taylorwessing.com — AI Site Grade

Analysis

I have enough data now. Let me compile the audit.

Taylor Wessing's site is fully open to every AI crawler tested — yet the navigation links to at least five major landing pages that return 404 errors, creating a hollow shell for any bot that follows internal links.

Crawler Access

All 11 AI crawlers tested — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, anthropic-ai, Bytespider, Applebot-Extended, Perplexity-User — receive a 200 OK with full HTML content on the homepage, identical to a browser baseline (~143KB). The robots.txt uses a single catch-all User-agent: * rule with no AI-specific directives; no bot is blocked, rate-limited, or redirected. The site runs on Azure (Microsoft CDN, x-azure-ref header) with a strong Content-Security-Policy and HSTS. No JS-rendering risk exists — all pages tested return rich server-rendered HTML.

Broken Navigation Architecture

The homepage navigation prominently links to /en/insights, /en/news, /en/events, /en/media-centre, and /en/locations — all five return 404 errors with noindex meta tags. The sitemap at sitemapindex.ashx (referenced in robots.txt) correctly points to language-specific sub-sitemaps, but the English sitemap at /en/sitemap.ashx timed out on fetch. The /llms.txt endpoint returns a 404. This means any AI crawler that follows the site's own navigation links to discover content depth will hit dead ends, limiting what gets indexed to the homepage, about, expertise, people, and careers pages.

Cold-Knowledge Gap

The LLM prior knows Taylor Wessing as a UK-German merger (2005) with strong IP/tech practices and mentions "high-profile data breach litigation" as a recent reputational signal. The actual site positions itself as a "global law firm that serves the world's most innovative people and businesses" with heavy emphasis on AI advisory, life sciences, and cross-border M&A (the "Dealonomics" campaign). The site contains zero mention of the 2005 merger origin, zero mention of data breach litigation, and zero FAQ or comparison-table content that would help AI engines extract structured answers. The cold knowledge is stale by at least 2-3 years on the firm's current messaging.

Schema Posture

The homepage carries only a minimal Organization schema (name + URL). Deeper pages add BreadcrumbList. The people directory uses rich Person schema with jobTitle, worksFor, image, and sameAs — this is the strongest structured data on the site. However, no WebPage, Article, FAQPage, Service, or LegalService schema exists anywhere. The AI advisory service page (/en/expertise/services/artificial-intelligence) — a 3,981-word deep-dive — has no service schema, no FAQ schema, and no article schema despite containing extensive Q&A-style content about AI governance and regulatory compliance.

External Signals

The site has an OpenAI domain verification TXT record (openai-domain-verification=dv-1AEgUDzXHHTclHfAlWN9ClnM), confirming proactive engagement with OpenAI's crawler ecosystem. DNS records show Microsoft 365 / Azure hosting, Hornetsecurity for mail, and multiple verification tokens (Apple, Google, Atlassian, DocuSign). No recent press or review content surfaced in web search queries, suggesting limited third-party citation volume that AI engines can draw on for authority signals.

Findings

1. Five major navigation links return 404 errors High

   The homepage navigation links to /en/insights, /en/news, /en/events, /en/media-centre, and /en/locations, all of which return 404 errors with noindex meta tags. AI crawlers following internal links hit dead ends, limiting indexed content to a few pages.

   What to change: Restore these pages or remove the broken links from navigation. Ensure all linked pages return 200 status with meaningful content.

2. English sitemap times out High

   The sitemap at /en/sitemap.ashx timed out on fetch, preventing crawlers from discovering all English-language pages efficiently.

   What to change: Ensure the English sitemap loads quickly and is accessible. Consider using a static XML sitemap instead of a dynamic handler.

3. No /llms.txt file Medium

   The /llms.txt endpoint returns a 404, missing an opportunity to guide AI crawlers to key content.

   What to change: Create an /llms.txt file listing important pages like the AI advisory service page and people directory.

4. Cold knowledge is 2-3 years out of date Medium

   LLM prior knowledge describes Taylor Wessing as a UK-German merger with data breach litigation, but the current site emphasizes AI advisory, life sciences, and Dealonomics. The mismatch means AI engines may present outdated information.

   What to change: Publish more current content about the firm's strategy, practice areas, and recent achievements to update AI training data.

5. No service or FAQ schema on AI advisory page High

   The 3,981-word AI advisory page lacks Service, FAQPage, or Article schema, reducing its ability to appear in rich AI-generated answers.

   What to change: Add Service schema with description and area served, and FAQPage schema for the Q&A sections.

6. Homepage has only minimal Organization schema Medium

   The homepage only includes Organization schema with name and URL, missing description, logo, sameAs, and other properties that help AI engines understand the firm.

   What to change: Expand Organization schema to include description, logo, sameAs (LinkedIn, Twitter), and founding date.

7. No Article schema on content pages Medium

   Even if insights pages existed, they lack Article schema, which helps AI engines identify and cite articles.

   What to change: Implement Article schema on all news and insight pages with headline, datePublished, author, and image.

8. Low third-party citation volume Medium

   Web searches for Taylor Wessing in context of AI visibility, reviews, and insights returned zero results, indicating limited external content that AI engines can use for authority signals.

   What to change: Encourage third-party publications, guest articles, and press releases to increase external citations.

9. No FAQ or comparison content Medium

   The site lacks FAQ pages or comparison tables that AI engines often extract for structured answers, reducing chances of appearing in featured snippets or AI summaries.

   What to change: Create FAQ pages for key practice areas and comparison tables for services.

What's working

• All 11 AI crawlers receive full HTML content — The homepage returns 200 OK with rich server-rendered HTML to all tested AI crawlers, with no blocking or rate-limiting.
• OpenAI domain verification TXT record present — The DNS includes an openai-domain-verification TXT record, indicating proactive engagement with OpenAI's crawler ecosystem.
• People directory uses rich Person schema — The people pages include Person schema with jobTitle, worksFor, image, and sameAs, helping AI engines understand attorney profiles.
• BreadcrumbList schema on deeper pages — Deeper pages include BreadcrumbList schema, aiding navigation understanding for crawlers.
• Azure hosting with strong security headers — The site runs on Azure with Content-Security-Policy and HSTS, ensuring reliable and secure delivery.
• No JavaScript rendering required — All tested pages return rich server-rendered HTML, eliminating JS rendering risks for AI crawlers.