At the British Beauty Council’s industry day during British Beauty Week, Grace Vernon described today’s beauty shopper as a “skintellectual”.
She knows her retinal from her retinol, she reads the ingredient list, and she expects the science to stack up.
That’s good news for brands that have done the work, but also means a claim built on a supplier’s datasheet is easier to spot than it used to be, by shoppers and by regulators.
For years, a familiar route to a claim looked like this: A raw material supplier tests their ingredient and publishes an impressive result, the brand adds that ingredient to a formula and carries the result onto the pack: “clinically proven to boost hydration by 70%”. The study was real for that inci, but the trouble is that nobody tested the product the customer is actually buying.
A raw material result belongs to the raw material
Supplier data is generated under the supplier’s conditions: a set concentration, often a simple test base, and a panel chosen for that study. Your product is different in almost every way that matters; the concentration may be different, other ingredients can help or hinder the active, the pH, the texture, the packaging and the way people use it all change what the consumer experiences.
The rules have always pointed at the finished product. UK cosmetic claims must meet the common criteria carried over from EU Regulation 655/2013: legal compliance, truthfulness, evidential support, honesty, fairness and informed decision-making. Ingredient data can support a claim, but only where it genuinely represents what the finished product does.
Whilst the rules haven’t changed, the enforcement of these rules have, and brands are under much more scrutiny. Two ASA rulings this spring show where the line now sits:
- Eucerin, April 2026. A poster claiming “look up to 5 years younger” and “clinically proven” was banned. Among the problems, the peer-reviewed paper offered as support hadn’t tested the actual product, the main study relied on self-assessment with no control group, and the trial ran in a hotter, sunnier country than the UK (Lewis Silkin).
- 111skin, May 2026. Claims that a product was “powered by exosomes” and “clinically proven to lift the skin by 20%” were upheld against. There was no evidence the exosomes did what the ad said, and the delivery-system test used one person applying product to their arm (Osborne Clarke).
The ASA now scans ads at scale through its Active Ad Monitoring system, and the CMA can fine businesses directly for misleading consumers under the Digital Markets, Competition and Consumers Act 2024. A claim that might once have gone unnoticed is far more likely to be picked up today.
What good substantiation looks like
The good news is that strong substantiation is mostly about doing things in the right order. Start with the claim you want to make, then build the evidence for that claim, on that product.
The stronger the claim, the stronger the evidence it needs. As a rough guide:
| Claim type | Example | Evidence that usually fits |
| Basic function | “Cleanses”, “moisturising” | Formulation rationale, supported by testing on the finished product where needed |
| Perception | “9 out of 10 said their skin felt softer” | Consumer use test on the finished product, with a representative panel, reported as what people said |
| Measured result | “Hydration up 40% after two weeks” | Instrumental testing on the finished product, against a baseline |
| Clinical | “Clinically proven to reduce fine lines” | Controlled, blinded study on the finished product, with objective measures and an adequate panel |
A few principles hold across all of them:
- Test the finished formula. Ingredient data is still useful for choosing actives and explaining how a product works. It just can’t stand in for testing the product itself.
- Match the panel to your customer. Skin types, ages and conditions should reflect the people who’ll actually buy the product in the UK.
- Use the right measure for the words. Self-assessment supports perception claims. A measured or clinical claim needs objective measurement and, ideally, a control.
- Don’t underrate modest claims. A well-supported “moisturising” often carries less risk, and more credibility with a skintellectual shopper, than an inflated percentage.
- Keep the evidence together. Your claims evidence belongs in the Product Information File, ready to share if a retailer, the ASA or Trading Standards asks.
What this means for your brand
Claims risk scales with you. A claim that sat quietly on your own website gets a lot more attention once it’s on a retailer’s shelf, in a paid social ad or in front of a competitor with a reason to complain.
Three things worth doing now:
- Audit what’s already live. For every claim on pack, online and in ads, ask one question: can we point to evidence on the finished product? Where the honest answer is “the supplier’s data”, either test the product or soften the wording.
- Put claims in the NPD brief from day one. Deciding the claim after the formula is locked often means retesting, or settling for a weaker claim than you wanted.
- Bring your manufacturer in early. Formulation choices such as active level, pH and packaging decide whether a claim is achievable at all.
How we can help
At Group55 we have an in-house Regulatory Affairs & Product Safety Manager, Adiato Baldė-Wharton, who helps guide our customers through the regulatory and claim substantiation process. We run stability and compatibility testing in-house, manage third-party claims testing on our customers’ behalf, and advise on claims wording. Because we don’t sell claims testing ourselves, our advice is about the evidence your claim needs, nothing more.
Have a project that you’d like to discuss? Get in touch.
