A material name is not a unit of comparison
Procurement teams are often asked to choose between labels: bamboo or plastic, recycled or virgin, reusable or single-use, repaired or replaced, local or imported. The labels sound like alternatives, but they do not yet describe the same job.
A bamboo tray may be lighter than a reusable container but deliver fewer uses. A repaired machine may avoid producing a new machine but require replacement parts, transport, testing, and more electricity during operation. A refill pouch may use less packaging per sale while changing product loss or shelf life. Comparing one kilogram, one item, or one purchase can hide the service that the decision is supposed to protect.
The first control is a functional unit: a clear statement of the quantified performance that all options must deliver. It turns the question from 'Which material sounds circular?' into 'What resources and impacts are associated with delivering the same result?'
Start with the decision, not the dataset
A comparison only needs to be as broad as the decision it must support, but it must be broad enough to avoid a false answer. Before requesting supplier figures, write who will use the result and what action may follow: shortlist a packaging format, redesign a service, extend an asset's life, commission a formal assessment, or narrow an environmental claim.
The Life Cycle Initiative's criteria for good practice say that goal and scope - including the functional unit, system boundary, and impact methods - should align with the decision context. For comparisons, the alternatives need equivalent goal and scope. That does not mean every option has identical processes. It means the study asks the same decision question of each option.
A screening worksheet can reveal whether the current evidence is strong enough to proceed. It is not a substitute for a formal life-cycle assessment, independent review, engineering judgement, safety review, or legal substantiation when those are required.
Write the functional unit in four parts
The European Commission's Product Environmental Footprint method defines the functional unit through four questions: what service is provided, how much of it is provided, how well it performs, and for how long. Leaving one part vague can move the result more than changing the material.
What names the function: protect a shipment, provide a drink, clean a floor, illuminate a workspace, keep food usable, or move a passenger. How much quantifies the service: litres served, square metres covered, tonne-kilometres moved, or hours of operation. How well sets the performance threshold: barrier quality, hygiene, strength, brightness, temperature, reliability, or another relevant requirement. How long sets the period or lifetime over which the function must continue.
A useful sentence could be: 'Protect and deliver 10,000 litres of beverage at the required food-safety and damage threshold across the defined route for one year.' The exact unit depends on the decision. The discipline is to make every option answer the same sentence.
- What: the function or service the user actually needs.
- How much: the quantified amount of that service.
- How well: the quality or performance level that must be met.
- How long: the service period, lifetime, or number of uses covered.
Reference flow turns equal service into product quantities
Once the functional unit is fixed, each option receives a reference flow: the amount of product needed to deliver the defined function. The European Commission method distinguishes the service being compared from the products required to provide it.
For a durable option, the reference flow may include the original unit, expected replacements, spare parts, losses, repair materials, and cleaning inputs over the service period. For a single-use option, it may include every item consumed to deliver the same number of services. For a coating, it may be the kilograms needed to cover the same area at the same opacity for the same years.
This is where circular claims become testable. A product designed for one hundred uses should not be credited with one hundred uses if the operating system records only twelve before loss or failure. A lighter product should not be rewarded if more units, more product loss, or additional protection are needed to provide equal service. Use measured operation where available and label estimates clearly.
Align the boundary before comparing the result
Equal service does not guarantee a fair comparison if one option carries a wider system boundary. The boundary identifies which life-cycle stages and processes are included: raw materials, manufacturing, transport, use, maintenance, return, washing, repair, product loss, and end of life.
The Commission's method follows a supply-chain logic from material acquisition through production, distribution, use, and end-of-life treatment, while requiring exclusions to be justified. For an operator screening, the immediate task is to place the same relevant stages on both sides and make any exclusion visible.
Do not include washing and return transport for a reusable system while ignoring replacement purchases for single-use packaging. Do not count production savings from repair while omitting a change in energy use during the extended operating period. Do not credit a recycling route without evidence that the material reaches that route. The boundary should follow the function, not protect a preferred answer.
Before and after: comparing service packaging
Before: a buyer compares the mass of one compostable fibre bowl with one reusable polypropylene bowl. The fibre bowl is lighter. The reusable bowl contains more material. The buyer concludes that the lighter item is environmentally preferable.
The comparison misses the service period, number of meals, return rate, rejected units, washing, transport, breakage, food loss, and actual end-of-life routes. It compares objects, not meal-service delivery.
After: the team defines a functional unit such as serving 100,000 meals at the required hygiene, leak-resistance, and temperature performance across named locations for one year. The fibre option's reference flow counts every bowl needed. The reusable option counts the starting pool, replacements, losses, washing, collection, and units retired. Both options include relevant production, transport, use, and end-of-life stages under aligned assumptions.
The answer may differ by route, return behaviour, energy mix, product loss, and impact category. That is not a failure of the method. It is evidence that the decision depends on how the system actually operates.
Durability, repair, and reuse need operating evidence
Circular strategies often aim to extend product life, increase utilisation, or keep materials in use. Those are operating hypotheses until records show what happens. Design life is not achieved life. Repairability is not a completed repair. Reusability is not a completed rotation.
For durable and repairable products, record failure rates, maintenance, replaced components, downtime, efficiency changes, quality limits, and the point at which another intervention is needed. For shared or reused products, record utilisation, return, loss, inspection, cleaning, and retirement. These fields determine the reference flow and may also change transport, energy, or service quality.
The useful comparison is therefore not 'new versus repaired' in the abstract. It is the amount of new and repaired equipment, parts, service activity, and operating resources needed to deliver the defined output and reliability for the same period.
Test uncertainty instead of hiding it
Early circular projects rarely have perfect data. Return rates may come from a pilot, product lifetime from a supplier test, end-of-life outcomes from regional averages, and transport from a route model. The honest response is not to present one precise number as settled truth. It is to show which inputs are measured, estimated, assumed, or unknown.
Run simple sensitivity cases around the inputs most likely to change the decision: lifetime, number of uses, loss rate, repair frequency, electricity, washing, transport distance, recycled content, product loss, and end-of-life route. If a small change reverses the ranking, the decision is sensitive and should be described that way.
The Life Cycle Initiative also warns against relying on a single impact score or mismatched boundaries. Climate change may be important, but water use, resource use, toxicity, pollution, land use, or other impacts can matter for the decision. The right set depends on the product and context. A carbon-only screen should not be promoted as proof of overall environmental superiority.
A practical comparison worksheet
Use this sequence before procurement approval, supplier comparison, product redesign, or a public environmental statement. It is a scoping and evidence-control tool, not a certification method.
- Decision: name the person who will use the result and the exact action the comparison can support.
- Alternatives: include the current system and realistic alternatives that can deliver the required service.
- Functional unit: write what, how much, how well, and how long in one reviewable sentence.
- Reference flows: calculate the products, replacements, parts, losses, returns, cleaning, and other flows each option needs.
- Boundary: align relevant production, distribution, use, maintenance, return, repair, product-loss, and end-of-life stages.
- Evidence status: mark every input as measured, supplier-provided, modelled, estimated, assumed, or missing.
- Context: record geography, technology, energy, route, user behaviour, and time period that could limit transferability.
- Impact coverage: choose the environmental indicators relevant to the decision and state what is not assessed.
- Sensitivity: test whether realistic changes in major assumptions reverse or materially alter the result.
- Review: assign an owner, competent reviewer, evidence date, version, limitations, and correction trigger.
- Claim boundary: communicate only the conclusion supported by the defined options, scope, data, and review.
What this method can and cannot establish
A well-defined functional unit can expose false equivalence, direct data collection, and make supplier proposals more comparable. Reference flows can show how many products and supporting processes each option needs. Aligned boundaries and sensitivity checks can reveal where a preferred circular strategy depends on uncertain operating assumptions.
This worksheet cannot establish universal environmental superiority. A conclusion remains specific to the alternatives, function, geography, technology, data period, impact categories, and assumptions assessed. A screening result may show where to investigate next without being sufficient for an external comparative claim.
Formal comparative assertions can require method-specific rules, qualified practitioners, critical review, and current legal substantiation. The Product Environmental Footprint method itself includes detailed modelling, data-quality, impact-assessment, interpretation, reporting, and verification requirements beyond this operator guide.
The practical principle is simple: equal service first. Once the service is fixed, material choice, durability, reuse, repair, and end-of-life evidence can enter a comparison that another person can inspect and challenge.
Start with one live decision
Choose one decision already waiting for approval: a packaging tender, a repair-or-replace choice, a recycled-content option, a refill system, or a material substitution. Ask the project owner to write the functional unit before asking vendors for a footprint number.
If the team cannot agree on what, how much, how well, and how long, pause the comparison. That disagreement is useful. It shows that the alternatives are being judged against different services or expectations.
When the functional unit is stable, calculate one transparent reference flow per option, align the boundaries, mark weak evidence, and test the assumptions that could reverse the decision. A circular comparison earns trust by making its conditions visible, not by making the material label louder.