This page describes how a review or comparison on this site is actually produced. It’s deliberately specific: since we don’t have clinical credentials to point to, the only reasonable basis for trusting us is being able to inspect our process and check it against what we publish.
If you find a page that doesn’t follow what’s described here, that’s a defect. Please tell us.
The short version
- We record what the manufacturer states, as fact about the label — not as fact about the world.
- We look up what published research says about the key ingredients at the stated doses.
- We rate how strong that evidence is, using a fixed four-level scale.
- We score the product against criteria we publish alongside the score.
- We state the limitations, including what we couldn’t check.
- We verify that any commercial link still goes where we said it goes.
- Nothing publishes until the required pieces above are present.
How we rate evidence
Every review carries an evidence classification for the claims it discusses. This describes the state of the research, not our confidence in the product:
| Rating | What it means |
|---|---|
| Strong | Multiple well-conducted human trials, generally consistent results, adequate sample sizes, ideally with meta-analysis or systematic review support. |
| Moderate | Human trials exist and lean in a consistent direction, but are limited by size, duration, funding source, or replication. |
| Limited | Sparse human data. May rest on small or short trials, or on mechanistic and animal research that hasn’t been confirmed in people. |
| Anecdotal | Testimonials, tradition of use, or marketing claims, without controlled human evidence. |
Two things this scale deliberately does not do. It doesn’t tell you a product will work for you — individual response varies, and evidence about an ingredient is not a prediction about a person. And it doesn’t rate the product; it rates the state of research on what’s in it. A well-made supplement can carry a “Limited” rating simply because its ingredient hasn’t been studied much.
How we score products
Where a review carries a numeric score, the criteria and their weights are published with it. We don’t use a hidden formula, and we don’t use a single global scale across unrelated categories.
Scores reflect things we can actually check — label transparency, dose relative to what’s been studied, availability of third-party testing documentation, value at the current price — rather than a judgment about efficacy we’re not positioned to make.
Every scoring rubric is versioned. When we change how we score, existing reviews keep the version they were scored under until they’re re-reviewed, and the version is shown on the page. This is so a score from six months ago means what it meant six months ago.
What “verified” means here
We use the word narrowly.
Verified means we located documentation from a party other than the seller — a certificate of analysis, a third-party testing organization, a regulatory filing — and we cite it.
Unverified means the manufacturer states it and we found no independent confirmation. Most label claims sit here. Unverified is not an accusation; it’s the normal state of most supplement information, and we mark it rather than let it read as fact.
We do not commission laboratory testing. If we ever do, we’ll say so explicitly on the pages it applies to.
Sources we use, and what we avoid
We prefer peer-reviewed human research, systematic reviews, and regulatory or standards-body documentation. Where a manufacturer is the only available source for a factual detail — a dose, a serving size — we use it and attribute it to them.
We avoid citing other affiliate sites, content farms, and press releases as evidence. Where research was funded by a party with a commercial interest in the result, we note that.
Every review lists its references. If a claim on our site has no source behind it, treat that as a defect and tell us.
Keeping pages current
Supplements change: formulas get reformulated, prices move, offers expire, and research accumulates.
Each review records when it was last fact-checked and when its content was last updated, and both are shown on the page. Commercial links are re-checked on a schedule; when a link’s destination can no longer be verified, the link stops being displayed rather than sending you somewhere unverified. You’ll see an editorial note instead of a broken or stale commercial link.
We’d rather show you nothing than show you something we can’t stand behind.
What we can’t tell you
Being direct about the boundaries, because they’re substantial:
- We can’t tell you whether a supplement will work for you. Nobody honest can, from a webpage.
- We can’t confirm what’s actually in the bottle. We don’t test. Independent testing organizations have repeatedly found discrepancies between supplement labels and contents; that risk is real and applies to products we review.
- We can’t account for your situation. Interactions with medication, existing conditions, pregnancy, age, and dosage all matter and are outside what a review can address.
- We have no clinical expertise. See About. We read the research; we’re not qualified to practice medicine, and we don’t.
Independence from our commercial arrangements
We earn commissions on some links. To keep that from steering conclusions, our publishing system separates the two: commercial eligibility of an offer is tracked independently from editorial approval of the content, and a review can’t be published on the strength of a commercial relationship. A product with no affiliate offer at all can be reviewed and recommended; a product with a lucrative one can be rated poorly.
The mechanics are described in our Editorial Policy, and the commercial relationships themselves in our Affiliate Disclosure.
Methodology version 1.0. When this page changes materially, we note the date and what changed.