bcautoplex.com — AI Site Grade

Analysis

I have all the data I need. Let me compile the audit.

The site is entirely invisible to browsers and most AI crawlers — but selectively visible to three OpenAI and Perplexity user-agent bots — while the domain name "bcautoplex.com" bears no relation to the actual business name "Barnes Crossing Auto Group," creating a cold-knowledge vacuum where LLMs know nothing about the brand.

Crawler Access

Cloudflare blocks every browser request and most AI crawlers (GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, Bytespider, anthropic-ai, PerplexityBot) with a 403 "Attention Required" page. Only three user-agents pass through to real content: ChatGPT-User, OAI-SearchBot, and Perplexity-User — all returning 200 with 190KB+ of HTML. This means OpenAI's search and ChatGPT can index the site, but Google's own Google-Extended crawler cannot. The robots.txt and llms.txt are both inaccessible (403), so no crawler directives exist anywhere. The sitemap is also blocked.

Content & Schema

The site is a multi-franchise auto group operating as Barnes Crossing Auto Group with dealerships across Mississippi, Tennessee, Alabama, Indiana, Louisiana, and North Carolina. Brands include Chevrolet, Ford, GMC, Hyundai, Kia, Mazda, Mitsubishi, Toyota, and Volkswagen. The homepage contains zero JSON-LD schema — no AutoDealer, Organization, or LocalBusiness markup. The Wayback snapshot (Dec 2024) confirms the same absence. The page is a JS-heavy dealer platform (DealerEProcess) that renders a large vehicle model grid, location list, and service scheduling links, but the live HTML served to bots is a 190KB shell with minimal visible text before JavaScript execution.

Cold-Knowledge Gap

A frontier LLM queried cold about "BC Autoplex" reports no knowledge of the brand — not even that it exists. The model guessed it might be a local used car dealership in British Columbia. The actual business is a major multi-state auto group in the southeastern US with 15+ dealership locations. The domain bcautoplex.com gives no clue to the real business name, and the site has zero external search presence — DuckDuckGo returns no results for the domain, the business name, or any combination. This is a complete knowledge vacuum.

External Signals

The domain has no indexed external mentions, reviews, Reddit threads, or press coverage discoverable via web search. The DNS TXT records include Facebook domain verification and Google site verification, indicating the business has claimed these profiles, but they do not surface in search. The site runs on Cloudflare (NS: georgia.ns.cloudflare.com / phil.ns.cloudflare.com) with Google Workspace mail (MX: aspmx.l.google.com). The Cloudflare block page itself references dealereprocess.org — the dealer platform vendor — rather than the actual business, further confusing any crawler that hits the 403 wall.

Findings

1. Cloudflare blocks all browser requests and most AI crawlers with 403 High

   Every browser request and most AI crawlers (GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, Bytespider, anthropic-ai, PerplexityBot) receive a 403 "Attention Required" page. Only ChatGPT-User, OAI-SearchBot, and Perplexity-User pass through.

   What to change: Configure Cloudflare to allow Google-Extended, GPTBot, ClaudeBot, and other major AI crawlers through to the real content, or serve a static HTML version to blocked user-agents.

2. robots.txt and llms.txt are inaccessible (403) High

   Both robots.txt and llms.txt return 403 errors, meaning no crawler directives are available. The sitemap is also blocked.

   What to change: Ensure robots.txt and llms.txt are publicly accessible and include directives for AI crawlers. Submit a sitemap to Google Search Console.

3. Homepage has zero JSON-LD schema markup High

   The homepage contains no AutoDealer, Organization, or LocalBusiness schema. The Wayback snapshot confirms the same absence in December 2024.

   What to change: Add JSON-LD structured data for AutoDealer, Organization, LocalBusiness, and Vehicle for each inventory item.

4. Content is JS-heavy and renders as a shell to bots High

   The live HTML served to bots is a 190KB shell with minimal visible text before JavaScript execution. The actual vehicle inventory and location data require JavaScript to render.

   What to change: Implement server-side rendering or pre-rendering for critical content so that crawlers can index vehicle inventory and location data without JavaScript.

5. Domain name "bcautoplex.com" bears no relation to actual business name High

   The domain gives no clue to the real business name "Barnes Crossing Auto Group," causing LLMs queried about "BC Autoplex" to report no knowledge of the brand.

   What to change: Consider using a domain that includes the business name, or add prominent branding and schema that associates the domain with the actual business name.

6. Zero external search presence for domain and business name High

   DuckDuckGo returns no results for the domain, the business name "Barnes Crossing Auto Group," or any combination. No reviews, Reddit threads, or press coverage are discoverable.

   What to change: Build external signals by claiming and optimizing Google Business Profile, Yelp, and other directories; encourage customer reviews; and publish press releases or blog content.

7. Cloudflare block page references dealer platform vendor, not the business Medium

   The 403 block page references "dealereprocess.org" rather than the actual business, further confusing any crawler that hits the wall.

   What to change: Customize the Cloudflare block page to display the business name and a brief message, or remove the block for legitimate crawlers.

8. Sitemap is blocked (403) Medium

   The sitemap returns a 403 error, preventing crawlers from discovering the site's URL structure.

   What to change: Ensure the sitemap is publicly accessible and submitted to Google Search Console.

9. Inventory pages lack vehicle schema markup Medium

   The new and used inventory pages (accessible to ChatGPT-User) contain large HTML but no Vehicle or Offer schema, reducing the chance of rich results in AI responses.

   What to change: Add JSON-LD structured data for each vehicle, including Vehicle, Offer, and Dealer markup.

10. No llms.txt or AI-specific directives Low

    The site does not provide an llms.txt file, missing an opportunity to guide AI crawlers to key pages and provide context.

    What to change: Create an llms.txt file listing important pages (inventory, locations, about) and a brief description of the business.

What's working

• Selective access granted to ChatGPT-User, OAI-SearchBot, and Perplexity-User — Three important AI user-agents (ChatGPT-User, OAI-SearchBot, Perplexity-User) can access the full site content, returning 200 status with large HTML payloads.
• Allowed bots receive large HTML payloads with inventory data — The inventory pages served to ChatGPT-User contain 1.7MB and 2.4MB of HTML, indicating rich vehicle data is present in the source.
• Google Workspace and domain verification records present — DNS TXT records include Google site verification and Facebook domain verification, indicating the business has claimed these profiles.
• Wayback Machine snapshot exists from December 2024 — A snapshot from December 2024 is available, providing a historical record of the site's content.