shippersautomotive.com — AI Site Grade

Analysis

Shippers Automotive Group — AI-Visibility Audit

GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, and Bytespider all receive zero-byte disconnects (status 0 / RemoteProtocolError) from the nginx server, while ChatGPT-User, OAI-SearchBot, PerplexityBot, and Perplexity-User get full 200 responses with identical content to a browser — a selective bot-blocking pattern that silently starves the most influential AI crawlers of content.

Crawler Access

The robots.txt at shippersautomotive.com/robots.txt contains only a generic User-agent: * rule disallowing /wp-admin/ — no AI-bot directives whatsoever. Despite this permissive stance, the server (nginx on Google Cloud, IP 35.209.33.122) actively terminates connections for GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, and Bytespider. The disconnect is not a Cloudflare challenge (no CF headers present) but a server-side UA filter. ChatGPT-User, OAI-SearchBot, PerplexityBot, and Perplexity-User all pass through cleanly, receiving the full 156KB HTML payload. This creates a paradoxical situation: OpenAI's search and chat products can index the site, but GPTBot itself cannot.

Cold-Knowledge Gap

A frontier LLM queried cold about "Shippers Automotive" returned: *"I do not have specific, verifiable information about a company named 'Shippers Automotive'... It may be a small or regional company... I cannot confirm any notable products, services, or well-known facts."* The site itself describes a national 3PL operating 1.4 million square feet across six states (Ohio, Michigan, Alabama, Tennessee, Texas, Oklahoma), with ISO 9001 certification, an AIAG membership, a proprietary WMS called WDLS, and a 2023 CSR report. The gap between the company's actual operational footprint and what AI models know cold is total — the brand is effectively invisible to LLM training data.

Content & Schema Posture

The site runs WordPress with Divi theme and All in One SEO Pro. Every page carries consistent Organization JSON-LD schema with name, telephone (+19374847780), logo, and LinkedIn sameAs. However, no LocalBusiness, Service, or FAQPage schema types are present — the services page describes seven distinct offerings (line-side delivery, sequencing, kitting, re-packing, sub-assembly, cross-docking, container management) with zero structured markup. The homepage has no FAQ schema despite describing a service business. The llms.txt exists (auto-generated by AIOSEO) but is polluted with raw Divi shortcodes ([et_pb_section fb_built="1"...]) on the disclaimer, terms, privacy, code-of-conduct, and diversity pages, making those entries useless to LLMs.

External Signals

Multiple web searches for the company name, brand + "logistics", brand + "3PL", brand + city names, and even the domain itself returned zero indexed results across DuckDuckGo. The only external citation is a single LinkedIn company page and a link to hemlockstrategy.com (likely a web development partner). No press mentions, no Reddit threads, no industry directory listings, no customer reviews, no news coverage appear in search. The CSR PDF (2023) exists on the server but is not linked from any discoverable external source. The site has no blog, no case studies, no whitepapers, and no news section — nothing that would generate inbound links or social signals.

Surprising Details

The sitemap's lastmod dates show the homepage was last modified May 13, 2026 — a future date, likely a WordPress plugin quirk. The dateModified in the homepage JSON-LD reads 2026-05-13T18:27:23-04:00, which is also in the future. The site has only 14 total URLs (all flat pages), zero blog posts, and the aioseo-location-sitemap.xml sub-sitemap exists but appears empty. The employment page accepts file uploads directly — a potential security surface. The corporate office is in Jacksonville, Florida, but all six operational locations are in the Midwest and South, with no Florida warehouse listed.

Findings

1. Selective bot blocking starves GPTBot, ClaudeBot, and Google-Extended High

   The nginx server terminates connections for GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, and Bytespider with zero-byte disconnects, while allowing ChatGPT-User, OAI-SearchBot, PerplexityBot, and Perplexity-User to receive full content. This pattern silently blocks the most influential AI crawlers despite a permissive robots.txt.

   What to change: Remove the server-side user-agent filter that blocks GPTBot, ClaudeBot, and Google-Extended, or configure it to serve content to all AI crawlers equally.

2. Brand is completely absent from LLM training data High

   A frontier LLM queried cold about 'Shippers Automotive' returned no verifiable information, despite the company operating a national 3PL with 1.4 million square feet across six states, ISO 9001 certification, and a proprietary WMS. The brand is effectively invisible to AI models.

   What to change: Build external signals through press releases, industry directory listings, guest posts, and social media activity to establish the brand in training data.

3. Zero indexed results across web searches for brand and domain High

   Multiple web searches for the company name, brand plus 'logistics', '3PL', city names, and the domain itself returned zero results on DuckDuckGo. No press mentions, reviews, or directory listings exist.

   What to change: Create and distribute press releases, claim business listings on industry directories, and publish case studies or blog content to generate inbound links.

4. No structured markup for seven distinct services High

   The services page describes line-side delivery, sequencing, kitting, re-packing, sub-assembly, cross-docking, and container management, but none of these offerings have Service or LocalBusiness schema. The homepage also lacks FAQ schema.

   What to change: Add Service schema for each offering on the services page and FAQ schema on the homepage to improve AI understanding.

5. llms.txt contains raw Divi shortcodes making entries useless Medium

   The auto-generated llms.txt includes raw Divi shortcodes (e.g., [et_pb_section fb_built="1"...]) on disclaimer, terms, privacy, code-of-conduct, and diversity pages, rendering those entries unreadable to LLMs.

   What to change: Clean the llms.txt file by removing shortcodes and providing plain-text summaries for each page.

6. Sitemap and schema contain future dates (2026) Low

   The sitemap's lastmod and the homepage JSON-LD dateModified both show May 13, 2026, a future date likely caused by a WordPress plugin quirk, which may confuse crawlers.

   What to change: Correct the date settings in WordPress or the All in One SEO plugin to output accurate dates.

7. No blog, case studies, or news section to generate content signals Medium

   The site has only 14 flat pages with no blog, case studies, whitepapers, or news section, limiting opportunities for inbound links and fresh content for AI crawlers.

   What to change: Add a blog or news section with regular posts about industry insights, company updates, and case studies.

8. Location sitemap exists but appears empty Low

   The aioseo-location-sitemap.xml sub-sitemap exists but contains no URLs, missing an opportunity to help crawlers discover location pages.

   What to change: Populate the location sitemap with URLs for each operational location page.

9. Employment page accepts file uploads directly Low

   The employment page allows file uploads, which could be a security surface if not properly sanitized, though not directly an AI visibility issue.

   What to change: Review file upload security on the employment page to prevent potential abuse.

What's working

• Consistent Organization JSON-LD schema on all pages — Every page carries Organization schema with name, telephone, logo, and LinkedIn sameAs, providing a baseline identity signal to AI crawlers.
• llms.txt file is present and served — The site has an llms.txt file (auto-generated by AIOSEO) that provides a starting point for LLM content discovery, despite shortcode pollution.
• Robots.txt allows all AI bots to crawl — The robots.txt file has no AI-bot-specific disallow rules, only blocking /wp-admin/ for all user-agents, which is a correct baseline.
• CSR report PDF is accessible on the server — A 2023 CSR report PDF (3271 words) is hosted on the server, providing substantive content that could be indexed by crawlers.
• All discovered pages return 200 status — Every page on the site returns a successful HTTP 200 response, ensuring no broken links or errors for crawlers.