schomp.com — AI Site Grade

Analysis

Schomp.com: A Next.js JS Shell Masquerading as a Multi-State Dealer Network

The homepage delivers 56 words of visible text to all AI crawlers despite returning a 200 status — a Next.js client-side rendered shell that loads 746KB of JavaScript bundles but produces almost no parseable content for any LLM crawler.

Crawler Access

All major AI bots — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, Applebot-Extended, Bytespider — receive a 200 status with identical byte sizes (~746KB) and the same JS-heavy shell. No bot is blocked at the server level. The robots.txt contains no AI-specific directives; the wildcard * rule has a Crawl-delay: 10 and blocks only inventory-feed, 404, terms, privacy, and print-related paths. The site runs on Google Frontend (GCP load balancer) with x-dealer-domain and x-dealer-name custom headers pointing to the Ridemotive platform. No Cloudflare or WAF is present. The llms.txt exists and is substantial (52KB), listing hundreds of inventory URLs — a rare and forward-looking asset — though every entry uses the same generic description "New quality vehicles in Aurora, Littleton, and Highlands Ranch."

Cold-Knowledge Gap

The LLM knows Schomp as a Colorado-only, family-owned dealer group founded in 1925, known for no-haggle pricing and brands like BMW, MINI, Honda, Subaru, and Hyundai. The actual site reveals a much larger footprint: Schomp now operates Mercedes-Benz of Farmington (UT), Porsche of Salt Lake City (UT), Volkswagen of Fairfield (CA), BMW of Fairfield (CA), and Audi Napa Valley (CA) — none of which the model mentioned. The cold knowledge also dates the founding to 1925, while the site's history page says the business started in 1941. The "One Price. One Person. One Hour." trademarked philosophy is the brand's strongest differentiator, yet the model's prior only vaguely recalled "no-haggle pricing."

Schema Posture

The homepage and all subpages carry AutoDealer and WebSite JSON-LD schema with SearchAction — but the schema has critical flaws. The PostalAddress object lists streetAddress as empty string, addressLocality as "Aurora, Littleton, and Highlands Ranch" (three towns crammed into one field), and postalCode as 84025 — which is a Utah ZIP code (Farmington, UT), not Colorado. The GeoCoordinates point to roughly Denver, but the ZIP is wrong. The telephone field is empty across all schema instances. Location-specific pages (Schomp BMW, Porsche SLC, etc.) carry zero schema markup — no LocalBusiness, no GeoCoordinates, no OpeningHoursSpecification at the individual dealership level. The /one-price and /schompway-service pages do include FAQPage schema with well-structured Q&A.

Sitemap Anomalies

The sitemap.xml contains 4,201 URLs and reveals serious quality issues. Multiple URLs include zero-width space characters (\u200b, \u200e) in their paths — https://schomp.com/locations/audi-napa-valley-\u200b and https://schomp.com/locations/schomp-mini\u200e — creating duplicate, hard-to-crawl entries. The URL https://schomp.com/locations/schomp-bmw-(schompcom) redirects to the clean /locations/schomp-bmw but remains in the sitemap. The Audi Napa Valley page (/locations/audi-napa-valley) returns a blank title tag and no H1 heading. The sitemap also includes inventory filter URLs with encoded JSON query strings, which are unlikely to add crawl value.

External Signals

Web search returned zero indexed results for Schomp across multiple queries — no reviews, no Reddit threads, no press mentions surfaced through DuckDuckGo. This suggests the site has minimal external citation footprint, which limits the corpus data AI engines can draw on when generating answers about the brand.

Findings

1. Homepage delivers only 56 words of visible text to AI crawlers High

   The homepage returns a 200 status but serves a Next.js client-side rendered shell with 746KB of JavaScript, yielding only 56 words of parseable content for all major AI bots.

   What to change: Implement server-side rendering or static generation for key pages to ensure meaningful HTML content is delivered to crawlers.

2. Schema PostalAddress has empty streetAddress and telephone fields High

   The JSON-LD AutoDealer schema on the homepage lists streetAddress as an empty string and telephone as empty, which undermines local search visibility.

   What to change: Populate streetAddress and telephone with the correct physical address and phone number of the main dealership.

3. Schema PostalAddress uses wrong ZIP code 84025 (Utah) for Colorado dealership High

   The homepage schema lists postalCode as 84025, which corresponds to Farmington, UT, not the Colorado locations mentioned in addressLocality.

   What to change: Correct the postalCode to the appropriate Colorado ZIP code for the main dealership address.

4. Schema addressLocality crams three cities into one field Medium

   The addressLocality field contains 'Aurora, Littleton, and Highlands Ranch' instead of a single locality, which violates schema best practices and may confuse parsers.

   What to change: Use a single locality per schema instance or create separate schema blocks for each dealership location.

5. Individual dealership pages lack LocalBusiness schema High

   Location-specific pages such as Schomp BMW, Porsche of Salt Lake City, and Audi Napa Valley have no LocalBusiness, GeoCoordinates, or OpeningHoursSpecification schema markup.

   What to change: Add LocalBusiness schema with correct name, address, phone, geo coordinates, and opening hours to each dealership page.

6. LLM cold knowledge misses Schomp's multi-state expansion Medium

   The LLM's prior knowledge only covers Colorado locations, but the site now operates dealerships in Utah and California, which are absent from the model's knowledge.

   What to change: Increase external citations and press releases about the expansion to improve LLM knowledge coverage.

7. LLM cold knowledge cites wrong founding year (1925 vs 1941) Medium

   The LLM states Schomp was founded in 1925, but the site's history page says the business started in 1941, creating a factual discrepancy.

   What to change: Ensure the founding year is prominently and consistently mentioned across the site and in structured data.

8. Sitemap contains URLs with zero-width space characters Medium

   Multiple sitemap URLs include invisible Unicode characters (\u200b, \u200e), creating duplicate entries and potential crawl issues.

   What to change: Remove zero-width characters from all URLs and ensure the sitemap contains only clean, canonical URLs.

9. Sitemap includes a URL that redirects Low

   The URL /locations/schomp-bmw-(schompcom) redirects to /locations/schomp-bmw but remains in the sitemap, wasting crawl budget.

   What to change: Replace the redirecting URL with the canonical version in the sitemap.

10. Audi Napa Valley page has blank title tag and no H1 High

    The page /locations/audi-napa-valley returns a blank HTML title and lacks an H1 heading, severely harming SEO and crawlability.

    What to change: Add a descriptive title tag and an H1 heading to the Audi Napa Valley page.

11. Minimal external citations and indexed results Medium

    Web searches for Schomp returned zero indexed results across multiple queries, indicating a weak external signal footprint that limits AI training data.

    What to change: Invest in PR, backlinks, and local citations to increase the site's external presence.

12. llms.txt uses generic descriptions for all inventory URLs Low

    The llms.txt file lists hundreds of inventory URLs but each uses the same generic description 'New quality vehicles in Aurora, Littleton, and Highlands Ranch', reducing usefulness.

    What to change: Provide unique, descriptive summaries for each inventory URL in llms.txt.

What's working

• llms.txt file exists and is substantial — The site hosts a 52KB llms.txt file listing hundreds of inventory URLs, a forward-looking asset that helps AI crawlers discover content.
• All major AI bots are allowed access — No AI-specific directives in robots.txt block GPTBot, ClaudeBot, or other crawlers, and all receive 200 responses.
• FAQPage schema on /one-price and /schompway-service — The /one-price and /schompway-service pages include well-structured FAQPage schema, which can generate rich results and answer snippets.
• AutoDealer and WebSite schema present on homepage — The homepage includes JSON-LD AutoDealer and WebSite schema with SearchAction, providing basic structured data for search engines.
• History page provides detailed brand narrative — The /our-history page contains 2,350 words of rich content about the company's founding and growth, which can inform AI models.
• Blog section exists with regular content — The /blog page contains 236 words of content, providing a channel for ongoing content creation and external signals.
• Sitemap covers 4,201 URLs for broad crawl coverage — The sitemap includes a large number of URLs, indicating an attempt to provide comprehensive crawl coverage despite quality issues.
• One Price philosophy is a strong brand differentiator — The 'One Price. One Person. One Hour.' trademarked philosophy is clearly presented on the site and is a unique selling point that can be leveraged in AI answers.