bhhsgeorgia.com — AI Site Grade

Analysis

Cloudflare Blocks Every AI Crawler and Human Browser Alike

The entire bhhsgeorgia.com domain returns HTTP 403 from Cloudflare for every user-agent tested — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, Applebot-Extended, Bytespider, and even a standard desktop browser. No bot or human can reach a single page of real content. The robots.txt and llms.txt endpoints also return the same Cloudflare block page. The site has effectively zero AI-visibility posture: no crawlable content exists for any language model to index.

Cold-Knowledge Gap

The LLM model queried knows the brand as "Berkshire Hathaway HomeServices Georgia Properties" — a full-service residential brokerage in metro Atlanta, part of the Berkshire Hathaway HomeServices network, offering luxury listings, relocation services, and a proprietary digital marketing platform. This knowledge likely comes from the parent brand's reputation and offline signals, not from the site itself. The site's actual homepage (visible only via Wayback Machine) uses the tagline "Always a Good Move" and promotes services including mortgage, title, insurance, relocation, and a "QuickBuy" program — details the model did not mention. The gap between what the model knows (generic brand affiliation) and what the site actually offers (a full-service ecosystem with six distinct service lines) is wide.

Schema and Content Posture

The Wayback snapshot of the homepage reveals only a single WebSite schema with a SearchAction target — no RealEstateAgent, LocalBusiness, Organization, or Product schema types that would help AI engines understand the brokerage's services, office locations, or agent roster. The homepage has a clear heading structure (H1: "Always a Good Move." followed by H2s for services, market reports, and buyer/seller guides) and includes definition patterns and a table, but none of this is machine-readable to crawlers blocked by Cloudflare. Subpages like /homes-for-sale and /roster/offices are JavaScript-rendered even in archived snapshots, producing near-zero text extraction.

External Signals Absence

Web searches for the brand name, domain, and related terms returned zero results across multiple queries. No external reviews, Reddit threads, press mentions, or third-party citations were discoverable through the search tool used. The domain's DNS records show Microsoft 365 email, multiple Google site verification tokens, and integrations with SendGrid and Mailchimp — indicating active business operations — but none of these generate discoverable external signals that AI engines could use to supplement the site's blocked content.

Findings

1. Cloudflare blocks every AI crawler and human browser High

   The entire bhhsgeorgia.com domain returns HTTP 403 from Cloudflare for all tested user-agents, including GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and standard desktop browsers. No bot or human can access real content.

   What to change: Configure Cloudflare to allow AI crawlers (e.g., GPTBot, ClaudeBot) and search engine bots while maintaining security for other traffic.

2. Robots.txt and llms.txt endpoints return 403 High

   The robots.txt and llms.txt files are inaccessible, returning the same Cloudflare block page. This prevents crawlers from learning allowed paths and AI-specific directives.

   What to change: Serve a valid robots.txt that allows AI crawlers and create an llms.txt file with site overview and key URLs.

3. High

   No crawlable content exists for any language model to index. The site's homepage and subpages are either blocked or JavaScript-rendered, yielding near-zero text extraction.

   What to change: Ensure server-side rendering for key pages and allow crawlers to access them.

4. LLM knowledge of the brand is generic and outdated Medium

   The LLM knows the brand as a residential brokerage but lacks details about the site's actual offerings, including mortgage, title, insurance, relocation, and QuickBuy program. This gap exists because the site's content is inaccessible.

   What to change: Allow crawlers to index the site so LLMs can learn accurate, detailed information.

5. Homepage lacks relevant schema types Medium

   The Wayback snapshot shows only a WebSite schema with SearchAction. Missing RealEstateAgent, LocalBusiness, Organization, and Product schemas that would help AI engines understand services, offices, and agents.

   What to change: Add RealEstateAgent, LocalBusiness, and Organization schemas with office locations, agent rosters, and service descriptions.

6. Subpages are JavaScript-rendered with minimal text Medium

   Archived snapshots of /homes-for-sale and /roster/offices yield only 34 words each, indicating heavy JavaScript rendering that prevents content extraction even when accessible.

   What to change: Implement server-side rendering or static HTML for key pages to ensure content is crawlable.

7. No external signals found in web searches Medium

   Multiple web searches for the brand name, domain, and related terms returned zero results. No reviews, press mentions, or third-party citations are discoverable, limiting AI engines' ability to supplement blocked content.

   What to change: Build external signals through press releases, business listings, and social media presence.

8. No llms.txt file for AI guidance Low

   The llms.txt endpoint returns 403, meaning no AI-specific guidance file exists. This file could help AI engines understand the site's structure and key pages.

   What to change: Create an llms.txt file with a brief site overview and links to important pages.

What's working

• Brand known to LLMs via parent network — The LLM correctly identifies Berkshire Hathaway HomeServices Georgia Properties as a full-service residential brokerage in metro Atlanta, indicating strong offline brand recognition.
• Active business operations indicated by DNS records — DNS records show Microsoft 365 email, Google site verification, SendGrid, and Mailchimp integrations, confirming the business is operational and using standard tools.
• Wayback Machine snapshot preserves homepage content — A recent Wayback snapshot captures the homepage with heading structure, service descriptions, and a table, providing a fallback for content recovery.