Last reviewed August 29, 2026 · Reviewed against the ten standards below.
Every factual claim on The Attractiveness Report is backed by a named, dated citation, labeled as a heuristic or contested claim, or removed before publication. A named human reviews each claim before it ships and at least every 90 days.
What Is The Attractiveness Report's Editorial Policy?
The Attractiveness Report's editorial policy states that every published claim is either cited to a named source, labeled as a heuristic or contested claim, or deleted before publication, with no fourth, unlabeled state permitted.
The rule applies across the entire site, not only to the methodology page, though that page shows it working most visibly: every accuracy claim there is graded by the same tier system defined below. Content across the site, this page included, is drafted with AI assistance, and a person checks every citation before it ships. When a page states that photos never leave your browser during a free scan, that claim is graded and checked the same way a cited study is, never exempted because it describes the product instead of the research. A claim that clears none of the three tiers below does not publish in softened form; it is deleted. The accuracy of the measurement itself is not governed here; it is documented separately, on the methodology page.
Our Ten Editorial Standards
The three-state rule above (cited, labeled, or deleted) governs individual claims. The ten standards below are the fuller checklist every page on this site, including this one, is held to before and after it publishes.
How Do You Decide What Counts as Strong Evidence?
Evidence on this site is graded into three tiers: randomized trials and meta-analyses are stated as fact, single studies are attributed as "research finds," and industry heuristics carry a fixed downgrade label such as "contested in the literature."
The breakdown below states the language rule attached to each tier, not just its name.
- Tier 1 evidence: randomized controlled trials, meta-analyses, and systematic reviews, licensed for definitive claim language such as "X improves Y."
- Tier 2 evidence: single observational or rating studies, published only with attributed language such as "research finds" or "in a study of n = 200."
- Tier 3 evidence: styling conventions, industry heuristics, and this site's own tool data, published only beside one of three fixed labels: "styling heuristic, not clinical science," "our tool's output, not a clinical measurement," or "contested in the literature."
Every Tier 1 or Tier 2 citation on the site follows the same named-study format: author and year, the journal, the sample size, the quantified finding, and a one-line limitation. That rule is stated here so any reader can check it against any citation on the site, not just described as a general principle. The same tier system grades the What the Evidence Says section inside the attractiveness test's own evidence section, the identical rubric applied to every scored component, not a lighter version reserved for prose. A claim that clears none of the three tiers does not publish in a softened form; it is deleted before the page ships, the same rule stated above. The scan itself runs on the same standard applied to how we measure your face, not a separate one reserved for the words around it.
Who Reviews the Content on This Site?
A named, credentialed reviewer checks every science-facing page on this site before it publishes and again at least once every quarter, and a person checks every citation even when a draft starts with AI assistance.
Every science-facing page on this site is reviewed before it publishes and again at least once each quarter, checking the citation behind each claim and whether the underlying research has shifted since the last pass. Who reviews this site, and how is set out on the about page, not repeated here. The review credit and its last-reviewed date appear only on the pages that actually went through that pass, never stamped sitewide by default. Every page on the site, this one included, starts as an AI-assisted draft, and a person checks every citation and every quantified claim against its primary source before publication. That is a standing fact about how the site is built, not a defense offered after a complaint. Science-facing pages, meaning any page stating a research finding or a quantified claim, get the full quarterly pass; pages that are purely stylistic carry only the Tier 3 labeling requirement from the section above.
The rubric exists, and someone applies it. The last honest question is whether either one survives contact with time and with the mistakes every editorial process eventually makes.
How Often Is This Site Updated?
The Attractiveness Report reviews every evidence-bearing page, including this one, on a 90-day cycle, and a page's listed update date changes only when its content substantively changes.
This page follows the identical 90-day review cycle it requires of every other evidence-bearing page on the site, including every tool page and every guide that cites research. The policy does not exempt itself from its own rule. A page's listed update date changes only when something inside it substantively changes: a citation gets replaced, a tier label gets corrected, or a reviewer credit gets added. It never changes as a routine freshness signal with no real edit behind it. An unscheduled review also runs immediately on any change to the reviewer, the fixed Tier 3 label set, or the corrections mechanism described next, rather than waiting for the next quarterly pass. That discipline is what keeps a last-reviewed date meaningful instead of decorative.
How Do You Correct Mistakes?
Yes: every claim on this site lives in a maintained claims register that tracks its source and tier, so a correction or retraction updates every page that cited it in a single pass instead of a manual search.
The claims register records four things for every cited fact on the site: the claim, its source, its evidence tier, and every page currently citing it. When a study is corrected, retracted, or superseded, the register shows every page that needs the same fix in a single pass, instead of a manual search across the site that risks missing a page. A tier downgrade works the same way in reverse: when new research moves a claim from Tier 2 to Tier 3, every page carrying that claim picks up the correct label at the same time, not on a rolling schedule. A reader who spots a suspected error reports it through the site's contact page, the same channel as any other message, not a separate error-reporting form. A second inbox is one more place a report can go unread. A correction, once it ships, carries the same last-reviewed date logic described above, never backdated to look older than the fix actually is.
This same policy governs every claim behind your Attractiveness Report, the free score and the paid one alike: cited, labeled, or deleted, and checked by a named human before it publishes. Neither the free scan nor the report gets a softer standard than the other; the same rubric grades both, without exception.