back into the ordering model regularly.
- What should fresh produce shrink be?
- There is no published, comparable category benchmark for Türkiye; chains report loss on different bases. Build the target from your own data: take the product × store shrink matrix for the last twelve months and use the rate achieved by the best-quartile stores on each product as the target for the rest. Shrink is not driven to zero; a shelf empty at closing is a lost sale too.
Sources
- Migros, Gıda İsrafını Önleme Çalışmalarımız (food waste prevention section of the 2020 Sustainability Report) — 2020
- Law No. 6585 on the Regulation of Retail Trade, Art. 6 (consolidated text, Turkish) — 2014-01-29
- Regulation on the Principles and Rules for Retail Trade, Art. 4 (consolidated text, Turkish) — 2016-08-06
- Güneş, Akyol, Güven, Eren: Reducing waste rates in grocery chain inventory management, an application in Ankara, Verimlilik Dergisi — 2024-04-30