barrett-jackson.com — AI Site Grade

Analysis

Barrett-Jackson.com AI-Visibility Audit

Barrett-Jackson's site is functionally invisible to every major AI crawler — Cloudflare returns HTTP 402 or 403 to GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and OAI-SearchBot, while serving a full 1.1 MB page to browsers. The brand that pioneered "no reserve" collector car auctions has a de facto "no access" policy for the AI engines that increasingly shape how collectors discover auction houses.

Crawler Access

The robots.txt contains a contradiction: User-agent: * with Disallow: / blocks all crawlers, but a subsequent User-agent: Googlebot with Disallow: (empty) re-allows Googlebot specifically. No AI-specific user-agents are mentioned anywhere in the file. The compare_bot_access test confirms the Cloudflare WAF enforces the block: GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, ChatGPT-User, Bytespider, and Applebot-Extended all receive HTTP 402 with a 55-byte JSON denial. Google-Extended and anthropic-ai receive HTTP 403 with a Cloudflare challenge page. Only a standard browser UA gets past Cloudflare to the Next.js server. The llms.txt returns a 404 Next.js error shell. The sitemap-index.xml exists and lists ~60,000 vehicle URLs, but none of these are accessible to AI crawlers.

Content & Schema

The homepage and all subpages (company history, founding families, philanthropy, event pages) return rich, well-written text content to browsers — the company history page alone has 1,841 words. Vehicle detail pages (e.g., a 2001 Ford Mustang Bullitt Edition) contain 673 words with full specs, descriptions, and VIN data. However, zero pages contain any JSON-LD structured data — no Organization, Event, Product, Auction, or Vehicle schema types were detected on any page fetched. The homepage has no meta description, no OG tags, and no canonical tag. The results page and media/articles page are effectively empty shells (164 and 169 words respectively, mostly navigation boilerplate), suggesting client-side rendering for dynamic content.

Cold-Knowledge Gap

The LLM knows Barrett-Jackson's core positioning well: founded 1971 in Scottsdale, "no reserve" pioneer, flagship Scottsdale auction televised on Velocity/Discovery, charity partnerships, and even recent criticism about high buyer/seller fees. This prior knowledge is richer than what the site itself communicates to AI crawlers — the model can describe the brand's reputation and business model, but it learned that from external sources (news, Wikipedia, forums), not from barrett-jackson.com. The site's own content about its $171+ million raised for charity, its four auction locations, and its founding families' stories is completely inaccessible to the AI engines that would cite it.

External Signals

The site runs on Cloudflare (A records point to 104.20.37.171 and 172.66.146.237) with Next.js server-side rendering. The showroom subdomain (showroom.barrett-jackson.com) hosts vehicle inventory on a separate platform with no AI-crawler testing performed. DNS records show integrations with Mandrill, Mailgun, Zendesk, Litmos, and Azure — a complex tech stack. The sitemap-index.xml references four sub-sitemaps totaling over 60,000 URLs, including vehicle detail pages from past auctions dating back to 2021, but these are all behind the Cloudflare WAF for non-browser UAs.

Findings

1. Cloudflare WAF blocks all major AI crawlers with HTTP 402/403 High

   GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, ChatGPT-User, Bytespider, and Applebot-Extended receive HTTP 402; Google-Extended and anthropic-ai receive HTTP 403. Only standard browser UAs reach the server.

   What to change: Update Cloudflare WAF rules to allow AI crawler user-agents (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, etc.) to access the site, or serve them a static HTML version.

2. Robots.txt contains contradictory rules blocking all crawlers High

   The robots.txt has 'User-agent: *' with 'Disallow: /' blocking all crawlers, but then 'User-agent: Googlebot' with empty Disallow re-allows Googlebot. No AI-specific user-agents are mentioned.

   What to change: Remove the global 'Disallow: /' rule and explicitly allow AI crawlers, or use a more permissive default.

3. llms.txt returns 404 error page Medium

   The llms.txt file is missing, returning a 404 Next.js error shell. This prevents AI crawlers from discovering the site's content inventory.

   What to change: Create an llms.txt file listing key pages and summaries for AI crawlers.

4. Zero JSON-LD structured data on any page High

   No Organization, Event, Product, Auction, or Vehicle schema types were detected on any fetched page, including homepage, vehicle detail pages, and event pages.

   What to change: Add JSON-LD structured data for Organization, Event, Product, and Vehicle schemas on relevant pages.

5. Homepage lacks meta description, OG tags, and canonical tag Medium

   The homepage has no meta description, no Open Graph tags, and no canonical URL, reducing its visibility in search and social previews.

   What to change: Add a meta description, OG tags, and a canonical URL to the homepage.

6. Results and articles pages are empty JS shells High

   The /results and /media/articles pages return only 164 and 169 words respectively, mostly navigation boilerplate, indicating client-side rendering that AI crawlers cannot execute.

   What to change: Implement server-side rendering or static generation for these pages to ensure content is available to crawlers.

7. Sitemap with 60,000+ vehicle URLs is inaccessible to AI crawlers High

   The sitemap-index.xml exists and lists ~60,000 vehicle URLs, but all are behind the Cloudflare WAF that blocks AI crawlers.

   What to change: Ensure AI crawlers can access the sitemap and vehicle pages by adjusting Cloudflare rules.

8. LLM knowledge of Barrett-Jackson is richer than site content accessible to AI Medium

   The LLM knows about Barrett-Jackson's history, 'no reserve' model, charity work, and fee criticism from external sources, but the site's own content on these topics is blocked from AI crawlers.

   What to change: Allow AI crawlers to access the site so that its rich content can be cited directly.

What's working

• Key pages contain rich, well-written text content — Pages like company history (1,841 words), founding families (2,019 words), and vehicle details (673 words) provide substantial information to browsers.
• Sitemap index exists with 60,000+ vehicle URLs — The sitemap-index.xml is accessible and lists four sub-sitemaps covering vehicle detail pages, providing a comprehensive content inventory.
• Vehicle detail pages contain full specs and descriptions — Vehicle pages include VIN, year, make, model, and detailed descriptions, which are valuable for AI understanding if accessible.
• LLM has strong prior knowledge of Barrett-Jackson's brand and history — The LLM correctly identifies Barrett-Jackson as a 'no reserve' auction pioneer, founded in 1971, with televised events and charity partnerships.