Editorial policy
How an article here gets chosen, researched, sourced, reviewed and corrected — and what this site cannot honestly claim yet.
Written by Sarah CollinsLast updated
Every article on PCOSguides is researched and written by Sarah Collins, cites at least three primary sources linked to the study itself, and carries the date it was last reviewed. No article is reviewed by a named clinician yet — until one is, each states the guideline it was checked against. Corrections are published, not quietly made.
That paragraph is the whole policy. The rest of this page is the detail, including the parts that are still a standard rather than a finished system.
The six standards, and which are live
5 of the 6 standards below are enforced today. Anything marked in progress is marked that way because it is not true yet. Stamping it enforced would be the exact failure this site exists to avoid.
Minimum three primary sources
Every article links to the study itself, never to another blog. The build fails if an article carries fewer than three.
Every protocol names who it excludes
Interactions, contraindications, and the people who should skip it. That section is mandatory, not a footnote.
No cure claims, no outcome promises
PCOS is managed, not cured. Nothing here promises weight loss, pregnancy, or a symptom outcome — and no page will ever show a before-and-after photo.
Named medical review
A registered dietitian and a clinician are being recruited to review this content. Until an article carries a named reviewer, it states which clinical guideline it was checked against.
Reviewed on a schedule
Anything carrying a dose is re-reviewed every 12 months, and immediately when a guideline changes. Every article shows its last-reviewed date.
Corrections are published
When we get something wrong we change the article and say what changed. We do not quietly edit.
How a topic gets chosen
One article answers one question. Before anything is written, the question goes on a single master keyword list alongside the 1,000 articles planned across the site's 8 sections — so nothing is commissioned twice under a different name. That planned count is the build target, not what is published today.
The list exists to prevent cannibalisation: two of our own pages competing for the same query split the signal and beat each other. So a new question earns its own URL only when the answer is genuinely different.
| The case | What happens |
|---|---|
| A question no existing article answers | New article, one target question, added to the master list first |
| A modifier that changes the recommendation — a different phenotype, a medication interaction, pregnancy | New article, because the answer is not the same answer |
| A modifier that does not change the recommendation | A section inside the article that already covers it. No new URL |
| A question an existing article half-answers | The existing article is expanded and re-reviewed. The URL does not change |
Rejected topics stay on the list, marked rejected with the reason. Keeping the rejects is what stops the same weak idea being re-proposed in six months.
How an article gets researched
Research starts at the source, not at the search results. The first pass is the literature and the guidelines; what other sites have written is read afterwards, to find what they left out. An article that only restates the first page of Google has no reason to exist.
Sources are ranked. Where two sources disagree, the higher rank wins and the disagreement is named in the article rather than hidden.
| Rank | Source type | What counts |
|---|---|---|
| 01 | Systematic reviews & meta-analyses | Cochrane, PubMed |
| 02 | Randomised controlled trials | population and n stated |
| 03 | Clinical guidelines | ESHRE, ACOG, Endocrine Society, NIH |
| 04 | Observational studies | labelled as observational |
| 05 | Expert clinical opinion | labelled as opinion |
Ranks 04 and 05 are labelled in the article as observational or as opinion, every time. An observational finding is never written as if it were a trial result.
What is never cited here
- Other blogs
- Supplement brand pages
- Social media
- AI summaries
Each of those is somebody else's summary of a study, and a summary of a summary is where doses and populations get lost. If a claim can only be traced to one of them, it does not go in the article.
What has to be true before an article publishes
Three primary sources is the floor, and it is enforced by the build rather than by memory. Every article passes its citations to the article template as structured data; the template renders them as a linked source list at the foot of the page and emits them as a machine-readable citation list in the page's structured data. An article carrying fewer than three fails the build and never reaches the site.
Four more conditions apply to every article:
- Every number is attributed. A specific figure carries the named study, its population and its year. An unsourced figure is flagged as needing a source, and that line does not publish.
- Every link goes to the study. Not to a press release about the study, and not to a blog post about the press release.
- Every protocol names who it excludes. Interactions, contraindications, and the people who should skip it — a required section, not a footnote. Supplement articles give the studied dose and the study it came from, and never rank a supplement above a prescribed medication.
- No cure claims, no outcome promises. PCOS is managed, not cured. Nothing here promises weight loss, pregnancy or a symptom outcome, and no page will ever carry a before-and-after photograph.
Who medically reviews this, and who does not
There is no medical review board. Nobody credentialed has signed off the articles on this site, and no byline here says otherwise. A registered dietitian and a clinician are being recruited; until one of them agrees in writing, the reviewer line on every article stays empty rather than getting filled with a plausible-looking name.
What happens instead is stated on the article itself. In place of a reviewer, each article names the clinical guidance it was checked against — currently the 2023 International Evidence-Based Guideline for the Assessment and Management of PCOS — so you can see the standard being applied and go and read it yourself.
Sarah is not a doctor or a dietitian. Nothing on this site is a diagnosis or a treatment plan, and every article names the point at which you should be talking to a clinician instead of reading.
When an article gets reviewed again
Anything carrying a dose is re-reviewed at least every 12 months, and immediately when a guideline it relies on changes. Every article shows its last-reviewed date at the top, so a stale page cannot hide.
| Trigger | Deadline | What you see |
|---|---|---|
| An article names a dose, a lab range or a supplement protocol | Every 12 months | Last-reviewed date moves; sources re-checked and re-linked |
| A guideline or study the article relies on is updated or retracted | Immediately | Article updated, plus a correction note if the recommendation changed |
| A reader reports an error | As soon as it is verified | See corrections, below |
| Nothing has changed | No republish | Nothing. The date is never bumped to make an article look fresh |
What happens when we get something wrong
Corrections are published on the article, not made quietly. The scale of the change decides what appears:
- A typo, a broken link, a formatting fix. Corrected without a note. Nothing about the meaning has changed.
- A number, a dose or the meaning of a claim. The article is corrected and a dated correction line is added at the foot of it, saying what it previously said and what it says now.
- A claim that can no longer be supported. The claim is removed, and the correction says it was removed and why. It is not silently deleted.
If something on this site looks wrong, say so. Emailsupport@pcosguides.com with the page address and, if you have it, the source that contradicts it. A correction backed by a better source is the single most useful thing anyone can send this site.
Who pays for this, and what that does not buy
PCOSguides is independent and published by one person. Two revenue sources are allowed here: products this site makes and sells itself, and affiliate commission on products it already recommends on the evidence. Two are not allowed: sponsored articles and paid links.
- No sponsored posts. No brand has paid for an article, a mention, a ranking or a placement, and none will. There is no rate card.
- No paid links. Links are not bought, sold, exchanged or accepted in return for coverage.
- Affiliate links are disclosed at the link. Not only in the footer — beside the link itself, where the decision is actually made. Full terms are on theaffiliate disclosure page.
- Commission never changes a recommendation. The comparison tables include products that pay nothing, and say plainly when the cheapest option is the one to buy — or when the honest answer is to buy nothing.
How to check any of this
None of the above is meant to be taken on trust. Every claim on this page is visible on the site itself:
- Open any article. The author, the review status and the last-reviewed date sit above the first paragraph, and the full source list sits at the foot.
- Click the sources. They go to the study, the guideline or the trial record.
- Read /llms.txt. It states the same review status in machine-readable form, generated from the same data this page renders from, so the two cannot drift apart.
- Read who writes this, including what she does not claim.
If you find a gap between what this page promises and what an article does, that gap is the defect. Send it to support@pcosguides.com.