samesauto.com — AI Site Grade

Analysis

Sames Auto Group is Texas' oldest dealership (founded 1910) and a fifth-generation family business, yet its AI crawler posture is a study in contradictions — excellent bot access paired with a JS-rendered shell that delivers zero content to AI on key pages.

Crawler Access

The robots.txt is unusually sophisticated for a regional auto group. Every major AI crawler — GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, anthropic-ai, PerplexityBot, Perplexity-User, Google-Extended, Applebot-Extended, Bytespider — is explicitly allowed with Allow: / and only narrow disallows for /login, /print, /scheduletestdrive?*, and /reserve-your-vehicle/. The compare_bot_access test on the homepage confirmed every single AI bot receives a 200 with full content (217KB, identical to browser baseline). No UA-based blocking, no Cloudflare challenge, no WAF gate. The site runs on Fastly CDN behind Google (Kestrel server, varnish cache) with HSTS enabled. This is best-in-class AI accessibility for the homepage.

The JS-Rendering Trap

The /about page returns title and meta tags but zero visible text to any crawler — including GPTBot and ClaudeBot — because the body content is rendered entirely by client-side JavaScript. The page is 205KB of HTML but the extracted text is empty. The /about-us page (a different URL) does render 532 words of rich history, but /about and /community-programs are JS shells. The sitemap lists both /about and /about-us as separate URLs — a canonical duplication that confuses crawlers. The homepage itself renders 388 words of text, but the "Our Mission" section is duplicated verbatim (appears twice in the extracted text), wasting the limited token budget AI models allocate.

Cold-Knowledge Gap

The LLM knows Sames Auto Group as a "family-owned" Laredo dealer group with "mixed" reviews and "some complaints about sales pressure." The actual site tells a far more compelling story: founded in 1910, Texas' oldest dealership, fifth-generation leadership under Evelyn Sames (the company's first female leader), 13 dealerships across 5 cities, 11 franchise brands, and a $2M+ scholarship program (Sames Scholars at TAMIU) with 95 graduates since 2002. None of this differentiation appears in the model's cold knowledge. The site's most AI-valuable pages — /our-story (407 words of founding narrative), /about-us (532 words of detailed history), and /community-programs (547 words of philanthropic programs) — are not surfaced in any llms.txt file (which 404s/redirects to homepage).

Schema Posture

Every page carries AutoDealer JSON-LD with PostalAddress, GeoCoordinates, OpeningHoursSpecification, and priceRange. However, the telephone field is empty on every page ("telephone": ""), and both ContactPoint entries have "telephone": "+1 " with no actual number. The sameAs array is empty — no social media profiles linked despite the group operating Facebook, Instagram, and YouTube presences. The inventory pages embed rich Vehicle schema with VIN, price, model, brand, and Offers — this is strong and well-structured, but the dealer-level schema is missing critical contact data.

External Signals

The group operates a fragmented domain strategy: individual dealership sites at samesford.com, sameshonda.com, sameskia.com, samesmazda.com, samesbastropford.com, etc. — each with separate schema and content. Web search returned zero indexed results for "Sames Auto Group Laredo reviews" or "Evelyn Sames fifth generation," suggesting the brand's external citation footprint is thin relative to its 115-year history. The sameAs schema gap means AI models cannot reliably connect the parent group to its individual dealership domains.

Findings

1. About page renders as empty JS shell to AI crawlers High

   The /about page returns 205KB of HTML but zero visible text to GPTBot and ClaudeBot because content is rendered client-side. AI crawlers see only title and meta tags.

   What to change: Implement server-side rendering or pre-rendering for the /about page so that AI crawlers receive the full text content.

2. Community programs page is a JS shell High

   The /community-programs page renders 547 words of content to a browser but delivers zero visible text to AI crawlers due to client-side rendering.

   What to change: Implement server-side rendering or pre-rendering for /community-programs.

3. Duplicate about URLs confuse crawlers Medium

   The sitemap lists both /about and /about-us as separate URLs. /about is a JS shell, while /about-us has rich content. This duplication wastes crawl budget and creates confusion.

   What to change: Consolidate /about and /about-us into a single canonical URL with server-rendered content, and redirect or remove the duplicate.

4. Homepage mission section duplicated verbatim Low

   The homepage's 'Our Mission' section appears twice in the extracted text, wasting AI token budget and reducing content efficiency.

   What to change: Remove the duplicate mission text from the homepage.

5. No llms.txt file published Medium

   The llms.txt endpoint returns a 200 with 2454 bytes but the content is not a standard llms.txt file; it likely redirects to the homepage. This misses an opportunity to guide AI crawlers to key pages.

   What to change: Publish a proper llms.txt file that lists the most important pages: /our-story, /about-us, /community-programs, and inventory pages.

6. AutoDealer schema has empty telephone field Medium

   Every page carries AutoDealer JSON-LD with an empty telephone field ("telephone": "") and ContactPoint entries with incomplete phone numbers. This reduces schema completeness for AI models.

   What to change: Populate the telephone field in AutoDealer schema with the correct phone number for each location.

7. sameAs array empty in schema Medium

   The AutoDealer JSON-LD has an empty sameAs array despite the group having active Facebook, Instagram, and YouTube presences. This prevents AI models from linking the brand to its social profiles.

   What to change: Add the correct social media URLs to the sameAs array in the JSON-LD schema.

8. Brand's 115-year history absent from LLM knowledge High

   LLMs know Sames Auto Group only as a 'family-owned' dealer with 'mixed' reviews. The site's compelling story (founded 1910, Texas' oldest, fifth-generation, $2M+ scholarship program) is not surfaced in AI training data or via structured signals.

   What to change: Publish a llms.txt file, add more structured data (e.g., Organization schema with founding date and awards), and ensure key pages are server-rendered and indexed.

9. Low external citation footprint Medium

   Web searches for 'Sames Auto Group Laredo reviews' and 'Evelyn Sames fifth generation' returned zero indexed results, indicating thin external signals despite the brand's long history.

   What to change: Encourage customer reviews on third-party sites and publish press releases or articles about the brand's history and community programs.

10. Fragmented domain strategy dilutes brand authority Low

    The group operates separate domains for individual dealerships (samesford.com, sameshonda.com, etc.) with separate schema, making it harder for AI models to connect them to the parent brand.

    What to change: Consider consolidating dealership sites under the main domain or using cross-domain schema linking (e.g., sameAs) to connect them.

What's working

• Robots.txt allows all major AI crawlers — The robots.txt explicitly allows GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others with Allow: /, and only narrow disallows for login/print pages.
• Homepage returns full content to all AI bots — The compare_bot_access test confirmed all 11 AI bots receive a 200 with full content (217KB) on the homepage, identical to browser baseline.
• Inventory pages embed detailed Vehicle schema — Inventory pages include Vehicle JSON-LD with VIN, price, model, brand, and Offers, providing rich structured data for AI models.
• AutoDealer schema with location data on every page — Every page carries AutoDealer JSON-LD with PostalAddress, GeoCoordinates, and OpeningHoursSpecification, providing consistent location data.
• Fastly CDN with HSTS ensures fast, secure delivery — The site uses Fastly CDN behind Google with Kestrel server and HSTS, providing fast and secure content delivery to all crawlers.
• Key pages contain rich historical and community content — Pages like /our-story (407 words), /about-us (532 words), and /community-programs (547 words) contain compelling narratives about the brand's history and philanthropy.
• Sitemap contains 80 URLs and is properly indexed — The sitemap at sitemap.xml returns 80 URLs and is an index sitemap, helping crawlers discover pages.
• Locations page renders 207 words of text — The /locations page returns 207 words of text, providing address and contact information for dealership locations.