1. Product claims are treated as leads, not evidence.
Brand pages, ads, retailer pages, reviews, and trend signals can help identify what to investigate. They cannot prove efficacy.
Methodology
How CosmeticFountain classifies products
Our goal is simple: identify which cosmetic products and beauty devices have support from peer-reviewed and validated science, and where evidence is missing or overstated.
2. Sources must pass verification.
We prioritize peer-reviewed human data, DOI/PMID/registry identifiers, source freshness, retraction checks, and direct relevance to the product or device parameters.
3. Expert routing is domain-specific.
Dermatology, epidemiology, biostatistics, chemistry, toxicology, device engineering, photobiology, claims, and legal experts review the claims relevant to their scope.
4. Evidence and safety are separate axes.
Evidence strength grades the claim. Safety status flags context, adverse potential, and regulatory classification risk. Strong evidence is not the same thing as safe for everyone.
5. Grades are conservative.
A product can be popular and still receive a weak evidence grade. Ingredient-level evidence is not treated as finished-product proof.
6. Human review is mandatory before public claims.
Automated workflows draft evidence packets, but sensitive claims, strong verdicts, devices, and safety concerns require human review before publication.