nielsenautos.com — AI Site Grade

Analysis

Akamai WAF Grants Exclusive AI Access to Anthropic

The site's Akamai Web Application Firewall returns 403 Access Denied to every user-agent tested except anthropic-ai, which reaches a backend nginx server and receives a full 200 response with 314KB of HTML. This means GPTBot, Google-Extended, PerplexityBot, OAI-SearchBot, ChatGPT-User, and even a standard browser all see a 15-word "Access Denied" page, while only Claude's crawler sees the actual dealership content. The robots.txt is also blocked by Akamai for non-Anthropic bots, making it effectively invisible to the broader AI ecosystem.

Crawler Access

The robots.txt (accessible only via anthropic-ai UA) explicitly disallows GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, PerplexityBot, CCBot, Baiduspider, and Yandex from /api/, /apis/, /pixall/, and static asset paths. However, this is moot: the Akamai WAF blocks these bots at the edge before they ever reach the nginx server to read the robots.txt. The llms.txt file exists (870KB, listing hundreds of pages with descriptions) and is served by nginx, but only to anthropic-ai — no other AI crawler can reach it. The sitemap (569KB, 2026-05-30 lastmod) is similarly gated.

Cold-Knowledge Gap

LLM knowledge describes "Nielsen Autos" as a family-owned New Jersey dealership group founded in 1965 by John Nielsen, operating Ford, Lincoln, and Kia franchises with a "Nielsen's Way" no-haggle philosophy. The actual site brands itself as Nielsen Automotive Group with 15+ dealership locations across 11 brands (Chevrolet, Chrysler, Dodge, FIAT, Ford, Hyundai, INFINITI, Jeep, Kia, Mitsubishi, Nissan, RAM) — a much larger operation than the cold knowledge reflects. The cold knowledge mentions "mixed reviews on Google and Yelp" but the site itself contains no review or reputation signals.

Schema and Content Posture

The homepage contains no JSON-LD structured data — no AutoDealer, Organization, or LocalBusiness schema was detected. The site runs on the Dealer.com (DDC) platform and is JS-heavy (prefetching multiple JS bundles for inventory faceting and filtering). The /about/index.htm page listed in the sitemap returns a 404 ("We Messed Up! Navigate to a Different Page, Please!"), as does /about-us.htm and /why-choose-us.htm. The homepage heading structure is flat: a single H1 ("Nielsen Automotive Group") with H3 subheadings for inventory, service, financing, and about sections. No FAQ, comparison tables, or answer-format signals are present.

External Signals

The site has zero indexed external mentions in search results — no Reddit threads, press coverage, review aggregator pages, or dealer directory listings surfaced. The domain has been archived by the Wayback Machine since 2013 with 173 captures, indicating long-standing operation, but the current Akamai WAF configuration has effectively removed the site from the visible web for non-Anthropic crawlers. DNS records show Sophos email security (Hydra platform) and GoDaddy nameservers, with the site hosted behind Akamai's edge network.

Findings

1. Akamai WAF blocks all non-Anthropic AI crawlers High

   The Akamai Web Application Firewall returns 403 Access Denied to every user-agent except anthropic-ai. GPTBot, Google-Extended, PerplexityBot, OAI-SearchBot, ChatGPT-User, and standard browsers all see a 15-word denial page, while only Claude's crawler receives the full 314KB HTML.

   What to change: Reconfigure the Akamai WAF to allow access for major AI crawlers (GPTBot, Google-Extended, PerplexityBot, etc.) or implement a rate-limiting approach instead of blanket blocking.

2. Robots.txt inaccessible to non-Anthropic bots High

   The robots.txt file returns 403 for standard user-agents and is only reachable by anthropic-ai. It disallows GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot, and others from certain paths, but the WAF makes these rules irrelevant for most crawlers.

   What to change: Ensure robots.txt is publicly accessible and update it to allow AI crawlers on relevant paths.

3. LLMs.txt only served to Anthropic High

   The llms.txt file (870KB, listing hundreds of pages) is served by nginx but only to anthropic-ai. Other AI crawlers cannot access it due to the WAF block.

   What to change: Make llms.txt publicly accessible so all AI crawlers can discover the site's content.

4. Sitemap only accessible to Anthropic High

   The sitemap.xml (569KB, 2026-05-30 lastmod) is served by nginx but only to anthropic-ai. Other crawlers cannot reach it.

   What to change: Make sitemap.xml publicly accessible to all crawlers.

5. No JSON-LD structured data on homepage High

   The homepage contains no JSON-LD structured data such as AutoDealer, Organization, or LocalBusiness schema, which limits AI understanding and rich result eligibility.

   What to change: Add JSON-LD structured data for AutoDealer, Organization, and LocalBusiness to the homepage and key pages.

6. Key pages return 404 errors High

   Pages listed in the sitemap such as /about/index.htm, /about-us.htm, and /why-choose-us.htm return 404 errors with a 'We Messed Up!' message, indicating broken internal links.

   What to change: Fix or remove broken links to these pages and ensure all sitemap URLs resolve correctly.

7. Zero indexed external mentions High

   No external mentions of the site or dealership group were found in search results, including Reddit, reviews, press, or directory listings, indicating extremely low online visibility.

   What to change: Build external signals through local business directories, press releases, and social media engagement.

8. Cold knowledge understates dealership size Medium

   LLM knowledge describes a smaller operation (Ford, Lincoln, Kia) while the site actually represents 15+ locations across 11 brands, creating a gap between AI perception and reality.

   What to change: Publish comprehensive brand and location information on the site and ensure it is crawlable to update AI knowledge.

9. JS-heavy page may hinder content extraction Medium

   The site runs on Dealer.com platform and prefetches multiple JS bundles for inventory faceting, which may delay or complicate content extraction for AI crawlers that do not execute JavaScript.

   What to change: Ensure critical content is server-side rendered or available in static HTML to improve crawlability.

10. Flat heading structure with single H1 Low

    The homepage uses a single H1 and multiple H3 subheadings, lacking a clear hierarchical outline that helps AI understand content organization.

    What to change: Use a proper heading hierarchy (H1, H2, H3) to structure content sections.

11. No FAQ or comparison content for AI answers Medium

    The site lacks FAQ, comparison tables, or answer-format content that AI systems often use to generate direct answers.

    What to change: Add FAQ sections and comparison tables for vehicles and services to improve AI answer generation.

What's working

• Anthropic crawler receives full site content — The anthropic-ai user-agent is allowed through the WAF and receives complete HTML, robots.txt, sitemap, and llms.txt, ensuring Claude can index the site.
• LLMs.txt exists with extensive page listing — An 870KB llms.txt file lists hundreds of pages with descriptions, providing a comprehensive content map for AI crawlers that can access it.
• Large sitemap with recent lastmod date — The sitemap.xml is 569KB and has a lastmod date of 2026-05-30, indicating active site maintenance and a large inventory of pages.
• Domain has long-standing web presence — The domain has been archived by the Wayback Machine since 2013 with 173 captures, indicating a well-established online presence.
• Robots.txt disallows AI crawlers on sensitive paths — The robots.txt explicitly disallows GPTBot, OAI-SearchBot, and others from /api/, /apis/, and /pixall/, which is a good practice to prevent crawling of non-public endpoints.