Standards
Editorial policy.
The supplement industry buys plenty of coverage. We do not sell ours.
Independence
Supplement Skeptic does not accept paid reviews, sponsored placements, gifted product in exchange for coverage, or editorial input from any brand we cover. No vendor sees a review before it publishes. No vendor gets a right of reply built into the verdict. If a brand contacts us after publication with a factual correction, it goes through the same correction process as a reader email, and it has to be a fact, not a feeling.
The desk chooses what to review based on reader interest and category relevance — which products are being marketed hardest, which are ranking for high-intent searches, which readers email us about. Commission potential is not an input to that decision, and the library's verdict distribution is the proof: of 491 published reviews, 147 say Avoid and 241 say Skeptical. Not one says Recommend.
Purchase-link independence
Most purchase links on this site compensate us when a reader buys. Reviews are assigned, researched, scored, and edited independently from that compensation. Payment potential does not determine product selection, rating, or wording.
The structural safeguard is simple: our commission is identical whether a review says buy or says walk away, because the commission is set by the vendor's affiliate terms, not by our verdict. There is no version of this site where a kinder review pays better per sale. What a kinder review would buy us is more sales — and the cost of that is the only asset this desk has, which is being believed the next time we say something is fine. The affiliate disclosure shows the full money path.
Sourcing standards
A claim in a review has to be traceable to something a reader can open. In descending order of weight, we rely on:
- Randomised controlled trials in humans, ideally independent of the ingredient supplier.
- Systematic reviews and meta-analyses, including Cochrane reviews where one exists.
- Reference monographs, principally the NIH Office of Dietary Supplements fact sheets.
- The product's own label, sales page, and checkout flow, read as primary source documents.
Mechanistic hand-waving, rodent studies, in-vitro work, and "clinically studied ingredient" badges do not carry a purchase decision on their own, and we say so in the review when a sales page leans on them. Where a sales page cites a real paper, we open the paper and check three things: whether it studied the claimed outcome, whether it studied humans, and whether its dose is within reach of what the product actually delivers. That third check kills most of them.
How the rating works
Every review carries a verdict on a four-point scale — Recommend, Conditional, Skeptical, Avoid — plus a numeric rating out of ten, a short verdict summary, and explicit pros and cons. The rating is a summary of the verdict, not a separate opinion; it exists so that two products with the same verdict can still be ranked against each other. Ratings across the library currently run from 1.2 to 6.9, with a mean of 3.9. The full scoring path is set out on the methodology page.
Corrections
If we're wrong, we fix it — visibly. Corrections are appended to the top of the article with the date and a description of what changed. We do not silently edit published material to make an old take look better than it was.
Routine maintenance is different from a correction and is treated differently: prices move, refund windows change, and vendors reshuffle checkout flows, so reviews carry a last-updated date and get re-checked. A changed price is an update. A wrong dose, a misread study, or a verdict built on a bad premise is a correction, and it gets labelled as one. Send either to [email protected].
Medical review, stated plainly
No review on this site has been reviewed by a licensed clinician. Every article carries a
medicallyReviewed flag in its source data and it is false on all 491 of
them. We keep the flag rather than deleting it so that clinician sign-off, if it ever happens,
appears per-article and is checkable, instead of as a vague sitewide trust badge.
What you are reading is an evidence review written by research analysts, citing published sources. It is journalism about products, not clinical advice about you. Nothing here diagnoses, treats, or substitutes for a clinician who knows your medications and history. The sitewide medical, editorial, and affiliate disclaimer applies to every page.
Limits we disclose
- We do not lab-test products. No assays, no commissioned certificates of analysis. Manufacturing claims are reported as label claims and labelled as such.
- We do not buy and consume every product we review. A review is a documentary analysis of the formula, the evidence, the price, and the terms — not a personal trial, and never presented as one.
- We publish under a desk byline rather than named individual credentials. We would rather say that than manufacture a credentialed author.
- Absence of evidence is reported as absence of evidence, not as proof a product fails. Several reviews say exactly that: nobody has tested this, so nobody knows.
What we will not publish
- Sponsored "best of" lists dressed up as editorial
- AI-generated review copy without human evidence review
- Before/after photos supplied by the seller
- Doctor endorsements without verified credentials, including our own
- Invented testimonials, review counts, or purchase figures
- Fabricated urgency — countdown timers, fake stock warnings, or "price rises tonight" framing
- A verdict written before the evidence check that is supposed to produce it
Reader data
Editorial independence is not much use without the boring version of the same promise. We do not sell reader data, run advertising networks, or install retargeting pixels. What the site collects and who receives it is itemised on the privacy policy page.