hummingbird.healthcare — AI Site Grade

Analysis

Hummingbird Healthcare has registered domain verification tokens with both OpenAI and Anthropic, yet Cloudflare rate-limits GPTBot and ClaudeBot to 429 while allowing ChatGPT-User, OAI-SearchBot, anthropic-ai, and PerplexityBot through — a fragmented bot policy that contradicts the DNS-level opt-in signals.

Crawler Access

The robots.txt is a bare Yoast default with a single User-agent: * allow-all rule and no AI-bot-specific directives. DNS TXT records include openai-domain-verification and anthropic-domain-verification, signaling explicit intent to be crawled by both companies. Yet compare_bot_access on the homepage reveals GPTBot and ClaudeBot both receive Cloudflare 429 (rate-limited), while ChatGPT-User, OAI-SearchBot, anthropic-ai, PerplexityBot, Perplexity-User, Google-Extended, and Applebot-Extended all get 200 with full content. The /llms.txt returns 404. The /blog/ page returns only 5 words of visible text ("Hummingbird Healthcare News Load More") via plain GET — a JS-rendering dependency that means AI crawlers that do not execute JavaScript see an empty shell. The site runs on WP Engine behind Cloudflare with nginx, and security headers (HSTS, CSP, X-Frame-Options) are entirely absent.

Cold-Knowledge Gap

The LLM's prior knowledge describes Hummingbird Healthcare as founded in 2019 by former athenahealth executives, offering a "Patient Access Platform" with AI/automation, and appearing on the Inc. 5000 list. The actual site says the company was established in 2022, lists CEO Jeremy Schwach (no athenahealth background mentioned), and makes no mention of an Inc. 5000 ranking. The site describes itself as a managed services company (not a platform vendor) offering "Patient Access as a Service" and "Epic as a Service" — a business-process outsourcing model, not a software product. The cold model has the founding date wrong by 3 years, the product category wrong (platform vs. managed services), and cites a growth-list credential the site never claims.

Schema Posture

Every page carries the same Yoast-generated WebSite + Organization + WebPage schema with SearchAction. The Organization block includes a logo and a single sameAs link to LinkedIn. No LocalBusiness, Service, FAQPage, Product, or MedicalBusiness schema exists anywhere. The services page describes 10+ specific offerings (Primary Care Scheduling, Referral Management, Nurse Triage, Epic upgrades, etc.) with zero structured markup — a significant gap for AI answer engines trying to map capabilities. No CaseStudy schema on the Cone Health results post, which contains concrete metrics (80% of calls answered in 30 seconds, 40% increase in new patient scheduling).

External Signals

The site references UCHealth as both investor and strategic client, Cone Health as a client with published results, and Henry Ford Health as a partner. The Series B ($20M, Feb 2025, led by Health Velocity Capital) and Series A ($10M, June 2024, led by UCHealth) are documented on the blog but have zero external press pickup found via search. No Reddit threads, Glassdoor reviews, or third-party articles surfaced. The "Good Jobs" philosophy is a distinctive differentiator (Good Jobs Institute membership, MIT/Zeynep Ton framework) that the cold LLM did not mention at all.

Findings

1. GPTBot and ClaudeBot receive Cloudflare 429 rate limits despite DNS opt-in High

   DNS TXT records include openai-domain-verification and anthropic-domain-verification, signaling intent to be crawled. Yet GPTBot and ClaudeBot both receive Cloudflare 429 (rate-limited) on the homepage, while ChatGPT-User, OAI-SearchBot, anthropic-ai, and PerplexityBot get 200 with full content.

   What to change: Remove Cloudflare rate limits for GPTBot and ClaudeBot, or ensure consistent access for all AI crawlers that have DNS verification tokens.

2. Blog page renders as empty JS shell for non-JavaScript crawlers High

   The /blog/ page returns only 5 words of visible text ('Hummingbird Healthcare News Load More') via plain GET, indicating a JavaScript-rendering dependency. AI crawlers that do not execute JavaScript see an empty shell.

   What to change: Implement server-side rendering or static generation for the blog page so that content is available in the initial HTML response.

3. No /llms.txt file published Medium

   The /llms.txt endpoint returns a 404. This file is used by AI crawlers to discover content and context, and its absence limits AI visibility.

   What to change: Create an /llms.txt file that lists key pages and a brief description of the site's content for AI crawlers.

4. Cold LLM knowledge has incorrect founding date and product category High

   The LLM prior knowledge states Hummingbird Healthcare was founded in 2019 and offers a 'Patient Access Platform,' but the site says it was established in 2022 and describes itself as a managed services company (Patient Access as a Service, Epic as a Service). This mismatch can cause AI-generated answers to fabricate facts.

   What to change: Ensure the site's About page and structured data clearly state the founding year and business model to correct the LLM's prior.

5. Services page lacks structured markup for 10+ offerings High

   The /our-services/ page describes multiple specific services (Primary Care Scheduling, Referral Management, Nurse Triage, Epic upgrades, etc.) but has no Service schema or other structured data. This limits AI answer engines from mapping capabilities.

   What to change: Add Service schema markup for each offering, including name, description, and relevant properties.

6. Case study post lacks CaseStudy schema despite concrete metrics Medium

   The Cone Health results post contains specific metrics (80% of calls answered in 30 seconds, 40% increase in new patient scheduling) but has no CaseStudy or Article structured data. This reduces the chance of being featured in AI-generated answers.

   What to change: Add CaseStudy schema with the key metrics and results to the Cone Health post.

7. Funding announcements have zero external press pickup Medium

   Web searches for 'Hummingbird Healthcare funding' and related terms returned no results. The Series A and Series B announcements exist only on the company blog, limiting external signals and credibility for AI systems.

   What to change: Proactively pitch funding news to healthcare and business media to generate third-party coverage.

8. Security headers (HSTS, CSP, X-Frame-Options) are absent Low

   The site does not set HSTS, Content-Security-Policy, or X-Frame-Options headers. While not directly impacting AI visibility, this can affect trust signals for some crawlers and users.

   What to change: Add HSTS, CSP, and X-Frame-Options headers to improve security posture.

9. No Reddit or Glassdoor presence found Low

   Searches for 'Hummingbird Healthcare' on Reddit and general web returned no results. This limits social proof and external signals that AI systems may use for credibility assessment.

   What to change: Encourage employee and customer reviews on platforms like Glassdoor and Reddit to build external signals.

What's working

• DNS verification tokens for OpenAI and Anthropic are registered — The domain has openai-domain-verification and anthropic-domain-verification TXT records, signaling explicit intent to be crawled by both companies. This is a strong opt-in signal.
• robots.txt allows all crawlers with no AI-bot restrictions — The robots.txt has a single User-agent: * allow-all rule, meaning no AI crawlers are explicitly blocked at the robots.txt level.
• Sitemap is present and contains 33 URLs — The sitemap is accessible and lists 33 URLs, helping crawlers discover content.
• Every page has consistent WebSite, Organization, and WebPage schema — Yoast-generated schema provides a baseline structured data foundation across all pages, including logo and LinkedIn sameAs link.
• Blog posts contain substantive content with metrics and details — Individual blog posts (e.g., Cone Health results, funding announcements) have 600-1100 words of detailed content, providing rich material for AI crawlers that can render JavaScript.
• Good Jobs philosophy is a distinctive differentiator documented on site — The 'Good Jobs at Hummingbird Healthcare' page explains the company's commitment to the Good Jobs Institute framework, a unique angle that could differentiate the brand in AI-generated summaries.
• Key client partnerships (UCHealth, Cone Health, Henry Ford Health) are documented — The site includes case studies and partnership announcements with notable healthcare systems, providing credibility signals for AI systems.