leaplegalsoftware.com — AI Site Grade

Analysis

I have enough data to write a sharp audit. Let me compile.

The homepage redirects to /uk/ with no region selector for non-UK visitors, and the site's llms.txt returns a 404 Nuxt.js shell — a missed opportunity for a company aggressively marketing AI tools.

Crawler Access

All 11 tested AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User) receive 200 status with identical byte payload (202,076 bytes) as a browser — no UA-based blocking exists. The robots.txt is a single User-agent: * rule allowing everything, with no AI-bot-specific directives whatsoever. The site runs on Netlify (Nuxt.js SSR), meaning content is server-rendered and fully readable by crawlers. However, several internal URLs (/uk/company/, /uk/company/leadership/, /uk/about-us/, /uk/thought-leadership-blog/) return 404 Nuxt.js shells ("Loading...") — these are linked from the homepage navigation but lead to dead ends for any crawler.

Cold-Knowledge Gap

The LLM prior knows LEAP as "practice management for small to mid-sized law firms" with "built-in legal forms" and "Microsoft Office integration," and recalls the TPG Capital acquisition in 2018 and "user complaints about customer support responsiveness and pricing increases." The actual site makes zero mention of TPG Capital or private equity ownership. The site heavily promotes Matter AI, AI Prompts, and LawY integration — three AI products the cold model did not reference at all. The model's knowledge of LEAP is stuck at the pre-AI-product era, while the site positions AI as its primary differentiator ("Legal AI" appears in the title tag).

Schema Posture

The homepage carries rich JSON-LD: Organization, SoftwareApplication, WebSite, WebPage, and a FAQPage with 4 questions answered inline. The SoftwareApplication schema correctly lists operatingSystem as "Windows, macOS, iOS, Android, Web" and applicationCategory as "BusinessApplication." The Organization block includes areaServed (UK, Ireland), sameAs links to 10 profiles, and a knowsAbout array covering 16 practice areas. However, the @id values reference leap.co.uk (a separate domain) rather than leaplegalsoftware.com, creating a cross-domain schema identity mismatch. The pricing page and feature pages also use leap.co.uk in their @id fields.

External Signals

DNS TXT records reveal OpenAI domain verification (openai-domain-verification=dv-...) and two Anthropic domain verifications — confirming LEAP has proactively registered with both OpenAI and Anthropic for crawler trust. The site also has cursor-domain-verification and launchdarkly-domain-verification TXT records. Despite this, no llms.txt exists (404). The sitemap contains 976 URLs across global, UK, NZ, and other regional subdirectories, but the root domain (/) immediately 301-redirects to /uk/ with no country detection, meaning crawlers hitting the bare domain see only the UK site.

Findings

1. Key internal pages return 404 Nuxt.js shells High

   Several pages linked from the homepage navigation, including /uk/company/, /uk/company/leadership/, /uk/about-us/, and /uk/thought-leadership-blog/, return 404 Nuxt.js shells with only the text 'Loading...'. These dead ends waste crawler budget and prevent AI systems from indexing company information and thought leadership content.

   What to change: Restore these pages with meaningful content or remove the links from navigation to avoid dead ends.

2. No llms.txt file published Medium

   The site returns a 404 for /llms.txt, missing an opportunity to guide AI crawlers to key pages and provide structured context about the company's AI products. Given the proactive domain verification with OpenAI and Anthropic, this omission is notable.

   What to change: Create an llms.txt file listing important pages (e.g., AI features, pricing, blog) and a brief summary of the company.

3. Schema @id values reference leap.co.uk instead of leaplegalsoftware.com Medium

   JSON-LD blocks on the homepage, pricing page, and feature pages use @id values pointing to leap.co.uk, a separate domain. This creates a cross-domain identity mismatch that may confuse knowledge graphs and dilute entity authority.

   What to change: Update all @id values to use the leaplegalsoftware.com domain consistently.

4. LLM knowledge gap on AI products High

   The cold LLM prior does not reference Matter AI, AI Prompts, or LawY integration, despite the site heavily promoting these AI features. The model's knowledge is stuck at the pre-AI-product era, indicating that AI systems have not yet indexed or associated these products with the brand.

   What to change: Ensure AI product pages are well-indexed and consider adding structured data (e.g., SoftwareApplication schema) for each AI tool. Publishing an llms.txt may also help.

5. Homepage redirects to /uk/ without region selector Low

   The root domain immediately 301-redirects to /uk/ with no country detection or region selector. Crawlers hitting the bare domain only see the UK site, potentially missing content for other regions (e.g., NZ, Australia) that exist in the sitemap.

   What to change: Implement a region selector or use hreflang annotations to serve appropriate regional content based on crawler IP or user preference.

6. Robots.txt lacks AI-bot-specific directives Low

   The robots.txt uses a single User-agent: * rule allowing everything, with no specific directives for AI crawlers like GPTBot or ClaudeBot. While this does not block access, it misses the chance to guide crawlers to important pages or away from low-value ones.

   What to change: Consider adding specific directives for AI crawlers to prioritize key pages (e.g., AI features, blog) and disallow broken pages.

7. Site omits TPG Capital ownership Low

   The cold LLM prior recalls the TPG Capital acquisition in 2018, but the site makes no mention of private equity ownership. This omission may reduce trust and transparency for AI systems that surface ownership information.

   What to change: Add a brief mention of the company's ownership or history on the About or Company page.

What's working

• All tested AI crawlers receive full access — All 11 tested AI crawlers receive a 200 status with full content, and robots.txt allows all bots without restriction. This ensures AI systems can index the site's content without barriers.
• OpenAI and Anthropic domain verification in place — DNS TXT records include openai-domain-verification and two anthropic-domain-verification entries, confirming proactive registration with major AI providers for crawler trust.
• Homepage includes comprehensive JSON-LD structured data — The homepage carries Organization, SoftwareApplication, WebSite, WebPage, and FAQPage schemas with detailed information including operating systems, practice areas, and social profiles.
• Site uses Nuxt.js SSR for readable content — The site runs on Netlify with Nuxt.js server-side rendering, meaning content is fully rendered and readable by crawlers, not hidden behind JavaScript execution.
• Sitemap contains 976 URLs across regions — The sitemap includes URLs for global, UK, NZ, and other regional subdirectories, providing a broad index of the site's content for crawlers.
• Dedicated pages for Matter AI and AI Document Drafting — The site has well-structured pages for AI products like Matter AI and AI Document Drafting, with descriptive content that can be indexed by AI crawlers.