wardhadaway.com — AI Site Grade

Analysis

Ward Hadaway: JS-shell homepage with rich article content — AI crawlers see a ghost town

The homepage and all top-level navigation pages return fewer than 30 words of visible text from a plain GET, while individual insight articles contain 500–1,500 words of substantive legal content. AI crawlers are technically allowed (200 OK, no blocks) but receive a JavaScript shell on the site's most important pages.

Crawler Access

All major AI bots — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended, anthropic-ai, ChatGPT-User — receive a 200 status with identical byte size (135,922 bytes) to a browser baseline. No UA-based blocking exists. The robots.txt is minimal: a single User-agent: * rule disallowing only /wp-admin/, with no AI-bot-specific directives. The site runs on nginx with no CDN/WAF layer (no Cloudflare, no Akamai). No llms.txt exists (404). The homepage, /who-we-are/, /what-we-do/, /what-we-do/businesses/, and /insights/ all return under 30 words of extracted text — the content is loaded client-side via JavaScript, meaning AI crawlers that do not execute JS see essentially blank pages.

Content & Schema

The homepage has Organization and WebSite JSON-LD schema with SearchAction but no LegalService or Attorney schema type, despite being a law firm. Individual insight articles use Article schema with wordCount, datePublished, author, and image — well-structured. The site has a dedicated FAQ custom post type with ~200+ FAQ entries (mostly from 2020, COVID-era content) but the /faqs/ landing page is mis-titled "Insights & events" and shows stale pandemic content first. The FAQ pages lack FAQPage schema — they use CollectionPage or Article instead, missing a key opportunity for AI answer extraction. Heading structure is clean (single H1, H2/H3 hierarchy). Answer-format signals (FAQ, definition patterns, lists) are present on insight articles.

Cold-Knowledge Gap

The LLM prior knows Ward Hadaway as a North of England firm with offices in Leeds, Newcastle, and Manchester, founded in 1988, a top-100 UK firm, winner of Yorkshire Legal Awards "Law Firm of the Year" (2022) and British Legal Awards "Regional Law Firm of the Year" (2023). The actual site reveals five offices — Birmingham and Teesside are missing from the model's knowledge. The site also shows a Birmingham office prominently in navigation and contact pages, suggesting a geographic expansion the model has not absorbed. The model knows "FastFacts" legal updates but the site calls them "Insights & events." The model mentions healthcare, education, and social housing focus — the site confirms this with dedicated /what-we-do/healthcare/ and /what-we-do/public-third-sector/ sections.

External Signals

External search results for the firm are thin. Searches for reviews, Reddit mentions, and award citations returned zero results from DuckDuckGo. The firm's external footprint consists primarily of its own LinkedIn page and Vimeo channel. No third-party review sites, press coverage, or forum discussions surfaced in the search. This sparse external signal landscape means AI models have little off-domain material to supplement the thin JS-shell homepage content.

Surprising Findings

The FAQ archive is a time capsule: the vast majority of ~200+ FAQ entries date from March 2020 and concern COVID-19 business support (CBILS, furlough, remote working). These are five years stale and irrelevant to current queries. The site publishes fresh insight articles weekly (dates in May 2026) but the FAQ section has not been updated since 2020. The /faqs/ landing page H1 reads "Insights & events" — a copy-paste error from the insights template. The site has a 3,000+ URL sitemap (14 sub-sitemaps) indicating substantial content volume, but the top-level navigation pages that AI crawlers hit first are all JS shells. The dateModified on the homepage is set to 2026-05-28 (a future date relative to typical crawl cycles), which may confuse freshness signals.

Findings

1. Homepage and top-level pages render as JavaScript shells High

   The homepage, /who-we-are/, /what-we-do/, /what-we-do/businesses/, and /insights/ all return fewer than 30 words of visible text from a plain GET request. AI crawlers that do not execute JavaScript see essentially blank pages, severely limiting the site's AI visibility.

   What to change: Implement server-side rendering or static pre-rendering for all top-level navigation pages so that AI crawlers receive meaningful text content without requiring JavaScript execution.

2. Homepage lacks LegalService and Attorney schema Medium

   The homepage uses Organization and WebSite JSON-LD schema but does not include LegalService or Attorney schema types, despite being a law firm. This reduces the site's ability to be correctly classified by AI systems for legal queries.

   What to change: Add LegalService and Attorney schema markup to the homepage and relevant practice area pages.

3. FAQ pages lack FAQPage schema Medium

   The site has a dedicated FAQ custom post type with approximately 200 entries, but these pages use CollectionPage or Article schema instead of FAQPage schema. This misses a key opportunity for AI answer extraction and rich results.

   What to change: Apply FAQPage schema to all FAQ entries and ensure the FAQ landing page uses appropriate schema.

4. FAQ archive contains outdated COVID-19 content Medium

   The vast majority of the approximately 200+ FAQ entries date from March 2020 and concern COVID-19 business support (CBILS, furlough, remote working). This content is five years stale and irrelevant to current queries, potentially harming the site's credibility with AI systems.

   What to change: Review and update FAQ content to reflect current legal topics, or remove outdated entries. Implement a regular content freshness review process.

5. FAQ landing page mislabeled as 'Insights & events' Low

   The /faqs/ landing page has an H1 that reads 'Insights & events', which appears to be a copy-paste error from the insights template. This mislabeling confuses both users and AI crawlers about the page's purpose.

   What to change: Correct the H1 on the /faqs/ page to accurately reflect its content, e.g., 'Frequently Asked Questions'.

6. Homepage dateModified set to future date Low

   The homepage's dateModified is set to 2026-05-28, a future date relative to typical crawl cycles. This may confuse freshness signals for AI crawlers and search engines.

   What to change: Ensure dateModified reflects the actual last modification date of the page.

7. Sparse external footprint limits AI supplementation Medium

   External searches for reviews, Reddit mentions, and award citations returned zero results. The firm's external footprint consists primarily of its own LinkedIn page and Vimeo channel, providing AI models with little off-domain material to supplement the thin JS-shell homepage content.

   What to change: Encourage client reviews on platforms like Google, Trustpilot, or legal directories. Engage in legal forums and press outreach to build a richer external signal landscape.

8. No llms.txt file published Low

   The site returns a 404 for /llms.txt, missing an opportunity to provide AI crawlers with a curated list of important pages and guidance.

   What to change: Create an llms.txt file that lists key pages (e.g., practice areas, insights, people) to guide AI crawlers to the most valuable content.

9. AI model knowledge missing two offices Medium

   The LLM prior knows Ward Hadaway as having offices in Leeds, Newcastle, and Manchester, but the site shows five offices including Birmingham and Teesside. This geographic expansion is not reflected in the model's knowledge.

   What to change: Ensure office locations are prominently featured in structured data (e.g., LocalBusiness schema) and in visible text on the homepage and contact pages.

What's working

• All major AI bots allowed access — All tested AI crawlers (GPTBot, ClaudeBot, PerplexityBot, etc.) receive a 200 status with no UA-based blocking. The robots.txt only disallows /wp-admin/.
• Insight articles contain substantive legal content — Individual insight articles contain 500–1,500 words of substantive legal content with well-structured Article schema including wordCount, datePublished, author, and image.
• Clean heading hierarchy on content pages — Content pages use a single H1 with H2/H3 subheadings, providing clear document structure for AI crawlers.
• Comprehensive sitemap with 3,000+ URLs — The site has a sitemap index with 14 sub-sitemaps covering over 3,000 URLs, indicating substantial content volume and good crawl coverage.
• Fresh insight articles published weekly — The site publishes new insight articles regularly, with dates in May 2026, demonstrating ongoing content creation.
• Dedicated pages for key practice areas — The site has dedicated sections for healthcare and public/third sector, confirming the firm's focus areas as known by the LLM.