ourismancars.com — AI Site Grade

Analysis

The site is a Gatsby SPA shell served from Amazon S3 + CloudFront — every AI crawler gets a 200 with full HTML content, but the content itself is structurally hollow.

Crawler Access

All 11 tested AI bots (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User) receive a 200 status with the identical 761KB HTML payload as a browser. The robots.txt is a bare-bones User-agent: * Allow: / with zero AI-specific directives. No bot is blocked at the UA or network level. However, the site is a Gatsby 5.15.0 SPA — the 761KB includes all JS bundles and CSS, and several key pages (/ourisman-cares/, /ourisman-promise/, /meet-our-executive-team/) return only 129 words of visible text after the navigation chrome. An AI crawler that does not execute JavaScript will find those pages effectively empty.

Cold-Knowledge Gap

The LLM prior knows Ourisman as a 100-year-old family-run group founded in 1921 serving MD/VA/DC with brands including Mercedes-Benz and BMW — but the actual site carries neither Mercedes nor BMW. The site lists 18 dealerships across Chevrolet, Ford, Honda, Kia, Hyundai, VW, Mazda, Subaru, Genesis, Lotus, Chrysler-Dodge-Jeep-Ram-Fiat, and Toyota. The cold model also has no awareness of the Ourisman Promise loyalty program (the site's flagship differentiator) or the Lotus of Washington DC franchise — a notable gap since Lotus is a high-signal, AI-interest brand.

Schema Posture

Zero JSON-LD schema exists on any page examined — not on the homepage, not on the about page, not on vehicle detail pages, not on dealer pages. The homepage has no Organization, AutoDealer, LocalBusiness, or Product schema. Vehicle detail pages (e.g., the 2025 Ford Escape VDP) have no Vehicle or Car schema despite containing structured data like VIN, mileage, MSRP, and trim. The blog page has no BlogPosting schema. This is a complete schema vacuum across a site with 4,000+ indexed URLs.

Content Rot and Hollow Pages

The blog contains exactly one post from November 10, 2021 titled "The all new blog post is the best yet!" — a placeholder that was never replaced. The /ourisman-cares/ community outreach page and /ourisman-promise/ loyalty page are content shells (129 words each, just navigation and footer). The /meet-our-executive-team/ page is similarly empty. The sitemap contains 4,171 URLs including stale test pages (/about-us-test/, /about-second-test/, /test/) and an /ourisman-towson-md-html/ page that appears to be a draft. External search returns zero results for "Ourisman Cars reviews" or "Ourisman Automotive Group reviews" on DuckDuckGo — the brand has essentially no discoverable off-domain reputation footprint in current search indices.

Findings

1. Key pages render as content shells with under 130 words of visible text High

   Pages like /ourisman-cares/, /ourisman-promise/, and /meet-our-executive-team/ return only 129 words of visible text after navigation chrome, making them effectively empty for AI crawlers that do not execute JavaScript.

   What to change: Add substantive content to these pages — at least 300 words of unique, descriptive text about the program or team.

2. Zero JSON-LD schema across all examined pages High

   No Organization, AutoDealer, LocalBusiness, Product, Vehicle, or BlogPosting schema exists on the homepage, about page, vehicle detail pages, dealer pages, or blog. This is a complete schema vacuum across a site with 4,000+ indexed URLs.

   What to change: Implement JSON-LD structured data: Organization and AutoDealer on the homepage, Vehicle on each VDP, LocalBusiness on dealer pages, and BlogPosting on the blog.

3. Blog contains only one placeholder post from November 2021 High

   The blog has a single post titled 'The all new blog post is the best yet!' dated November 10, 2021 — a placeholder that was never replaced. No other content exists.

   What to change: Remove the placeholder post and either launch a real blog with regular, relevant content or remove the blog section entirely.

4. No discoverable off-domain reputation in search indices Medium

   External searches for 'Ourisman Cars reviews' and 'Ourisman Automotive Group reviews' return zero results on DuckDuckGo, indicating the brand has no visible review or discussion footprint outside its own domain.

   What to change: Encourage customers to leave reviews on major platforms (Google, Yelp, DealerRater) and engage in community discussions to build an off-domain presence.

5. Sitemap contains stale test and draft pages Medium

   The sitemap includes URLs like /about-us-test/, /about-second-test/, /test/, and /ourisman-towson-md-html/ that appear to be drafts or test pages, polluting the crawl budget and index.

   What to change: Remove stale test and draft pages from the sitemap and either delete them or add noindex tags.

6. LLM cold knowledge lists brands not on the site Medium

   The LLM prior associates Ourisman with Mercedes-Benz and BMW, but the actual site does not carry those brands. The cold model also lacks awareness of the Ourisman Promise loyalty program and the Lotus of Washington DC franchise.

   What to change: Ensure the site prominently and accurately lists all dealership brands and key differentiators (like the Ourisman Promise) in visible, crawlable text and structured data.

7. No /llms.txt file published Low

   The site does not serve an /llms.txt file, missing an opportunity to guide AI crawlers to key content and provide a structured overview.

   What to change: Create an /llms.txt file listing important pages (dealers, about, promise, cares) and a brief site description.

8. Robots.txt lacks AI-specific directives Low

   The robots.txt is a minimal 'User-agent: * Allow: /' with no AI bot-specific rules, missing the chance to prioritize or restrict certain crawlers.

   What to change: Add explicit directives for AI bots (e.g., GPTBot, ClaudeBot) to allow or disallow specific paths as needed.

What's working

• All 11 tested AI bots receive 200 status with full HTML — Every major AI crawler (GPTBot, ClaudeBot, PerplexityBot, etc.) is allowed access and receives the same 761KB HTML payload as a browser, with no UA or IP blocking.
• Site served via AWS CloudFront CDN for fast global delivery — The site uses Amazon CloudFront, ensuring low-latency content delivery worldwide, which benefits both human users and AI crawlers.
• Sitemap with 4,171 URLs is accessible and well-structured — The sitemap at /sitemap-pages.xml lists 80+ pages and is properly served, helping crawlers discover the site's URL inventory.
• Vehicle detail pages contain structured data like VIN, mileage, and MSRP in visible text — Pages like the 2025 Ford Escape VDP include key vehicle attributes (VIN, mileage, MSRP, trim) in plain text, which is crawlable even without JavaScript execution.
• About page provides 770 words of company history and context — The /about-us/ page contains 770 words detailing the company's history, family ownership, and values, offering substantive content for AI crawlers.
• Lotus of Washington DC dealer page exists with 201 words of content — The dealer page for Lotus of Washington DC provides 201 words of text, including location and brand details, which is crawlable and informative.