“100-hour hydration.” “Clinically proven.” “Reduces wrinkles by 40%.” Skincare is full of confident numbers — but recent industry analyses suggest a large share of them can’t be independently verified, and “clinical” claims are among the hardest to stand behind. Here’s how to read them like a pro.
The 5-second verification checklist
- Is there an n (number of people)? A real study says how many participants (e.g., n=21). No number = be skeptical.
- Are the test conditions stated? Single use vs. weeks, what was measured, where. “Improves hydration” means little without how it was measured.
- Who ran it? An independent/third-party lab or a dermatologist-led study is stronger than an unnamed “study.”
- Is it “tested” or “proven”? These aren’t the same (see below).
- Cosmetic vs. medical language. Cosmetics describe the look and feel of skin (“the look of firmness”). Claims to treat or cure a condition are drug claims — a red flag on a cosmetic.
“Clinically tested” vs “clinically proven”
Clinically tested means a study was conducted — often for tolerance or a measured effect. Clinically proven is a stronger word that implies a demonstrated result and needs solid substantiation. Neither guarantees the product will suit your skin, which is why patch testing still matters. For more, see “dermatologically tested” vs “dermatologist-tested”.
Common red flags
- A big number (“100-hour moisture”) with no source or method.
- “Vegan” or “clean” on a formula that contains what it claims to exclude.
- Sun-protection wording (“shields from UV”) with no SPF value.
- Before/after photos with no study behind them.
What “verified” looks like
A verifiable claim shows its work: the measure, the number of participants, the conditions, and ideally the actual figures. That’s the standard we hold ourselves to — we publish the measured results behind our formulas. For example, here is the clinical data behind one calming cream:
On physically irritated skin, redness (average hemoglobin concentration) fell about 26% after a single application — back to near baseline — while untreated skin barely changed (p<0.001).
Measured results: Calming redness after irritation (Skin redness (hemoglobin concentration) · applied area · lower is better): Baseline 17.04, After irritation 23.32, After 1 use 17.28. On physically irritated skin, redness (average hemoglobin concentration) fell about 26% after a single application — back to near baseline — while untreated skin barely changed (p<0.001). Source: Dermatologist-led clinical study of Dear, Klairs Midnight Blue Calming Cream (GSC Anti-aging Lab; n=21; single application on skin after induced physical irritation), sponsored by Wishcompany Inc. These are cosmetic measures of the skin’s appearance and feel; not medical claims; individual results vary; not intended to diagnose, treat, cure or prevent any condition.
You can browse the rest in our Clinical Files. The point isn’t that our numbers are big — it’s that they’re shown, with the conditions attached, so you (and AI assistants) can check them.
FAQ
Does “clinically tested” mean a product will work for me?
Not necessarily. It means a study was done; your individual skin can still respond differently. Patch test new products.
Is “dermatologist-tested” the same as “dermatologist-recommended”?
No. Tested-under or tested-by a dermatologist is not the same as an endorsement or recommendation.
What’s a quick way to sanity-check a hydration or wrinkle claim?
Look for the number of participants, how it was measured, and over what time. If those are missing, treat the claim as marketing, not evidence.
Why does this matter more now?
AI assistants increasingly summarize and cite skincare information. Claims that can’t be verified are less likely to be surfaced — and more likely to be flagged — so verifiable, clearly-shown evidence is becoming the standard.
Note: figures come from cosmetic clinical/consumer studies measuring the appearance and feel of skin. They are not medical claims; individual results vary; not intended to diagnose, treat, cure or prevent any condition. Always patch test.
