hingehealth.com — AI Site Grade

Analysis

The robots.txt uses double-slash paths (//for/, //chat/) that no crawler interprets correctly — the disallow rules are effectively broken, and the sitemap URL also contains // syntax that may confuse parsers.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, anthropic-ai, Bytespider, Applebot-Extended, ChatGPT-User — receive a 200 with identical byte payload (215,774 bytes) as a browser. No UA-based blocking exists. The site runs on Cloudflare + AWS CloudFront with Gatsby 5.16.1 SSR. However, deep content pages (press releases, articles) return JS shells with 28–33 visible words to plain GET requests — the article body is client-rendered. GPTBot fetching the same pages also gets the shell, not the rendered article text. The robots.txt contains Disallow: //for/ and Disallow: //chat/ — the double-slash syntax is non-standard and ignored by all crawlers, so nothing is actually disallowed. No llms.txt exists (404). The sitemap-index.xml is valid and lists 1,732 URLs across one sub-sitemap.

Cold-Knowledge Gap

The cold LLM describes Hinge Health as a $6.2B unicorn (2021 valuation) with a 12-week program using a wearable motion sensor, and recalls a 2023 BIPA class-action lawsuit over biometric data collection. The actual site tells a different story: Hinge Health is now a public company (Q4 2025 earnings reported Feb 2026), has expanded far beyond MSK into pelvic health, migraine care, and fall prevention, and its core differentiator is Enso — an FDA-cleared wearable pain relief device — plus TrueMotion computer vision tracking. The cold model knows nothing about the IPO, the Enso device, the migraine program, or the public-sector expansion to 300+ organizations. The BIPA lawsuit is not mentioned anywhere on the site, creating a reputational blind spot for AI engines that may surface the lawsuit without the site's counter-narrative.

Schema Posture

The homepage carries a single Organization schema with founders, founding date, address, and social profiles — solid but missing WebSite schema, potentialAction (SearchAction), and hasPart references to sub-pages. The /for-individuals/ page has a FAQPage schema with 7 questions — a strong answer-format signal. The articles index uses ItemList with 976 entries. However, individual article pages lack Article schema — the migraine article has only VideoObject schemas for embedded exercises, no Article type, no author, no datePublished in structured data. Press release pages have zero schema at all. The case studies index has ItemList but with malformed URLs (https://www.hingehealth.com/https://www.hingehealth.com/...).

External Signals

DNS TXT records confirm Hinge Health has explicitly verified its domain with OpenAI (openai-domain-verification), Anthropic (anthropic-domain-verification), and Google (gc-ai-domain-verification) — indicating active AI ecosystem participation. The site references press coverage from Fierce Healthcare and CNBC (Fortt Knox interview with CEO Daniel Perez). The press page shows a record Q1 2026 and a migraine care program launch in April 2026 — content the cold model has no awareness of.

Findings

1. Deep content pages return empty JS shells to AI crawlers High

   Press releases and article pages deliver 28–33 visible words to plain GET requests and to GPTBot, with the article body client-rendered. This prevents AI crawlers from indexing the substantive content.

   What to change: Enable server-side rendering (SSR) or static generation for article and press release pages so that AI crawlers receive the full rendered HTML content.

2. Cold LLM knowledge omits IPO, Enso device, and migraine program High

   The cold LLM describes Hinge Health as a $6.2B unicorn with a 12-week program and a BIPA lawsuit, but knows nothing about the company's public status, the Enso FDA-cleared wearable, the migraine care program, or expansion into pelvic health and fall prevention.

   What to change: Publish an llms.txt file and ensure key pages (IPO, Enso, migraine) are fully crawlable and contain structured data to accelerate AI model updates.

3. Individual article pages lack Article schema High

   The migraine article page contains only VideoObject schemas for embedded exercises, with no Article type, author, or datePublished in structured data. Press release pages have zero schema at all.

   What to change: Add Article schema with author, datePublished, headline, and image to all article and press release pages.

4. Case studies index has malformed URLs in ItemList schema Medium

   The case studies index page uses ItemList schema but with malformed URLs that duplicate the domain (e.g., https://www.hingehealth.com/https://www.hingehealth.com/...).

   What to change: Fix the ItemList schema to use correct absolute URLs without domain duplication.

5. Homepage missing WebSite schema and SearchAction Medium

   The homepage has only Organization schema and lacks WebSite schema with potentialAction (SearchAction), which limits rich search result features like site links search box.

   What to change: Add WebSite schema with SearchAction to the homepage.

6. No llms.txt file published Medium

   The llms.txt endpoint returns a 404, missing an opportunity to guide AI crawlers to key pages and provide a structured summary for LLM context.

   What to change: Create an llms.txt file listing important pages (e.g., product, about, press releases) and a brief site summary.

7. Robots.txt uses non-standard double-slash paths Low

   The robots.txt contains Disallow: //for/ and Disallow: //chat/ with double slashes, which are ignored by all crawlers. The sitemap URL also contains // syntax that may confuse parsers.

   What to change: Correct the robots.txt paths to use single slashes (e.g., /for/, /chat/) and ensure the sitemap URL is properly formatted.

8. BIPA lawsuit surfaces in cold knowledge without site counter-narrative Medium

   The cold LLM recalls a 2023 BIPA class-action lawsuit over biometric data collection, but the site does not mention it anywhere, creating a reputational blind spot for AI-generated summaries.

   What to change: Consider publishing a statement or FAQ page addressing the lawsuit to provide context and counter-narrative for AI models.

What's working

• All major AI crawlers allowed with 200 status — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others receive a 200 response with identical content as browsers. No user-agent blocking exists.
• Domain verified with OpenAI, Anthropic, and Google for AI access — DNS TXT records show openai-domain-verification, anthropic-domain-verification, and gc-ai-domain-verification, indicating active participation in AI ecosystem programs.
• FAQPage schema on /for-individuals/ with 7 questions — The /for-individuals/ page includes a FAQPage schema with 7 questions, providing strong answer-format signals for AI crawlers.
• ItemList schema on articles index with 976 entries — The articles index page uses ItemList schema listing 976 articles, helping crawlers discover the breadth of content.
• Valid sitemap-index.xml with 1,732 URLs — The sitemap-index.xml is valid and lists 1,732 URLs across one sub-sitemap, ensuring good crawl coverage.
• Press page references Fierce Healthcare and CNBC coverage — The press page mentions coverage from Fierce Healthcare and CNBC, providing external credibility signals.