burtwatson.com — AI Site Grade

Analysis

The Entire Site Is Invisible to Every AI Crawler Except One

Burt Watson Auto Group operates a site that returns 403 Forbidden to every major AI crawler — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, OAI-SearchBot, and even standard browsers — while anthropic-ai alone receives a full 200 response with 407KB of real HTML. This creates a bizarre single-bot visibility funnel: only Claude's crawler can read the site, and only because the Akamai WAF (edgesuite.net) has an explicit allow-rule for that user-agent string.

Crawler Access

The robots.txt file is inaccessible from standard browsers (403 via Akamai) but loads for anthropic-ai. It contains rules blocking GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot from JS/CSS/JSON/API paths — though these restrictions are moot since the WAF blocks those bots entirely at the edge. The llms.txt file exists and returns 200 for anthropic-ai with 263KB of structured content listing inventory pages, but no other AI crawler can reach it. The sitemap (179KB, 200+ URLs) is similarly single-bot accessible. DNS points to Akamai IPs (2.19.248.x), and the server header AkamaiGHost confirms edge-level blocking.

Cold-Knowledge Gap

LLM knowledge of "Burt Watson" returns zero awareness of the auto group. The model describes Burt Watson as a UFC backstage coordinator famous for "It's time!" catchphrases — a completely different entity. The actual site is a multi-franchise Chevrolet/Ford/GMC dealership group with locations in Freeland, Reese, and Pinconning, Michigan. No external search results surfaced for the dealership name, no Reddit threads, no reviews, no press mentions. The brand has effectively zero off-domain footprint that AI engines can cite.

Schema Posture

The homepage HTML (fetched via anthropic-ai) contains no JSON-LD schema in the head section. The page uses ddc-site (Dealer.com platform) classes and loads inventory data via JavaScript (/static/ws/inv-facets/bundle.js, /static/ws/inv-listing/bundle.js). The <title> tag reads "Burt Watson Auto Group | Freeland, Reese & Pinconning, MI" and meta descriptions reference Chevrolet, Ford, and GMC, but structured data for AutoDealer, LocalBusiness, or Product is absent. The about-us.htm page returns a 404 — no "About" page exists for AI crawlers to learn the brand story.

External Signals

No external mentions of "Burt Watson Auto Group" were found across web search, Reddit, or review platforms. The domain's TXT records include Microsoft 365 (spf.protection.outlook.com) and a tmes tracking token, but no Google Business Profile or dealer-review citations surfaced. The only discoverable off-domain signal is the Dealer.com platform infrastructure (images.dealer.com, pictures.dealer.com). This total absence of external references means AI models have no third-party corroboration to cite when describing the dealership.

Findings

1. Akamai WAF blocks all major AI crawlers except anthropic-ai High

   The site returns 403 Forbidden to GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, OAI-SearchBot, and standard browsers. Only anthropic-ai receives a 200 response, creating single-bot visibility.

   What to change: Remove the Akamai WAF rules that block AI crawler user-agents, or add allow-rules for GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, and OAI-SearchBot.

2. robots.txt is inaccessible to standard browsers High

   The robots.txt file returns 403 Forbidden when fetched from a standard browser, though it loads for anthropic-ai. This prevents search engines from discovering crawl rules.

   What to change: Ensure robots.txt is publicly accessible (200) from all user-agents, not just anthropic-ai.

3. llms.txt is only accessible to anthropic-ai High

   The llms.txt file returns 200 for anthropic-ai with 263KB of content, but other AI crawlers cannot reach it due to WAF blocking.

   What to change: Make llms.txt publicly accessible to all AI crawlers by adjusting WAF rules.

4. No JSON-LD structured data on homepage High

   The homepage HTML contains no JSON-LD schema for AutoDealer, LocalBusiness, or Product. The site relies on JavaScript to load inventory data, which may not be rendered by all crawlers.

   What to change: Add JSON-LD structured data for AutoDealer, LocalBusiness, and individual vehicle listings to the homepage and inventory pages.

5. About Us page returns 404 Medium

   The /about-us.htm page returns a 404 error, preventing AI crawlers from learning the brand story and location details.

   What to change: Restore the About Us page with a proper 200 response and include structured data for LocalBusiness.

6. Zero external search results for the dealership High

   Multiple web searches for 'Burt Watson Auto Group' and related terms returned zero results. No Google Business Profile, reviews, or press mentions were found.

   What to change: Create and verify a Google Business Profile, encourage customer reviews, and build local citations to establish an off-domain footprint.

7. LLM knowledge confuses dealership with UFC personality High

   When queried about 'Burt Watson', LLMs describe a UFC backstage coordinator, not the auto group. The brand has no presence in AI training data.

   What to change: Publish authoritative content about the dealership (e.g., Wikipedia, local news) and ensure structured data is in place to help AI models disambiguate.

8. Sitemap only accessible to anthropic-ai High

   The sitemap.xml (179KB, 200+ URLs) is only accessible to anthropic-ai due to WAF blocking, preventing other crawlers from discovering the site structure.

   What to change: Make sitemap.xml publicly accessible to all crawlers by adjusting WAF rules.

9. No Reddit or review platform mentions Medium

   Searches on Reddit and review sites returned zero mentions of 'Burt Watson Auto Group', indicating no community or customer feedback online.

   What to change: Encourage customers to leave reviews on Google, Yelp, and DealerRater, and engage with local community forums.

10. Inventory data loaded via JavaScript Medium

    The homepage loads inventory data through JavaScript bundles (/static/ws/inv-facets/bundle.js, /static/ws/inv-listing/bundle.js), which may not be executed by all AI crawlers, potentially leaving vehicle listings invisible.

    What to change: Implement server-side rendering or include critical inventory data in the initial HTML to ensure crawlers can access it without JavaScript.

11. llms.txt not accessible to other AI bots High

    While llms.txt exists and is rich, it is only served to anthropic-ai. Other AI crawlers cannot access it, limiting the site's AI visibility.

    What to change: Allow all AI crawlers to access llms.txt by updating WAF rules.

12. No Google Business Profile or dealer citations found Medium

    No Google Business Profile or dealer-review citations were discovered, contributing to the lack of external signals.

    What to change: Create and verify a Google Business Profile and list the dealership on local directories.

What's working

• anthropic-ai crawler is explicitly allowed — The Akamai WAF has an explicit allow-rule for the anthropic-ai user-agent, enabling Claude to fully crawl the site with 200 responses on all tested pages.
• llms.txt file exists with 263KB of structured content — The site hosts an llms.txt file that lists inventory pages and other resources, providing a structured entry point for AI crawlers that can access it.
• Sitemap contains 200+ URLs — The sitemap.xml is 179KB and lists over 200 URLs, indicating a well-structured site with many pages available for crawling.
• Contact page is accessible and returns 200 — The /contact.htm page returns a 200 response for anthropic-ai, providing contact information that can be used by AI models.
• New and used inventory pages return 200 — The new-inventory and used-inventory index pages return 200 responses for anthropic-ai, allowing Claude to access vehicle listings.
• Homepage has descriptive title and meta description — The homepage title 'Burt Watson Auto Group | Freeland, Reese & Pinconning, MI' and meta description reference Chevrolet, Ford, and GMC, providing context for crawlers.
• DNS and hosting infrastructure is stable — DNS records resolve to Akamai IPs with multiple A records and TXT records for email authentication, indicating a professionally managed hosting setup.
• robots.txt includes AI crawler rules — The robots.txt file (accessible to anthropic-ai) contains directives for GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot, showing awareness of AI crawlers.