01Why we assess categories and not brands
A ranked list of named products would be more useful to a reader and more valuable commercially. We do not publish one, for a reason that is simple and non negotiable: we have not tested named products under controlled conditions, and publishing a ranking would mean publishing an assessment we did not make.
What we can do rigorously is assess how a category behaves. Whether shampoo bottles are typically collected, whether sunscreen claims are testable, whether serums typically disclose concentration, whether a refill option for body wash actually saves material. Those are answerable from published sources and from the structure of the market, and they are the questions that determine most of what a buyer can do.
02The five criteria
C1, formulation disclosure. Does the category routinely disclose more than the legal minimum. Concentration or form of an advertised active, pH where it determines performance, tested figures where a standard test exists, and a period after opening that is usable.
C2, claim substantiation. Are the category's headline claims testable and tied to a defined test or published standard, or do they rest on vocabulary with no legal definition. A tested SPF figure scores well. A clean claim scores badly, because there is nothing to test against.
C3, packaging recoverability. Is the dominant format actually collected and sorted through UK household recycling, and is mono material construction common. This is assessed against the real collection picture rather than against a recyclability symbol.
C4, lower material option. Does a refill, concentrate or waterless format exist for the category, and does it work without trading away performance. A format that saves packaging and loses a tested performance claim does not count.
C5, use phase burden. How much of the environmental impact sits with the user rather than with the manufacturer, and whether a behaviour change is available. For rinse off products heated water usually dominates, which is a category characteristic rather than a criticism.
03The four statuses, and why they are words
Each criterion gets one of four qualitative statuses: Strong, Mixed, Weak or Poor. They are words rather than numbers because a numerical score implies a precision we do not have and invites comparison arithmetic that the underlying assessment cannot support.
Strong means the category performs well and a buyer can identify a better than average product without difficulty. Mixed means practice varies widely and the difference is real but not visible from the front of the pack. Weak means the category norm falls short and exceptions take work to find. Poor means the category fails for structural reasons and individual product choice changes little.
04What moves a status
- Regulation changes. A new labelling requirement, a restriction, or an enforcement position that alters what a category must disclose.
- Collection changes. Greater consistency in UK household collection would move C3 for several categories at once.
- A format matures. A lower material option that moves from novelty to reliable performance moves C4.
- The evidence moves. Where an ingredient class acquires or loses supporting human evidence, the articles change, and where that alters a category's claim picture, C2 moves with it.
We record the change on the page rather than editing quietly, because a publication that revises silently is not offering you anything to check.
05How we weigh evidence in the articles
The same discipline applies to ingredients and claims. We describe the shape of an evidence base rather than inventing its contents: roughly how much controlled human research exists, whether it is consistent, whether it has been produced by parties with an interest in the outcome, and whether the outcome measured is one a user would notice.
We never invent a study, an author, a journal, a sample size or an effect size. Where we cannot describe an evidence base truthfully, we say it is thin or that we do not know, both of which are more useful than a confident sentence with nothing behind it.
06The limits of what we do
We do not run laboratory testing. We do not measure emissions. We do not audit supply chains. We do not evaluate the clinical practice of any clinic or practitioner, and where an article names an organisation, that is not an assessment of it.
Stating these limits is more useful than implying an authority we do not have. Where a question requires capabilities we lack, we describe what the relevant institutions have published and point you to them.
07Commercial separation
Assessments are written before any commercial arrangement is considered. A sponsor cannot see an assessment before publication, cannot request a change, and cannot be a sponsor while their own category is being assessed. An affiliate relationship never creates, changes or removes an assessment, and where one exists it is disclosed at the top of the page rather than in a footer.
If we ever cannot maintain that separation for a category, we will drop the category from the index rather than compromise the assessment. The commercial terms are published in full on the sponsorship page.