How we rate evidence

We want it to be easy to tell the difference between a product with substantial human research behind its intended use and one supported mainly by early or indirect evidence.

Strong evidence

Comparatively consistent human research supports the underlying ingredient, intervention or training method for the use described. This does not mean every formulation or individual result will be identical.

Moderate evidence

Human evidence is reasonably supportive or promising, but may be smaller, less consistent, or more dependent on dose and formulation.

Limited evidence

There is scientific rationale or early research, but not enough reliable human evidence for strong effectiveness claims.

Mixed evidence

Different ingredients or intended uses within the product have different levels of support. The product page explains these separately where possible.

What our ratings do not mean

An evidence rating is not a guarantee that a product will work for every person. It also does not replace medical advice. We distinguish evidence for an ingredient or intervention from evidence for the exact commercial product whenever that distinction matters.

We aim to use human studies and higher-quality evidence where available, and to state uncertainty rather than hide it.