closebrothers.com — AI Site Grade

Analysis

Close Brothers Group plc — AI-Visibility Audit

The site's canonical URLs point to http:// (non-HTTPS) while every page serves over HTTPS, creating a persistent mixed-protocol signal that undermines how AI crawlers and knowledge graphs resolve the entity's canonical identity.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Applebot-Extended, Bytespider, anthropic-ai — receive a full 200 response with identical content to a browser visit. The Incapsula (Imperva) WAF does not discriminate by user-agent. The robots.txt contains no AI-bot-specific rules; only a generic User-agent: * block disallowing Drupal admin paths. The llms.txt returns 404, meaning no AI-friendly content map exists. The site runs on Drupal behind nginx/Varnish, and several key pages (e.g., /investor-relations/company-information/regulatory-news) return under 10 words of visible text — likely JS-rendered tables or feeds that AI crawlers cannot parse.

Cold-Knowledge Gap

The LLM prior knows Close Brothers as a UK merchant bank founded in 1878, focused on SME lending, asset finance, motor finance, and wealth management, and flags the FCA motor finance commission investigation as a recent reputational signal. The actual site leans heavily into a purpose-driven narrative ("helping the people and businesses of Britain thrive") and never uses the term "merchant bank" on the homepage or key pages — it calls itself a "UK specialist banking group." The site prominently addresses the FCA motor finance redress scheme on the homepage and investor relations page, but the linked news update page returns a 404. The cold model knows about wealth management and securities trading, yet the site's navigation and sitemap contain no wealth management or securities trading pages — those divisions appear to live on separate subdomains or are absent from the crawled content.

Schema Posture

Zero JSON-LD structured data exists on any page examined — homepage, who-we-are, our-purpose, investor-relations, sustainability, and motor finance pages all lack Organization, FinancialService, FAQPage, or any schema markup. This is a critical gap for a London Stock Exchange-listed financial institution that AI knowledge graphs should be able to resolve confidently. The homepage has no og: tags either. The h1 heading ("Helping you thrive") is aspirational but does not contain the brand name, which weakens entity recognition signals.

Content & Technical Fragility

The canonical tag on every page uses http://www.closebrothers.com/ while the live site redirects to https://www.closebrothers.com/. This HTTP canonical on an HTTPS site creates a mixed-signal problem for search engines and AI crawlers resolving the canonical URL. The sitemap contains 252 URLs and includes a /404 page and a /403 page as crawlable entries. Several insight articles reference "Covid 19" in titles, suggesting content has not been refreshed. The regulatory news page and the motor finance commission update page both render as near-empty shells for plain HTTP fetches, indicating JS-dependent content that AI crawlers cannot access.

Findings

1. HTTP canonical URLs on HTTPS site create mixed-protocol signal High

   Every page uses an http:// canonical URL while the live site serves over HTTPS and redirects to https://. This mixed-protocol signal undermines how AI crawlers and knowledge graphs resolve the entity's canonical identity.

   What to change: Update all canonical tags to use the https:// version of the URL. Ensure the sitemap also references HTTPS URLs.

2. Zero JSON-LD structured data on any page High

   No page examined — homepage, who-we-are, our-purpose, investor-relations, sustainability, motor finance — contains any JSON-LD schema markup. For a London Stock Exchange-listed financial institution, this is a critical gap that prevents AI knowledge graphs from confidently resolving the entity.

   What to change: Add Organization, FinancialService, and FAQPage JSON-LD schema to relevant pages. Include brand name, logo, and key financial services.

3. Key pages render as near-empty shells for AI crawlers High

   Pages like /investor-relations/company-information/regulatory-news and /investor-news/update-in-relation-to-motor-finance-commissions return under 10 words of visible text when fetched without JavaScript. AI crawlers cannot parse JS-rendered content, so these pages are effectively blank.

   What to change: Implement server-side rendering or provide static HTML fallback for critical content on these pages.

4. No llms.txt file for AI-friendly content map Medium

   The site returns a 404 for /llms.txt, meaning no AI-friendly content map exists. This limits the ability of AI crawlers to discover and understand the site's content structure.

   What to change: Create an llms.txt file listing key pages and summaries for AI crawlers.

5. Motor finance FCA redress update page returns 404 High

   The homepage and investor relations page prominently reference an FCA motor finance redress scheme, but the linked update page at /investor-news/update-in-relation-to-motor-finance-commissions returns a 404. This creates a broken user and crawler experience and undermines trust.

   What to change: Restore the page or update the links to point to a working URL with the relevant information.

6. Homepage H1 does not contain brand name Medium

   The homepage H1 is 'Helping you thrive' and does not include 'Close Brothers' or 'Close Brothers Group'. This weakens entity recognition signals for AI crawlers.

   What to change: Update the H1 to include the brand name, e.g., 'Close Brothers Group: Helping you thrive'.

7. No Open Graph tags on homepage Medium

   The homepage lacks og:title, og:description, and og:image tags, which limits how the site appears when shared on social platforms and may affect AI crawlers that use these signals.

   What to change: Add Open Graph meta tags to all pages, especially the homepage.

8. Sitemap includes /404 and /403 pages Low

   The sitemap contains entries for /404 and /403 error pages, which wastes crawl budget and signals poor site maintenance.

   What to change: Remove error pages from the sitemap.

9. Insight articles reference outdated 'Covid 19' in titles Low

   Several insight articles still reference 'Covid 19' in their titles, suggesting content has not been refreshed. This can harm credibility and AI perception of timeliness.

   What to change: Review and update outdated content, or remove or archive old articles.

10. Wealth management and securities trading pages absent from crawled content Medium

    The cold LLM knowledge includes wealth management and securities trading, but the site's navigation and sitemap contain no pages for these divisions. They may live on separate subdomains or be missing from the crawled content.

    What to change: Ensure all business divisions have dedicated pages on the main site or are linked from the sitemap.

What's working

• All major AI crawlers receive full 200 responses — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others all receive identical content to a browser visit. The Incapsula WAF does not discriminate by user-agent, and robots.txt has no AI-bot-specific blocks.
• Clear purpose-driven narrative on homepage and key pages — The site consistently communicates its purpose ('helping the people and businesses of Britain thrive') across the homepage, who-we-are, and our-purpose pages, providing a coherent brand story.
• Sitemap contains 252 URLs covering key sections — The sitemap includes a broad range of pages across investor relations, sustainability, insights, and business divisions, aiding discovery.
• Fast server response with nginx/Varnish caching — The site returns 200 responses quickly, with nginx and Varnish caching, ensuring good performance for crawlers.