equip.health — AI Site Grade

Analysis

I have enough data now. Let me compile the audit.

Equip.health — AI-Visibility Audit

The homepage's single H1 heading reads "You won't believe what happens next" — a clickbait placeholder phrase that communicates nothing about eating disorder treatment to any AI crawler parsing the page structure.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended — receive a full 200 response with identical content to a browser baseline. No UA-based blocking exists. The site is hosted on Vercel (Next.js/Plasmic stack) with no WAF layer that discriminates by user-agent. The robots.txt is a bare Allow: / for all agents with zero AI-specific directives — no GPTBot restriction, no PerplexityBot crawl-delay, no Google-Extended exclusion. The llms.txt returns a 404 (serving the full JS shell of the 404 page, ~698KB). The sitemap lives at /api/sitemap (418 URLs) and is properly referenced in robots.txt.

Cold-Knowledge Gap

The LLM prior knows Equip as treating children, adolescents, and young adults (ages 6–24) — but the site now explicitly says it treats "patients of all ages" and the 2024 outcomes report covers ages 3 to 72. The prior also describes the co-founders as Kristina Saffran and Dr. Erin Parks, which is correct, but omits that Equip expanded to treat adults in September 2023 and now serves all 50 states. The site's own content about treating adults, athletes, boys and men, LGBTQIA+, and BIPOC populations is entirely absent from the cold model's knowledge. The model also does not know about the Kerry Washington investment or the 2025 outcomes report showing 5,000+ patients.

Schema Posture

Every page carries rich MedicalOrganization + Organization JSON-LD with founders, founding date (2019), location (San Diego), employee count (201-500), phone, insurance plans offered, and medical specialty. This is unusually thorough and consistent — the same block appears on every page. However, the homepage and treatment pages also inject malformed Article schema entries where the articleBody field contains raw HTML/JSX markup (e.g., <DIV class="plasmic_default__all...">), which will confuse parsers. The FAQ page has no FAQPage schema despite containing a genuine FAQ with Q&A pairs. The homepage has no LocalBusiness or HealthcareBusiness schema.

Content & Structure

The homepage delivers ~9,200 words of visible text — rich, substantive content about the 5-person care team, insurance coverage, and patient outcomes. But the heading hierarchy is broken: the sole H1 is the placeholder phrase, followed by only two H2 headings ("What Equip treatment looked like for real patients" and "Sign up for resources"). The actual value propositions ("Lasting recovery", "Covered by most insurance", "No waitlist") are rendered as unstyled text or SVG elements, not as headings. The FAQ page has no H1 at all — its first heading is an H2. The site uses Plasmic visual builder, which generates deeply nested <DIV> structures that may cause AI crawlers to miss semantic relationships.

External Signals

Equip has earned press coverage from Time, Vogue, Fast Company, Inc., Newsweek, and Today.com — all linked from the homepage. The company is backed by General Catalyst, Tiger Global, F-Prime Capital, Optum Ventures, Rock Health, and Hopelab. The press page lists 6 company announcements from 2022-2024 via GlobeNewswire. The research page shows 6 peer-reviewed publications (2023-2025) in journals including *International Journal of Eating Disorders*, *PLOS One*, and *Journal of Adolescent Health*. No negative reviews or controversies surfaced in search.

Findings

1. Homepage H1 heading is a clickbait placeholder phrase High

   The homepage's single H1 reads 'You won't believe what happens next', which communicates nothing about eating disorder treatment to AI crawlers parsing page structure.

   What to change: Replace the H1 with a descriptive heading such as 'Virtual Eating Disorder Treatment for All Ages'.

2. FAQ page lacks FAQPage schema High

   The FAQ page contains genuine Q&A pairs but has no FAQPage structured data, reducing its eligibility for rich AI-generated answers.

   What to change: Add FAQPage JSON-LD schema to the FAQ page with each question and answer.

3. Malformed Article schema with raw HTML in articleBody Medium

   Homepage and treatment pages inject Article schema entries where the articleBody field contains raw HTML/JSX markup, which will confuse parsers.

   What to change: Remove the malformed Article schema or ensure articleBody contains plain text only.

4. Broken heading hierarchy across key pages Medium

   The homepage has only one H1 (the placeholder) and two H2s; the FAQ page has no H1 at all. Value propositions are not marked as headings, reducing semantic structure for AI crawlers.

   What to change: Add descriptive H1 headings to every page and use H2-H6 to structure content hierarchically.

5. llms.txt returns 404 serving a heavy JS shell Medium

   The llms.txt endpoint returns a 404 page (~698KB JS shell) instead of a simple text file, missing an opportunity to guide AI crawlers to key content.

   What to change: Create a plain-text llms.txt file listing important URLs and a brief summary.

6. Cold LLM knowledge omits expanded adult treatment Medium

   LLM prior knowledge describes Equip as treating ages 6-24, but the site now treats all ages (3-72). The cold model also lacks awareness of adult-focused content, Kerry Washington investment, and 2025 outcomes report.

   What to change: Publish a dedicated page or press release about the adult treatment expansion and ensure it is well-linked and indexed.

7. Homepage lacks LocalBusiness or HealthcareBusiness schema Low

   Despite having MedicalOrganization schema, the homepage does not include LocalBusiness or HealthcareBusiness schema, which could improve local AI visibility.

   What to change: Add LocalBusiness or HealthcareBusiness schema with address, phone, and opening hours.

8. Deeply nested DIV structure from Plasmic builder Low

   The site uses Plasmic visual builder, generating deeply nested DIVs that may cause AI crawlers to miss semantic relationships between content sections.

   What to change: Use semantic HTML elements (section, article, nav) and reduce unnecessary nesting where possible.

What's working

• All major AI crawlers receive full access — GPTBot, ClaudeBot, PerplexityBot, and others receive a 200 response with identical content to a browser baseline. No UA-based blocking exists.
• Consistent MedicalOrganization JSON-LD on every page — Every page carries rich MedicalOrganization + Organization schema with founders, founding date, location, employee count, insurance plans, and medical specialty.
• Homepage delivers ~9,200 words of substantive content — The homepage contains rich, visible text about the 5-person care team, insurance coverage, and patient outcomes, providing ample material for AI crawlers.
• Press coverage and research publications boost authority — Equip has earned press from Time, Vogue, Fast Company, and others, plus 6 peer-reviewed publications. These external signals enhance AI trust.
• Sitemap properly referenced in robots.txt — The sitemap at /api/sitemap (418 URLs) is correctly referenced in robots.txt, aiding crawler discovery.