asx.com.au — AI Site Grade

Analysis

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

---

ASX.com.au serves every AI crawler a 200 with full HTML but zero structured data — a rare open-door with nothing inside the rooms

Every major AI crawler — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended — receives a 200 OK with ~163KB of server-rendered HTML from the homepage. No UA-based blocking, no Cloudflare challenge, no JS shell. The robots.txt is a single User-agent: * rule blocking only /search* paths, with zero AI-bot-specific directives. This is an unusually permissive posture for a top-10 global exchange. Yet the site squanders it.

Schema and Structured Data Vacuum

Across every page fetched — homepage, about, markets, CHESS project, FAQ, equity market prices — zero JSON-LD schema of any type was found. No Organization, WebSite, FAQPage, BreadcrumbList, NewsArticle, or FinancialProduct schema. The FAQ page at /investors/investment-tools-and-resources/faq contains 7 sections of genuine Q&A content (~6,000 words of visible text) but uses no FAQPage markup. The homepage meta description calls ASX "more than a stock market" and describes "a sustainable financial ecosystem," but no structured data reinforces that positioning for knowledge panels or AI answer surfaces.

Cold-Knowledge Gap

A frontier LLM queried cold about ASX recalls the blockchain CHESS replacement abandonment in 2022, the $1.05M ASIC fine for outage failures, and the 1987 merger origin — but the actual site makes no mention of these events. The CHESS Project page describes a "secure & contemporary replacement system using global standards" without acknowledging the abandoned blockchain effort. The site's narrative is forward-looking and aspirational; the model's prior is anchored to regulatory failures and a stalled tech project. This gap means AI engines retrieving the site will find content that contradicts or ignores the reputational context they already hold.

Content Architecture and JS-Rendering Risk

The site runs on Adobe Experience Manager (AEM) with server-side rendering. All pages return substantial visible text (~1,200-6,000 words) from a plain GET — no JS dependency for core content. However, the text is heavily diluted by repeated navigation markup: every page body begins with the same 800+ word nav tree (login state, market links, listings, investors, participants) before reaching unique content. The homepage's actual unique body content (excluding nav) is thin. The llms.txt returns a 404, meaning no AI-friendly content map exists despite the site's massive URL surface (3,447+ URLs in the sitemap).

External Signals and Infrastructure

DNS records reveal Adobe Experience Manager hosting (Adobe site verification TXT), Fastly CDN delegation, and Google Cloud DNS. Response headers show x-cdn: Fastly and x-served-by cache nodes. The site has strong security posture (HSTS, X-Frame-Options: SAMEORIGIN, X-Content-Type-Options: nosniff). External mentions are sparse in search results — no recent news articles surfaced about ASX in 2024-2025, suggesting the CHESS project and outage stories dominate the stale external narrative that AI models hold.

Findings

1. Zero JSON-LD schema across all pages High

   No Organization, WebSite, FAQPage, BreadcrumbList, or other schema markup found on any fetched page, including the FAQ page with 7 Q&A sections.

   What to change: Add JSON-LD schema for Organization, WebSite, FAQPage, BreadcrumbList, and FinancialProduct on relevant pages.

2. Cold-knowledge gap: site narrative contradicts AI model priors High

   AI models recall the abandoned blockchain CHESS replacement, ASIC fine, and 1987 merger origin, but the site omits these events, creating a mismatch that undermines AI-generated summaries.

   What to change: Add a page or section that transparently addresses the CHESS project history, including the abandoned blockchain effort and current status.

3. No llms.txt file for AI-friendly content map Medium

   The llms.txt endpoint returns 404, missing an opportunity to guide AI crawlers to key content across the site's 3,447+ URLs.

   What to change: Create an llms.txt file listing key pages (about, markets, CHESS project, FAQ) to help AI crawlers discover important content.

4. No AI-bot-specific directives in robots.txt Medium

   The robots.txt only has a single User-agent: * rule blocking /search*, with no explicit allowances or disallowances for AI crawlers like GPTBot or ClaudeBot.

   What to change: Add explicit directives for AI crawlers (e.g., Allow: / for GPTBot) to ensure continued access and signal intent.

5. Repetitive navigation markup dilutes unique content Medium

   Every page starts with an 800+ word navigation tree before unique content, reducing the signal-to-noise ratio for AI crawlers extracting page meaning.

   What to change: Consider using semantic HTML or lazy-loading navigation to reduce repeated boilerplate in the initial HTML.

6. Sparse recent external mentions reinforce stale narrative Medium

   Web searches for ASX news in 2024-2025 returned no results, meaning the dominant external narrative remains the CHESS project failure and outages from prior years.

   What to change: Proactively publish press releases or blog posts about current initiatives to generate fresh external signals.

7. FAQ page lacks FAQPage schema High

   The FAQ page contains 7 sections of genuine Q&A content (~6,000 words) but uses no FAQPage markup, missing a chance to appear in AI answer surfaces.

   What to change: Add FAQPage schema with Question/Answer markup to each Q&A section on the FAQ page.

8. Homepage unique body content is thin Low

   After removing navigation boilerplate, the homepage's unique content is minimal, reducing its value for AI crawlers seeking a summary of the organization.

   What to change: Expand the homepage with a concise, unique summary of ASX's mission, key services, and recent milestones.

What's working

• Unrestricted access for all major AI crawlers — Every major AI crawler receives a 200 OK with full server-rendered HTML, with no UA-based blocking or challenges.
• Server-side rendered HTML with substantial text — All pages return substantial visible text (1,200-6,000 words) from a plain GET, with no JavaScript dependency for core content.
• Robots.txt allows crawling of most content — The robots.txt only blocks /search* paths, allowing crawlers to access the vast majority of the site.
• Large sitemap with 80 URLs covering key pages — The sitemap at /sitemap.xml returns 80 URLs, providing good coverage of the site's content for crawlers.
• Strong security headers (HSTS, X-Frame-Options, X-Content-Type-Options) — Response headers include HSTS, X-Frame-Options: SAMEORIGIN, and X-Content-Type-Options: nosniff, indicating good security posture.
• Fastly CDN and Google Cloud DNS for performance — The site uses Fastly CDN and Google Cloud DNS, ensuring fast global delivery and reliability.