AI Site Grade
allstatepeterbilt.com — AI Site Grade
Allstate Peterbilt Group's llms.txt returns 1MB of homepage HTML instead of a structured AI content map, and the site lacks any AI-specific crawler directives.
The site has no AI-specific blocking but suffers from a misconfigured llms.txt, missing product schema on inventory, and zero external signals, limiting AI visibility.
- Findings
- 8
- Evidence checks
- 22
- Completed
- 30 May 2026
Analysis
llms.txt returns the full homepage HTML (1MB+) instead of a text file
The /llms.txt URL at allstatepeterbilt.com does not serve a plain-text AI content map. It returns the full HTML of the homepage (1,084,118 bytes), including the entire <head> with New Relic instrumentation, Google Tag Manager scripts, and the full page DOM. This is not a valid llms.txt — it is a server misconfiguration where the CMS (Dealerfire platform) is routing the /llms.txt path to the default page controller. Any AI crawler requesting llms.txt will receive an HTML blob it cannot parse as structured content.
Crawler Access
All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai — receive a full 200 response with the same byte payload (~695 KB) as a browser. No UA-based blocking exists. The robots.txt uses a blanket Allow: / with no AI-specific directives whatsoever. No AI bot is mentioned by name. The Crawl-delay: 10 for * is reasonable, though msnbot and bingbot get 60-second delays. The site runs on nginx with HSTS, X-Frame-Options SAMEORIGIN, and no CDN/WAF layer (no Cloudflare, no Akamai). The hosting is on Network Solutions (NS: worldnic.com), IP 136.179.129.162.
Cold-Knowledge Gap
A frontier LLM queried cold about "Allstate Peterbilt Group" described it as a network of Peterbilt dealerships serving the Midwest and Great Plains, mentioning Illinois, Indiana, Iowa, and Wisconsin — but the site's actual footprint covers Minnesota, Wisconsin, North Dakota, South Dakota, and Iowa only. The model hallucinated Illinois and Indiana locations. It also claimed the group has been a Peterbilt dealer "since the 1970s" — the site itself contains no founding date or history page to confirm or refute this. The model correctly identified the Model 579, 567, and 520 as key products, but the site's homepage prominently features the Model 589 (described as "this staple of the Peterbilt line") which the model did not mention. The model also described "proprietary upfit packages" — the site does not mention these anywhere.
Schema Posture
Every page carries the same duplicated JSON-LD block: an AutoDealer and AutomotiveBusiness schema for the corporate office in Eagan, MN, with five department entries (Main Office, Sales, Finance, Service, Parts). All department images point to https://cdn-ds.com/noimage/noimage.jpg — a placeholder. The priceRange field is set to " - " (empty range). No individual location schemas exist despite 20+ physical stores. No ItemList or Product schema is applied to inventory pages, meaning individual vehicle listings (299 used trucks, 231 new trucks) have no structured data for AI engines to index as products. No FAQPage, HowTo, or BreadcrumbList schema is present anywhere.
External Signals
The site has virtually no discoverable external footprint. Searches for reviews, Reddit threads, and press mentions returned zero results. The only off-domain signals are social media links in the footer: Facebook, Instagram, YouTube, and Twitter/X — all pointing to branded accounts. The site is built on the Dealerfire platform (footer credit: dealerfire.com), a common automotive dealership CMS. The sitemap is a sitemap index with three gzipped sub-sitemaps (static, vehicle, vehicleimages), but the vehicle sitemap could not be parsed as plain text — it is gzip binary that the fetch tool could not decompress. The vehicle-comparison page is an empty shell ("No vehicles found to compare") with only 6 words of visible text.
Findings
llms.txt returns full homepage HTML instead of a text file High
The /llms.txt URL serves the entire homepage HTML (1,084,118 bytes) including scripts and DOM, not a plain-text AI content map. This is a server misconfiguration that prevents AI crawlers from parsing structured content.
What to change: Configure the server to serve a valid llms.txt file with plain-text summaries of key pages, or remove the route if not intended.
No AI-specific directives in robots.txt Medium
The robots.txt uses a blanket Allow: / with no AI bot names mentioned. No GPTBot, ClaudeBot, or other AI crawlers are explicitly allowed or disallowed, leaving AI access uncontrolled.
What to change: Add explicit directives for AI crawlers (e.g., Allow or Disallow) to control AI access and signal intent.
No Product or ItemList schema on inventory pages High
Inventory pages listing 299 used and 231 new trucks lack any Product or ItemList structured data. AI crawlers cannot index individual vehicle details as products.
What to change: Add JSON-LD Product schema for each vehicle listing and ItemList schema for inventory pages.
Duplicate AutoDealer schema with placeholder images and empty price range Medium
Every page carries the same JSON-LD block for the corporate office with five department entries, all using placeholder images (noimage.jpg) and an empty priceRange field. No individual location schemas exist.
What to change: Remove duplicate schema from non-corporate pages, add individual LocalBusiness schemas for each dealership location, and replace placeholder images with real ones.
LLM cold knowledge hallucinates locations and history Medium
A frontier LLM incorrectly stated the group serves Illinois and Indiana (actual footprint is MN, WI, ND, SD, IA) and claimed dealership history since the 1970s without site confirmation. The site lacks a history page or location list that could ground AI knowledge.
What to change: Add a dedicated history/about page and a comprehensive locations page with structured data to correct AI hallucinations.
No discoverable external reviews or mentions Medium
Searches for reviews, Reddit threads, and press mentions returned zero results. The site has no external footprint beyond social media links, limiting AI citation sources.
What to change: Encourage customer reviews on third-party platforms and build a PR presence to generate external signals.
Vehicle comparison page is an empty shell Low
The /vehicle-comparison page contains only 6 words of visible text and no vehicles to compare, providing no value to AI crawlers or users.
What to change: Either populate the page with comparison functionality or remove it to avoid thin content.
Missing BreadcrumbList and FAQPage schema Low
The site does not use BreadcrumbList or FAQPage structured data, missing opportunities for rich results in AI search.
What to change: Add BreadcrumbList schema to all pages and FAQPage schema to relevant content pages.
What's working
- All major AI crawlers allowed with 200 responses — All tested AI crawlers (GPTBot, ClaudeBot, PerplexityBot, etc.) receive full 200 responses with the same content as browsers. No UA-based blocking exists.
- Sitemap index with sub-sitemaps for vehicles — A sitemap index exists with three gzipped sub-sitemaps (static, vehicle, vehicleimages), helping crawlers discover inventory pages.
- Social media profiles linked in footer — The site links to Facebook, Instagram, YouTube, and Twitter/X branded accounts, providing some off-domain signals.
- Reasonable crawl-delay for general bots — The robots.txt sets Crawl-delay: 10 for all bots, which is reasonable and prevents overloading.
Track allstatepeterbilt.com across AI search
This is one snapshot. Open the interactive report to inspect evidence, or grade another site free.