Two directories, barely overlapping

The Menopause Society directory is the big one. ISSWSH, the sexual-health society, is smaller but mostly different people: only 180 of its 489 practitioners also appear in the Menopause Society list, which leaves 309 clinicians the first scrape never saw.

Overlap is matched on name after stripping honorifics and middle initials, because ISSWSH writes "Dr. Jane Smith" and the Menopause Society writes "Jane Smith". Matching the raw strings returns zero overlap, which is wrong. Names are not unique identifiers, so treat 180 as close, not exact.

What their websites look like technically

4,920 practitioners went through an automated website audit. 935 of them have no site of their own: either a directory or hospital profile page, or nothing findable at all. That group is the sharpest pain and the easiest sale.

The quality score averages 96.3 out of 100 across 4,535 scored sites, and 4,460 grade as "good". Read that with suspicion. It is a shallow technical check (HTTPS, an h1, a viewport tag, load speed), not a judgment about whether the site wins patients. A clinic can score 100 and still be invisible and ugly. The design scoring further down is the honest half of the picture, and the two disagree constantly.

Do they ever say the word?

Every reachable page was fetched and its full text stored, then menopause and perimenopause mentions counted in the body copy. Not inferred from the page title or the directory listing, read off the page itself.

2,689 pages are stored with 19.8M characters of text kept, so new questions about what these practices say are a database query rather than another scrape of the web. An earlier version of this page called the smaller number a coverage gap. That was wrong: the first pass deliberately looked only at solo US homepages, and it had finished. The real gap was the pages it never considered, which are now included and broken out below. The curated outreach shortlist stands at 247 and is due a refresh against this wider pool.

What is in the stored corpus

The first pass kept only a clinician's own US homepage, which is a good filter for "is her site menopause-invisible?" and a bad one for a corpus: it threw away group practices, bio pages, and everyone outside the US. Every page is now kept and labeled, so the narrow view is still one query away.

Third-party directory profiles (Healthgrades, Zocdoc and similar) are deliberately excluded. They are templated, they block automated clients most aggressively, and they say nothing about how a clinician markets herself.

The sites worth stealing from

122 menopause and women's health clinic sites scored across visual beauty, layout interest, uniqueness, color, modernity, warmth, and hero treatment. The average lands at 58.5, so the top of this list is genuinely unusual rather than merely competent. Top 24 shown.

Where this actually lives

All of it now sits under one project folder, MenopausePracticeGrowth. The scrapers and raw data are in MenopausePortalScraper/, moved out of OneDrive on 2026-07-18 because OneDrive sync corrupts open SQLite files. The design research is in MenopausePracticeSites/research/. Both stay independent git repos with their own GitHub remotes, so nothing was flattened or lost in the move.

This page carries counts and public clinic websites only. Practitioner names, phone numbers, and street addresses are deliberately not published here; they stay in the local CSVs. Regenerate this page with build_consolidated_stats.py followed by build_data_room_page.py rather than editing it by hand.