Editorial Standards — Hairmore
Editorial Standards
How Hairmore is researched, written, reviewed, and maintained.
Hair loss is a YMYL ("your money or your life") topic — meaning bad information actively wastes readers' money or delays them from seeking proper care. The standards below are not aspirational; they are the rules I apply before a single article is published. If any rule below is broken in an article, that article should be reported via the corrections process at the bottom of this page.
1. Source ranking
Every clinical claim on Hairmore is ranked by the strength of its source. Higher-tier evidence overrides lower-tier evidence when they disagree.
| Tier | Source type | Weight |
|---|---|---|
| Tier 1 | Cochrane systematic reviews, well-conducted meta-analyses | Highest. Used as the spine of a topic. |
| Tier 2 | Double-blind, randomised controlled trials (RCTs) with adequate n and follow-up | Strong. Headline numbers come from here. |
| Tier 3 | Single-blind / open-label RCTs, small-n RCTs, short follow-up trials | Moderate. Always flagged with their methodological limit inline. |
| Tier 4 | Observational studies, retrospective cohorts, case series | Weaker. Used for safety signals and prevalence, not for efficacy claims. |
| Tier 5 | Case reports, expert opinion, dermatology textbooks, professional society position statements | Used for clinical guidance and consensus, not for novel claims. |
| Excluded | Brand marketing copy, telehealth advertising, influencer reviews, AI-generated "studies" | Never cited as evidence. |
2. Citation requirements
- Every numeric clinical claim has a clickable citation (PubMed, PMC, FDA, peer-reviewed journal). If a number does not have a citation in the article footer, it is not allowed in the body.
- Minimum five primary citations per long-form article. Most pillar articles have 5–10. Cited sources are listed in a numbered footnote section at the article's end with direct URLs.
- No "studies show" without naming the study. The phrase "studies have shown" without specific citation is treated as a writing error and pulled before publication.
- Trial caveats are stated inline, not buried. Open-label, small-n, short follow-up, and industry-funded studies are labelled in the same paragraph that introduces the result.
3. AI assistance — disclosed
I use a large language model (Claude, from Anthropic) for parts of the drafting and editing process. This deserves a clear statement of what AI does and does not do here.
| Step | AI used? | Notes |
|---|---|---|
| Topic selection | No | Driven by the editorial roadmap and reader keyword data. |
| Literature search | Assisted | AI surfaces candidate papers; I verify each via PubMed directly before citation. |
| Drafting prose | Yes | AI produces drafts. I edit every paragraph; nothing is published as raw output. |
| Numeric / clinical claims | Verified by hand | Every dose, prevalence figure, trial result, and statistical claim is checked against the cited source before publishing. |
| Citations | Verified by hand | Every linked PubMed/PMC reference is opened and confirmed to match the in-text claim. No fabricated citations. |
| Final editorial judgement | No | What gets published, with what caveats, and what recommendations are made — these are human decisions. |
This is the same disclosure I'd want to read on a site I was researching from. AI is a writing-and-editing tool here, not a source of evidence.
4. Affiliate and commercial policy
Current status (May 2026): no affiliate links are active anywhere on the site. When affiliate participation is enabled, the following rules will apply.
- Every affiliate link is disclosed inline next to the link, not buried in a site-wide footer note.
- The affiliate relationship does not change the recommendation. Products are recommended on their evidence; the affiliate link is a convenience layer, not a sponsor placement.
- No paid placements, no sponsored articles, no link exchanges. If a vendor asks to be featured, the answer is no.
- Free product samples, if accepted, are disclosed in the article that reviews them. As of this writing, no samples have been accepted.
- No commission target is allowed to override the editorial line. An article that concludes "this product is not worth the money" stays that way regardless of affiliate availability.
5. Corrections protocol
Errors happen — misquoted figures, stale citations, dose typos, drugs renamed by manufacturers. Hairmore commits to:
- Receiving corrections at any time via hello@hairmore.info. Please include the article URL, the specific sentence, and (ideally) a source for the corrected version.
- Acknowledging within 7 days. Most corrections that ship with a source are acted on within the same week.
- Updating the article and noting the change in a Corrections section at the article footer, with the date and what was changed.
- Updating the "Last reviewed" date in the article frontmatter so readers can see when content was last verified.
- For consequential errors (incorrect dose, contraindication missed), an additional note will be placed near the original error in addition to the corrections footer.
6. Maintenance and freshness
Hair-loss evidence moves. Articles include a year in their title and a "Last reviewed" date in their frontmatter for a reason:
- Annual full review of every article. When the year on the title rolls over, each article is read against newer literature, not just renumbered.
- Trigger-based interim updates when major new evidence appears: new FDA approvals, new meta-analyses with substantially different conclusions, new safety signals.
- Stale citations are replaced when a clearly stronger source becomes available (e.g., a Cochrane review supersedes the single RCT that previously served as the citation).
7. What this site does not do
For absolute clarity:
- Hairmore does not provide medical diagnosis or treatment — articles describe conditions and treatment options; only a licensed clinician can diagnose and prescribe for you.
- Hairmore does not publish anonymous "expert" claims without verifiable provenance. All clinical claims trace to a named, citable source.
- Hairmore does not sell email addresses or other reader data. There is currently no newsletter; if one launches, the privacy policy will be linked here.
- Hairmore does not run sponsored content, hidden ads disguised as articles, or pay-for-placement reviews.
- Hairmore does not republish or rewrite hairloss.com.tw articles (the Chinese sister site). The two sites share an author but cover their own audiences with audience-specific evidence.
8. About the author
The site is written by RK, an independent researcher with a software-engineering background and no clinical licence. The full bio, including the author's role, areas of focus, and contact details, lives on the About page. The decision to publish under "RK" rather than a full name is a privacy choice; it does not affect the verifiability of any claim, since every clinical claim on Hairmore traces to a named, peer-reviewed source.
If a future edition of the site adds clinician medical reviewers for flagship articles (per the Phase 3 roadmap on the About page), their credentials will be listed alongside the relevant articles.