Review methodology
Last updated: August 2026
Every product we review is scored on the same four dimensions, on a 1–5 scale each. We then weight them based on category — for example, third-party-testing weighs heavier for supplements than for fitness equipment. The composite score appears at the top of every review.
Efficacy
Does the product do what it claims? We grade against peer-reviewed evidence for the active ingredient(s), dose, and delivery format — not against marketing copy. Inert or under-dosed products get an honest low score.
Value
Cost per active dose, cost per serving, cost per scientifically meaningful unit. We compare to category alternatives. Premium pricing is fine if the product earns it.
Quality / label honesty
Is the formula transparent? Is it third-party tested by an independent lab (USP, NSF, Informed Sport, ConsumerLab)? Are the doses on the label what's actually in the bottle? Proprietary blends with hidden doses get downgraded.
UX
Taste, texture, smell, package, ordering experience, subscription cancellation experience. The best product on paper doesn't help if you stop taking it after three days.
What we don't score
We don't score products on brand story, founder credentials, marketing aesthetic, or influencer endorsements. We don't accept payment for higher rankings. Our affiliate relationships are disclosed; commissions never influence the score.
When a product doesn't earn a recommendation
We'd rather leave a category empty than recommend a product that doesn't meet our bar. “Best of” lists may include a “Skip” section when the category is full of low-evidence options.
Medical article review process
Product reviews use the four-dimension rubric above. Editorial health articles — the long-form pieces on medication, supplements, nutrition, sleep, and disease management — follow a separate four-step process:
Topic selection from an evidence hierarchy.
Every article starts from a defensible evidence base. We prefer, in order: systematic reviews and meta-analyses of randomized controlled trials; individual RCTs; well-designed cohort studies; then case reports and expert opinion only when higher-tier evidence doesn't exist. When we cite lower-tier evidence, we say so explicitly. Wire copy, press releases, and social-media claims never substitute for a primary source.
Automated claims-check.
Every draft runs through an automated claims-check before it enters the clinician review queue. The check flags prohibited or high-risk language (“cure,” “treat,” “prevent,” “guaranteed,” “miracle”), unapproved compounded GLP-1 references, off-label dosing claims, and missing or broken citations. Passing claims-check does not substitute for clinician review — it only clears the draft to enter the review queue.
Clinician review.
A licensed clinician then reviews the article end-to-end. Reviewers hold at least one of the following active licenses: MD (Doctor of Medicine), DO (Doctor of Osteopathic Medicine), RD (Registered Dietitian), PA-C (Certified Physician Assistant), or NP(Nurse Practitioner). The reviewer checks: medical accuracy against current clinical guidelines; safety of any dosing, timing, or usage guidance; absence of prohibited “cure / treat / prevent” language; currency of the cited research; and appropriateness of the audience framing.
Publication and annual re-review.
Clinician-reviewed articles publish with the byline “Medically reviewed by {Reviewer Name}, {Credential}”and the review date. Every reviewed article is re-reviewed at least annually — sooner if a clinical guideline changes materially or new high-quality evidence contradicts what we said. Substantive corrections are dated and appended visibly to the article. Articles awaiting clinician review are transparently labeled “pending medical review” on the page and in the RSS feed.
See our editorial standards for how sourcing, corrections, and disclosures are handled.