Methodology
How we grade
Every grade on Supplement Hype is set from the published human evidence — before any product is ever attached, and regardless of who pays. This is the standard the whole site rests on, so here it is in full.
1. We grade claims, not products
A supplement is not "good" or "bad." A claim is well-supported or it isn't. One product can be Strong for one use and Weak for another — creatine is excellent for strength but only Limited for cognition; omega-3 is Strong for triglycerides but Weak as "general heart health" from a tiny softgel. So we grade each claim on its own.
2. The four evidence grades
Each claim is graded on the strength and quality of human evidence — not popularity, not how good the mechanism sounds. Animal and test-tube data alone never push a claim past Limited.
Multiple good-quality human trials and meta-analyses converge. You can rely on it.
Real human evidence exists but it's mixed, smaller, or depends on dose, form, or person.
Early, small, or short human studies, or a strong mechanism with thin outcomes. Promising, unproven.
Mostly animal/lab data, marketing, or human trials that came back negative. Be suspicious.
3. The Hype Gap: two numbers, one verdict
Each entry gets two editorial scores from 0–100:
- Evidence tracks the strength of the best-supported use — anchored to the grade table above and the quality of the underlying trials and reviews.
- Hype tracks marketing intensity — search-trend and influencer volume, and how inflated the typical claims are relative to what's proven.
The gap (hype − evidence) drives a plain-language verdict. The satisfying part is the inversions: a negative gap means a supplement is actually better than its reputation.
These two scores are editorial judgements, kept deliberately public so you can argue with them. The claim grades are the rigorous core; the scores are the at-a-glance summary.
4. Evidence base: how much a grade could still move
A grade tells you how strong the evidence is today. The evidence base tells you how settled that is - whether the grade rests on a deep, consistent literature or on a couple of small, early studies. We show it on every entry so the uncertainty is visible, not buried.
Large, consistent human evidence (multiple trials/meta-analyses). This grade is unlikely to change.
Real human evidence, but mixed, smaller, or dependent on form/dose. The grade could shift as data accrues.
Thin, small, or lower-quality human studies. Treat this grade as provisional.
Very early — mostly preclinical or one or two small/industry-funded human trials. Expect this to change as research matures.
5. Which sources we trust, in order
Not all evidence is equal, and more papers is not the same as better evidence - a pile of small, industry-funded trials can mislead. So we weight sources in this order, and say so when a claim rests on weaker ground:
- Systematic reviews & meta-analyses (especially Cochrane) - the top of the hierarchy.
- Large, well-designed randomized controlled trials.
- Reputable evidence aggregators - Examine.com, the NIH Office of Dietary Supplements - to cross-check.
- Single small or industry-funded trials - used cautiously, and we flag the funding and downgrade the evidence base.
- Animal and test-tube data - never enough on its own to push a claim past Limited.
Honest limit: grades are currently compiled editorially from these sources with AI assistance, not signed off by a clinician. We're transparent about that (see About), we flag low-confidence and industry-funded claims, and clinician review of safety-critical entries is the next step on our roadmap.
6. The verification gate (the part that matters)
This is what separates Supplement Hype from the affiliate slop it competes with. No entry is allowed to go live without sources and a review date — the site literally won't build otherwise. Each entry goes through:
- Draft the entry and its claims.
- Source every claim against primary references — preferring systematic reviews and meta-analyses (e.g. Cochrane), then randomized trials, then reputable aggregators (Examine.com, the NIH Office of Dietary Supplements).
- Set a review date and reviewer. The schema blocks publish without non-empty citations and a non-null review date.
- Re-review on a schedule — sooner for trending items — and update the date.
Every dossier shows "Last reviewed: DATE" and exposes the same date and reviewer in structured data, so both Google and AI assistants can see the page is maintained and checkable.
7. How money is kept away from the grades
Grades are set before any product is attached. Affiliate links — when present — only ever point to third-party-tested products and are disclosed every time. Sponsors can pay to have a product lab-tested and listed, never to change a grade. We run no display ads. If a grade ever looked bought, the only thing this site sells — trust — would be gone. Read the conflict-of-interest policy →
None of this is medical advice. We describe what the research shows; we don't tell you what to take. Full disclaimer →