hbf.com.au — AI Site Grade

Analysis

HBF's homepage and all key pages carry zero JSON-LD schema — the only structured data on the entire domain lives on the 404 error page

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Applebot-Extended, Bytespider — receive a full 200 response with identical byte payload (~678KB) to a browser baseline. No UA-based blocking exists. The robots.txt uses a single User-agent: * catch-all with Allow: / and no AI-specific directives. No AI bot is mentioned by name. The llms.txt at /llms.txt returns a 404 (rendered as a Next.js error page with noindex,nofollow). The site runs on AWS CloudFront behind a Next.js frontend; the 404 page HTML reveals a Sitecore CMS backend (/sitecore path blocked in robots.txt). JS-rendering risk is low — plain GET returns ~1,000+ words of visible text on all key pages.

Cold-Knowledge Gap

The LLM knows HBF as a Western Australia-centric not-for-profit founded in 1941, serving primarily WA residents, with "Smart" product ranges and the "HBF Run for a Reason" event. The actual site tells a different story: HBF positions itself as "Australia's second largest member-based health fund" with "over one million members" nationally, serving singles, couples, families, young adults, and overseas visitors. The site never mentions "Smart" product names — the cold model's product knowledge is stale or hallucinated. The site prominently features Mozo 2025 People's Choice Award for Highly Trusted Hospital Cover and claims "lowest average premium increase of the 5 largest health funds – 2 years in a row" — neither fact appears in the model's prior knowledge. The cold model also understates HBF's national scope; the site actively targets overseas visitors (OVHC) and all Australians, not just WA.

Schema Posture

Every page fetched — homepage, about, health-insurance, hospital-cover, extras-cover, switch-to-hbf, young-adults, tax-hub, OVHC — has zero JSON-LD schema of any type. No Organization, HealthInsurancePlan, FAQPage, Product, WebSite, or BreadcrumbList. The only schema on the entire domain is an Organization block embedded in the 404 error page HTML (returned when hitting /llms.txt). That lone schema includes sameAs links, a ContactPoint with phone +61-133-423, and a logo URL. All canonical tags are missing. All meta descriptions are null. All OG tags are absent. The site has rich FAQ content (6-11 FAQ items per page with Q&A pairs) but none are marked up as FAQPage.

External Signals

The site references the Mozo 2025 People's Choice Award for Highly Trusted Hospital Cover on multiple pages (young-adults, OVHC). The tax-hub page claims HBF has the "lowest average premium increase of the 5 largest health funds – 2 years in a row". The DNS TXT records show an anthropic-domain-verification token, confirming HBF has proactively verified domain ownership with Anthropic — yet no ClaudeBot directives exist in robots.txt and no llms.txt exists. The site also has Dynatrace monitoring, Salesforce (service.force.com), Google Maps API, and Facebook/Instagram/LinkedIn/YouTube social profiles. No blog content was found in the sitemap despite the homepage linking to an "HBF Blog" section.

Surprising Contradictions

The sitemap contains URLs with the path segment ,-w-, (e.g., /corporate-health-cover/,-w-,/compare-extras-cover) which appear to be Sitecore encoding artifacts — these are indexed and crawlable but likely serve broken or unintended content. The robots.txt blocks /corporate-health-cover/* entirely, yet the sitemap index includes a corporate-health-cover/sitemap.xml sub-sitemap. The homepage <title> is simply "HBF" — no descriptive text, no brand extension, no keywords. Every page shares the same bare title tag. The Wayback Machine returned no snapshot, suggesting the site may block archiving or is relatively recent in its current Next.js form.

Findings

1. Zero JSON-LD schema on all live pages High

   Every fetched page — homepage, about, health-insurance, hospital-cover, extras-cover, switch-to-hbf, young-adults, tax-hub, OVHC — has no JSON-LD schema of any type. The only schema on the domain is an Organization block on the 404 error page.

   What to change: Add JSON-LD schema for Organization, WebSite, HealthInsurancePlan, FAQPage, BreadcrumbList, and Product to all relevant pages.

2. Cold LLM knowledge understates HBF's national scope and awards High

   The LLM knows HBF as a WA-centric not-for-profit, but the site positions itself as Australia's second largest member-based health fund with over one million members nationally, serving overseas visitors and all Australians. The site prominently features Mozo 2025 People's Choice Award and lowest premium increase claims, none of which appear in the model's prior knowledge.

   What to change: Publish an llms.txt file and add structured data (JSON-LD) that explicitly states HBF's national membership, awards, and premium claims to improve AI knowledge accuracy.

3. Rich FAQ content lacks FAQPage schema markup Medium

   Multiple pages contain 6-11 FAQ items with Q&A pairs, but none are marked up as FAQPage schema. This prevents AI crawlers from extracting structured FAQ data for rich results.

   What to change: Add FAQPage JSON-LD schema to all pages with FAQ content.

4. All pages missing canonical tags, meta descriptions, and OG tags Medium

   Every fetched page has no canonical tag, no meta description, and no Open Graph tags. This harms SEO and social sharing, and reduces clarity for AI crawlers about the preferred URL.

   What to change: Add canonical tags, meta descriptions, and Open Graph tags to all pages.

5. Homepage title is just 'HBF' with no descriptive text Medium

   The homepage <title> is simply 'HBF' — no brand extension, no keywords. All pages share the same bare title tag, missing opportunities for SEO and AI context.

   What to change: Update the homepage title to include descriptive text such as 'HBF Health Insurance | Australia's Second Largest Member-Based Health Fund'.

6. Sitemap contains Sitecore encoding artifacts in URLs Medium

   The sitemap includes URLs with path segment ',-w-,' (e.g., /corporate-health-cover/,-w-,/compare-extras-cover), which are Sitecore encoding artifacts. These are indexed and crawlable but likely serve broken or unintended content.

   What to change: Remove or fix the Sitecore encoding artifacts from the sitemap to prevent indexing of broken URLs.

7. Robots.txt blocks /corporate-health-cover/ but sitemap includes its sub-sitemap Low

   The robots.txt disallows /corporate-health-cover/*, yet the sitemap index includes a corporate-health-cover/sitemap.xml sub-sitemap. This inconsistency may confuse crawlers.

   What to change: Either remove the disallow rule for /corporate-health-cover/ or remove the sub-sitemap from the sitemap index.

8. No llms.txt file published Medium

   The /llms.txt endpoint returns a 404 error page. An llms.txt file would help AI crawlers discover key content and correct cold-knowledge gaps.

   What to change: Create and publish an llms.txt file listing important pages and a brief description of HBF's national scope and awards.

9. Blog section not included in sitemap Low

   The homepage links to an 'HBF Blog' section, but no blog URLs appear in the sitemap. This may limit discoverability of blog content.

   What to change: Include blog URLs in the sitemap to ensure they are crawled and indexed.

10. Wayback Machine has no snapshot of the site Low

    The Wayback Machine returned no snapshot for the homepage, suggesting the site may block archiving or is relatively recent in its current Next.js form.

    What to change: Ensure the site allows archiving by not blocking the Wayback Machine crawler in robots.txt or via headers.

What's working

• All major AI crawlers receive full 200 responses — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Applebot-Extended, and Bytespider all receive identical byte payload to a browser baseline with no UA-based blocking.
• Domain verified with Anthropic via DNS TXT record — The DNS TXT records include an anthropic-domain-verification token, confirming proactive domain ownership verification with Anthropic.
• Key pages contain substantial text content (1,000+ words) — Homepage and key landing pages return over 1,000 words of visible text, providing rich content for AI crawlers to index.
• Multiple pages include FAQ sections with Q&A pairs — Pages like hospital-cover, extras-cover, switch-to-hbf, young-adults, and OVHC contain 6-11 FAQ items each, providing valuable structured information that could be marked up.
• Site prominently displays Mozo 2025 People's Choice Award — The Mozo 2025 People's Choice Award for Highly Trusted Hospital Cover is mentioned on multiple pages, providing strong external validation that can be leveraged in schema.
• Active social media profiles on major platforms — The site links to Facebook, Instagram, LinkedIn, and YouTube profiles, providing external signals and brand presence.
• Simple robots.txt with catch-all allow rule — The robots.txt uses a single User-agent: * with Allow: /, ensuring all crawlers can access the entire site without complex rules.