Environmental decisions
Make the comparison smaller and more honest.
Use ChatGPT or Claude to structure environmental comparisons, inspect evidence boundaries and write narrower, supportable claims.
Updated
Name the decision and the equivalent service
‘Which packaging is greener?’ is too broad to produce a dependable answer. Define the product, service, location, period and options. A comparison should make clear what each option has to accomplish, such as protecting the same quantity of product through the same delivery journey.
This guide offers a questioning framework, not a lifecycle assessment or regulatory interpretation. A reliable conclusion requires appropriate measurements and expertise. AI can organise the evidence and expose gaps; it should not fill them with invented impact values.
Help define an environmental comparison for [decision]. The options are [options] and must provide [equivalent service]. List the relevant stages, local context, available evidence and missing measurements. Separate what we can compare now from what requires specialist assessment.
Keep the boundary attached to the number
A figure without its unit, period and included stages is difficult to interpret. Ask whether the source covers production, transport, use, disposal or only part of that sequence. Record exclusions beside the result rather than leaving them in a distant footnote.
When reports disagree, first check whether they answer the same question. Different boundaries or assumptions can explain a difference without either report being fraudulent or useless.
Create an evidence table from these sources. For each result record the unit, location, period, included stages, exclusions and stated assumptions. Reference the exact passage. Explain which results are comparable and which should not be combined.
Test a change in assumptions
Illustrative scenario: a reusable item is compared with a single-use alternative. The result may depend on actual reuse, transport and cleaning assumptions. Do not assign a universal break-even number without relevant data. Ask which assumption would most affect the decision and how you could measure it locally.
Using only the supplied evidence, identify assumptions that could change the comparison. Where the data permits, show alternative scenarios and calculations. Where it does not, specify the missing measurement. Do not invent environmental conversion factors or a break-even threshold.
Write a claim that does not outrun the evidence
A narrow, specific statement is easier to check than a broad claim of environmental superiority. Describe what changed, the comparison period and what was measured. A reduction per item can coexist with higher total use, so inspect both when relevant.
Before external publication, have consequential environmental claims reviewed for the applicable requirements. This workflow does not certify compliance.
Compare our proposed statement with the evidence. Mark wording that implies a wider benefit than measured. Draft a narrower alternative stating the scope, baseline and limitations. List unresolved questions for expert review before publication.
Common questions
Can AI tell me which option is most sustainable?
It can structure a comparison, but a defensible result depends on relevant evidence, equivalent service and explicit boundaries. Missing data should remain visible.
Is a lower per-item impact enough?
Not necessarily for the question you are answering. Check total activity and whether the apparent improvement shifts an effect outside the measured boundary.
Keep exploring
About this workflow
This is an original AI-assisted editorial workflow. Illustrative scenarios are not customer case studies, and the prompts have not been systematically evaluated on live models. Check consequential outputs against your sources and use qualified review where needed.