The first food-waste baseline: what a factory should actually measure
National food-waste figures describe a country. They tell a factory nothing about its own losses. A useful baseline has to be built where the material actually arises.
- Byline
- Gambit Reign analysis
- Period covered
- 2022
- Reviewed
- 6 October 2026
- Topic
- Resource efficiency
- Reading time
- 3 min read
Key takeaways
- The first EU-wide food-waste monitoring release covered reference year 2020. It is a national and EU-level figure, not a factory benchmark, and cannot be used to judge a site's performance.
- Measure where waste arises — by process step and by reason — rather than in a single site total that no one can act on.
- Distinguish edible from inedible material and define by-product handling explicitly, or the same tonne will be counted twice.
Why a national number cannot be your baseline
Eurostat's first EU-wide food waste monitoring release, published in September 2022 and covering reference year 2020, established a common measurement across member states. [B] That is a significant achievement and the basis for tracking progress at national and Union level.
It is not a benchmark for a factory. A per-capita national figure aggregates households, retail, food service, primary production and manufacturing into one number. It says nothing about what a particular plant, making a particular product range, on a particular process route, ought to be achieving. Adopting it as a target would give a site a number it could meet or miss for reasons entirely unrelated to its own operations.
The only useful factory baseline is one built from the factory's own material flows. That is more work than reading a published figure, and it is the only version that produces an actionable result — because it identifies which process step is losing material and why.
Measure by process step, not by site total
A site total — say, a percentage of input lost as waste — is easy to report and almost impossible to act on. It does not tell you whether to change a preparation method, adjust a filling tolerance, revisit a supplier specification or address a storage condition.
Structuring the measurement by process step changes this. Receiving, storage, preparation, processing, filling, packaging, and finished-goods handling each have characteristic loss patterns and characteristic causes. Measuring at that granularity identifies the step, and the step usually identifies the remedy.
Alongside the step, capture the reason. A short closed list is sufficient and more reliable than free text: specification rejection, trim, spillage, overfill, changeover, machine stop, quality failure, damaged packaging, out-of-spec batch. Reasons convert a volume into a problem with an owner.
| Field | Why it is captured | Example values |
|---|---|---|
| Process step | Locates the loss where it can be fixed | Receiving, preparation, filling |
| Material | Distinguishes what is being lost | Ingredient, finished product, packaging |
| Quantity and unit | Enables mass balance and comparison | Kilograms, litres |
| Basis | Prevents wet/dry confusion | As received, dry matter |
| Edible / inedible | Determines correct destination and reporting | Edible, inedible |
| Reason code | Converts volume into a fixable cause | Trim, overfill, changeover |
| Destination | Establishes the route and its cost | Recovery, by-product, disposal |
| Cost basis | Connects the loss to margin | Input cost, disposal cost, both |
This register is our own working structure for organising factory-level measurement. It is not a regulatory form, a reporting template or official guidance.
Edible and inedible are not the same problem
Food waste and inedible material are frequently aggregated, and doing so obscures the decision. Material that was fit for consumption and was lost represents a direct loss of the value embodied in it. Material that was never edible — trimmings that no process would have used, for instance — is a different question, often answered through recovery or by-product routes rather than through prevention.
Separating the two changes priorities. Prevention effort directed at inedible material yields little; the productive question there is whether the recovery route is the best available. Prevention effort directed at edible loss addresses value that was genuinely destroyed.
The separation also affects how the numbers can be used externally. Where a business reports food waste, the composition matters to how the figure is read, and presenting a combined number without its composition invites misinterpretation in either direction.
The double-counting trap with by-products
Many food factories send material to a by-product route rather than to waste. Whether that material should appear in a food-waste figure depends on where the scope boundary is drawn, and the boundary must be defined before measurement begins rather than after.
The trap is counting the same tonne twice — once as process loss within the factory boundary, and again as waste at the point where the by-product is processed or disposed of downstream. Equally, a site that excludes all by-product tonnage may be presenting a figure that omits material a reader would expect to see.
The practical resolution is to state the boundary explicitly, record by-product tonnage separately from waste tonnage, and show the relationship between them. A reader can then see what has been included and on what basis, rather than having to infer it.
Establishing the routine
A baseline is only a baseline if it is repeatable. That requires a defined period, consistent boundaries, the same measurement points and a documented method for any estimates. Where a figure is estimated rather than measured, the method and reason should be recorded so that measured and estimated values can be distinguished later.
The first pass will be imperfect. That is normal and is not a reason to delay. A rough baseline with known gaps, improved over subsequent periods, is far more useful than an extended attempt at precision that never produces a number. The value comes from the comparison between periods, and that comparison requires a first period to exist.
Limitations
- This article describes factory-level measurement design. It does not state regulatory obligations or reporting requirements, and it is not food-safety or compliance advice.
- The source cited reports an EU-wide monitoring release. It is a national and Union-level dataset and should not be used as a site benchmark.
- The appropriate boundary, material categories and destination classifications depend on the business, its reporting obligations and the intended use of the data. These should be defined with reference to the applicable requirements.
The next decision
Decide your scope boundary and material categories before measuring anything — so that the first period's figures are comparable with the second.
Discuss your projectTaking this into your own project?
Our scoping guide and worksheet walk through the questions that make a brief usable — the decision, the evidence, the options including doing nothing, and what still has to be established. No email required.
Sources
External sources are referenced above by letter. Our own recommendations are identified as such in the text and are not attributed to these sources.
- [B]Eurostat — First EU-wide food waste monitoring release (25 September 2022, reference year 2020)https://ec.europa.eu/eurostat/web/products-eurostat-news/-/ddn-20220925-2