The first time I wrote a product page for a new serum, I drafted it the way I had seen everyone else draft it. “Clinical studies show…” with no citation. “Proven to reduce…” with no source. “Up to 47% improvement…” with no paper named. I sent it to our science advisor. She returned it with one comment, written in red across the top: “where.”
That comment changed how we write everything now. If we cannot name the paper, we cannot make the claim. If the paper does not say what we want it to say, we cannot stretch it. If the paper says it for a different population, we say so. The discipline is harder than the alternative, and it is the only version of this brand we are willing to run.
Why brands hide the citation
Three reasons, in order of frequency.
First, the claim does not have a citation. The “clinical study” was an internal panel of twelve people, none of whom were blinded, conducted by the brand’s own marketing team. Naming it would reveal it. So it gets called “clinical” without identification.
Second, the citation exists but does not support the claim. The paper showed a 12% improvement in a parameter the brand is rounding to “up to 47%” by selecting the top-quartile subgroup. Naming the paper would expose the rounding. So the paper does not get named.
Third, the citation is real and supports the claim, but the brand does not want to engage in a literature conversation with a public it considers unsophisticated. This is the most defensible of the three reasons and also the most paternalistic. Readers are more sophisticated than brands assume.
The Tier-1 standard
When we cite a study, we follow a tier system, and we publish the tier alongside the citation when it matters. Tier 1 is peer-reviewed, published in a recognized dermatology, biochemistry, or cosmetic-science journal, with the full study available for inspection. Tier 2 is a credible industry source — AAD, BAD, EADV, NIH, government health agency — with a public document we can link. Tier 3 is internal data or a vendor’s white paper, and we only cite it if we explicitly label it as such and explain its limitations.
The Microbiome Glow Serum page is Tier 1 throughout. The peptide claims on the BioCell Renewal Cream are anchored to published papers in the Journal of Cosmetic Dermatology and the British Journal of Dermatology. The Mindful Masks line cites Tier 2 sources for the calming claims and is honest that the relaxation benefit, while real, is partly ritual and partly biochemistry.
The contrarian section: most “clinical” claims you read are marketing fiction
The word “clinical” in cosmetics is legally almost meaningless in most jurisdictions. A panel of fifty people in a room doing self-assessment is “clinical” by the working industry definition. A double-blind randomized trial with placebo controls is also “clinical.” The two are not the same kind of evidence, and the brand has no obligation to tell you which one applied. The Federal Trade Commission in the US and the Advertising Standards Authority in the UK have both fined brands for misleading clinical claims, but the policing is sparse compared to the volume of claims being made.
If a product page says “clinical study shows” without naming the study, you can assume one of three things. The study is bad. The study does not say what the page implies. Or the study does not exist as conventionally understood. The legal department’s job is to keep the wording close enough to true that it cannot be successfully sued. The wording is not the same as the science.
What open citation actually requires
This is the part nobody talks about because it sounds boring. Naming a paper is one thing. Being open to having that citation challenged is another. We have an email address for science questions. The advisor reads it. When a reader writes in pointing out that we have cited a paper in a way the paper does not quite support, we either update the page or write a longer note explaining why we read it differently. That has happened five times in the last year. Two of the five led to page edits.
The alternative is to write claims that nobody can challenge because nobody knows where they came from. That is the standard model in this industry. It is cheaper. It is also a slow corrosion of reader trust, and we are betting that trust compounds.
Where this becomes a business problem
Open citation slows us down. A product page that another brand would launch in two days takes us three to four weeks, because the claims have to be argued back to specific papers, the papers have to be re-read for any drift, and the science advisor has to sign off in writing. We ship fewer SKUs as a result. We make weaker headline claims, because the literature on most cosmetic actives is more modest than the marketing usually pretends. We give the reader more nuance than the reader necessarily asked for.
That is fine. The brand grows slower. The customers stay longer. The math has worked out in our favor every year so far.
The Tier-1 papers behind our hero claims
Two examples, anchored.
The postbiotic calming claim on the Microbiome Glow Serum is anchored to Gueniche et al., “Lactobacillus paracasei CNCM I-2116 (ST11) inhibits substance P-induced skin inflammation” (Experimental Dermatology, 2010) and follow-on work by the same lab. The mechanism is established. The dose-response curve we work from is in those papers.
The barrier-supporting peptide claim on the BioCell Renewal Cream is anchored to multiple peptide-on-collagen studies, including Schagen SK, “Topical peptide treatments with effective anti-aging results” (Cosmetics, 2017), and the broader peptide literature reviewed in the American Academy of Dermatology position papers.
If a reader writes us tomorrow saying one of those papers has been retracted or superseded, we will update the page. That is the deal.
For the editorial position behind this, see the slow skincare manifesto. The diminishing returns piece and TikTok editorial sit alongside this argument.
FAQ
Why don’t all brands do this? Cost, mostly. Open citation requires an in-house or contracted scientist, slower launch cycles, and a willingness to make weaker claims. Most brands optimize for launch velocity instead.
Are your claims weaker than competitors’ as a result? Often, yes. We will say “supported by published research” where another brand says “scientifically proven to deliver 47% improvement.” Our wording is closer to true. Theirs is closer to viral.
What if a paper we cite gets retracted? We update the page and write a note. It has not happened yet for a paper we rely on heavily, but the protocol exists.
How do I check a citation myself? PubMed for biomedical papers, Google Scholar for broader cosmetic-science. Most papers we cite have a free abstract; some have full text behind a paywall. We are happy to send PDFs to readers who write in.
Is this just marketing about being transparent? Possibly, in the sense that anything a brand says publicly is also marketing. The test is whether the policy is observable. Our citations are on every page. The science advisor is named. The email is monitored. If we ever stop doing those things, the policy is gone.
Sources
- Gueniche A, et al. “Lactobacillus paracasei CNCM I-2116 (ST11) inhibits substance P-induced skin inflammation.” Experimental Dermatology, 2010. pubmed.ncbi.nlm.nih.gov
- Schagen SK. “Topical peptide treatments with effective anti-aging results.” Cosmetics, 2017.
- American Academy of Dermatology. Position statements on cosmetic ingredient claims. aad.org
Tag hub: All skin-science coverage