sewell.com — AI Site Grade

Analysis

Sewell.com is a 403 wall — zero AI crawlers can reach any content.

Total AI Crawler Blackout

Every AI bot tested — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User — receives a 403 Forbidden from Cloudflare on the homepage. The browser baseline also gets 403. The site is not selectively blocking bots; it is blocking all HTTP traffic at the edge. No robots.txt is served (403), no llms.txt (403), no sitemap.xml (403). The only accessible version of the site lives in the Wayback Machine, where the last snapshot (December 2025) shows a functional AutoDealer schema-tagged page with 20 Texas locations, 14+ luxury brands, and a "Sewell Difference" service narrative.

Cold-Knowledge Gap

An LLM queried cold knows Sewell as a family-owned Texas luxury dealership group founded in 1911, famous for Carl Sewell's book *Customers for Life*, transparent/no-haggle pricing, and J.D. Power recognition. It also recalls a 2023 class-action lawsuit over alleged deceptive pricing. The live site — if it could be read — would confirm the brand heritage and "Obsessed with Service" tagline, but the model's knowledge of the lawsuit and the book comes from off-domain sources, not from sewell.com itself. The site's own content (archived) makes no mention of the lawsuit, the book, or transparent pricing. The model knows more about Sewell's reputation than Sewell's own website communicates.

Schema and Content Posture

The archived homepage and new-vehicles page carry AutoDealer and Organization JSON-LD schema with priceRange, name, and url — but address fields are empty strings and sameAs is an empty array. No social profiles, no physical address, no telephone number in the schema. The BreadcrumbList schema on subpages is well-formed. The site uses DealerInspire as its CMS platform (visible in image URLs). The homepage text is thin (~125 words visible in the archive), relying on a JS-driven vehicle search widget that would not render for AI crawlers even if the 403 were lifted.

External Signal Desert

Sewell's off-domain footprint is surprisingly sparse in search results. No recent press, no Reddit threads, no review-site citations surfaced. The class-action lawsuit and the *Customers for Life* book — both known to the LLM — appear to live entirely outside the indexed web's accessible surface. This means AI engines describing Sewell are drawing on training data alone, not live retrieval from any discoverable source. The gap between what the model knows (detailed, nuanced) and what the web publicly attests (near-zero) is extreme.

Structural Fragility

The site runs on Cloudflare with aggressive WAF rules (403 on every path, every UA). DNS points to a single IP (74.119.99.3) on SystemDNS nameservers. The Wayback Machine has captured the homepage and locations page but many interior paths (/about/, /service-parts/, /the-sewell-difference/) were never archived — they return 404 from the archive, suggesting those pages either did not exist or were also blocked from capture. The site has no discoverable blog, no FAQ pages, no comparison tables, no answer-format content that AI engines could extract for featured snippets or knowledge panels.

Findings

1. All AI crawlers blocked by Cloudflare 403 High

   Every tested AI bot receives a 403 Forbidden from Cloudflare on the homepage. The site blocks all HTTP traffic at the edge, including browsers. No robots.txt, llms.txt, or sitemap.xml is served.

   What to change: Allow AI crawler user agents through Cloudflare WAF rules and serve a robots.txt that permits GPTBot, ClaudeBot, and other major AI bots.

2. Robots.txt returns 403 High

   The robots.txt file is not accessible, returning a 403 Forbidden. This prevents crawlers from understanding which paths are allowed or disallowed.

   What to change: Serve a valid robots.txt at the root that allows AI crawlers and references a sitemap.

3. No llms.txt file Medium

   The llms.txt file returns a 403, meaning AI engines cannot discover a curated list of important pages or content summaries.

   What to change: Create an llms.txt file that lists key pages like new-vehicles, locations, and the Sewell Difference.

4. Sitemap.xml returns 403 High

   The sitemap.xml is blocked, preventing search engines and AI crawlers from discovering the site's URL structure.

   What to change: Serve a valid sitemap.xml that lists all important pages and submit it to search engines.

5. Homepage has only ~125 words of visible text High

   The archived homepage contains approximately 125 words of visible text, relying heavily on a JavaScript-driven vehicle search widget that AI crawlers cannot execute.

   What to change: Add substantial static text content to the homepage describing the brand, services, and value proposition.

6. JSON-LD schema has empty address and sameAs fields Medium

   The AutoDealer and Organization schema on the homepage and new-vehicles page contain empty strings for address fields and an empty array for sameAs, missing physical location and social media links.

   What to change: Populate all address fields (streetAddress, addressLocality, addressRegion, postalCode) and add social media URLs to sameAs.

7. Site content omits known brand narratives Medium

   The archived site does not mention the class-action lawsuit or Carl Sewell's book 'Customers for Life', which are known to LLMs from training data. This creates a gap between external reputation and on-site messaging.

   What to change: Add a page or section addressing the lawsuit and highlighting the book and transparent pricing philosophy.

8. Near-zero indexed web presence for Sewell High

   Multiple web searches for Sewell, its dealerships, and related terms returned zero results. No recent press, reviews, or social media citations were found in search engine indexes.

   What to change: Build an off-domain presence through press releases, guest articles, social media activity, and listings on review sites to create indexed references.

9. Key interior pages not archived by Wayback Machine Medium

   Pages like /about/, /service-parts/, and /the-sewell-difference/ return 404 from the Wayback Machine, suggesting they either never existed or were blocked from capture.

   What to change: Create and serve these pages with static content, and ensure they are accessible to crawlers.

10. No blog, FAQ, or answer-format content Medium

    The site lacks discoverable blog posts, FAQ pages, comparison tables, or other answer-format content that AI engines extract for featured snippets or knowledge panels.

    What to change: Add a blog with articles about vehicle comparisons, maintenance tips, and brand stories, plus an FAQ page addressing common customer questions.

What's working

• AutoDealer and Organization JSON-LD schema present on archived pages — The archived homepage and new-vehicles page include AutoDealer and Organization schema with priceRange, name, and url fields, providing structured data that AI engines can use when accessible.
• BreadcrumbList schema on subpages — The locations page includes a well-formed BreadcrumbList schema, aiding navigation understanding for crawlers.
• Locations page with 20 Texas locations and 14+ luxury brands — The archived locations page contains detailed information about 20 Texas locations and 14+ luxury brands, providing valuable content for AI engines if accessible.
• Strong brand reputation known to LLMs from training data — LLMs recall Sewell as a family-owned Texas luxury dealership group founded in 1911, famous for Carl Sewell's book 'Customers for Life' and transparent pricing, which can be leveraged for AI visibility.
• DealerInspire CMS platform used — The site uses DealerInspire, a specialized automotive CMS, which may offer built-in SEO and schema features that can be optimized.