spectrum.life — AI Site Grade

Analysis

Spectrum.Life — AI-Visibility Audit

The site's /robots.txt and /llms.txt both silently redirect to the homepage (HTTP 200 with full HTML), meaning no crawler directive file exists at either canonical path — a structural failure that leaves every AI bot without explicit guidance, despite the domain having verified OpenAI and Anthropic DNS TXT records.

Crawler Access

The main site (www.spectrum.life) runs on LiteSpeed behind no visible WAF, with no security headers (no HSTS, CSP, X-Frame-Options). Bot access is sharply fragmented: GPTBot, ClaudeBot, anthropic-ai, ChatGPT-User, OAI-SearchBot, and Bytespider all receive HTTP 403 (1,240-byte block page). Google-Extended, PerplexityBot, Perplexity-User, and Applebot-Extended get the full 200 (82KB) with real content. The pages.spectrum.life subdomain (hosted on Google Frontend) is fully open to every bot including GPTBot and ClaudeBot — a stark contrast. The /ai/ page is password-protected and noindex. The /sitemap.xml redirects to the homepage (zero URLs discoverable via standard sitemap protocol), though an HTML sitemap exists at /sitemap/ listing ~80+ pages.

Cold-Knowledge Gap

The LLM's prior knowledge describes Spectrum.Life as a "digital mental health and wellbeing platform" serving UK/Ireland employers and universities, with a 24/7 helpline and therapy access. The actual site reveals a far more ambitious B2B healthtech infrastructure play: a modular platform with SDK integration, AI-powered clinical review (AI Scribe, AI Review, Cara AI companion), a "MatchTech" health navigator, and recent expansion into Australia via three acquisitions (MindFit at Work, Valion Health, We Lysn). The cold model knows nothing about the Cara AI product, the Australia expansion (dated March 2026), the Bupa UK full-stack partnership, the 15 million member scale claim, or the CE Class IIa medical device pathway for Cara Plus. The gap is not subtle — the model describes a simple EAP provider; the site describes a multi-continent healthtech platform with AI clinical infrastructure.

Schema Posture

Only the homepage and AU landing page carry Organization + WebSite JSON-LD. Every sub-page examined — /about/, /what-we-do/, /sectors/insurer/, /insights/, the press release, the AI dev diary — has zero schema markup. The pages.spectrum.life subdomain (EAP page) is the only page with a Service schema type, correctly typed as Employee Assistance Programme. No FAQPage, Article, BreadcrumbList, Product, or MedicalWebPage schema exists anywhere. The homepage schema lists a single phone number and email but no foundingDate, numberOfEmployees, knowsAbout, or aggregateRating.

External Signals

Web search returned zero indexed results for site:spectrum.life, "spectrum.life" healthtech, or spectrum.life reviews — a complete absence of organic search visibility. The DNS TXT records confirm integrations with HubSpot, Salesforce, Zendesk, Barracuda, Pardot, and Tipalti. The Wayback Machine shows 280 historical captures of /robots.txt (which previously existed with a minimal Disallow: /wp-admin/ rule), confirming the current redirect is a regression. The site footer claims "2026 Spectrum Life" — a forward-dated copyright year that may confuse temporal signals.

Findings

1. Robots.txt redirects to homepage, providing no crawler directives High

   The canonical /robots.txt path returns HTTP 200 with full homepage HTML instead of a text file, leaving all bots without explicit crawl instructions. This is a regression from a previously functional robots.txt.

   What to change: Restore a valid robots.txt file at /robots.txt with appropriate directives for AI crawlers.

2. Sitemap.xml redirects to homepage, zero URLs discoverable High

   The standard /sitemap.xml path redirects to the homepage, making it impossible for crawlers to discover URLs via the sitemap protocol. An HTML sitemap exists at /sitemap/ but is not referenced in robots.txt.

   What to change: Create a valid XML sitemap and reference it in robots.txt.

3. LLMs.txt redirects to homepage, no AI guidance Medium

   The /llms.txt path also returns homepage HTML, missing an opportunity to provide structured AI-friendly content.

   What to change: Create an llms.txt file with a summary and links to key pages.

4. Major AI bots blocked on main site, open on subdomain High

   GPTBot, ClaudeBot, anthropic-ai, ChatGPT-User, OAI-SearchBot, and Bytespider receive HTTP 403 on www.spectrum.life, while Google-Extended, PerplexityBot, and Applebot-Extended get full content. The pages.spectrum.life subdomain is fully open to all bots.

   What to change: Allow all major AI crawlers on www.spectrum.life or provide equivalent content via the open subdomain.

5. Zero schema markup on most pages High

   Only the homepage and AU landing page have Organization+WebSite JSON-LD. All other pages examined lack any structured data, missing opportunities for rich results and AI understanding.

   What to change: Add relevant schema types (e.g., Article, Product, MedicalWebPage, BreadcrumbList) to all pages.

6. LLM prior knowledge severely outdated, missing key products and scale High

   The LLM describes Spectrum.Life as a simple EAP provider, but the site reveals a multi-continent healthtech platform with AI clinical tools (Cara AI, AI Scribe), Australia expansion, Bupa partnership, and 15 million member scale. This gap means AI-generated summaries will be inaccurate.

   What to change: Publish a comprehensive 'about' page with current products, partnerships, and scale, and ensure it is crawlable and well-structured.

7. No organic search presence detected High

   Web searches for site:spectrum.life, brand + healthtech, and reviews returned zero results, indicating the site is not indexed by major search engines.

   What to change: Investigate indexing issues, ensure sitemap and robots.txt are correct, and build backlinks.

8. Copyright year set to 2026, may confuse temporal signals Low

   The site footer displays '2026 Spectrum Life', which is a future date relative to the audit. This could confuse crawlers about content freshness.

   What to change: Update the copyright year to the current year or use a dynamic range.

9. AI page is password-protected and noindex Medium

   The /ai/ page requires authentication and has a noindex tag, making it invisible to crawlers despite containing relevant AI content.

   What to change: Make the AI page publicly accessible or move key content to a public page.

10. No security headers on main site Low

    The main site lacks HSTS, CSP, and X-Frame-Options headers, which may affect trust signals for some crawlers.

    What to change: Add security headers such as HSTS and CSP.

11. Homepage schema lacks key organization details Medium

    The Organization schema on the homepage includes only a phone and email, missing foundingDate, numberOfEmployees, knowsAbout, and aggregateRating.

    What to change: Expand the Organization schema with additional properties like foundingDate, numberOfEmployees, and description.

What's working

• Pages subdomain fully accessible to all AI crawlers — The pages.spectrum.life subdomain allows all 11 tested bots, including GPTBot and ClaudeBot, providing a channel for AI visibility.
• EAP page has correct Service schema — The Employee Assistance Programme page on pages.spectrum.life includes a Service schema typed as Employee Assistance Programme, aiding AI understanding.
• Homepage and AU page have Organization+WebSite schema — The homepage and AU landing page include JSON-LD for Organization and WebSite, providing basic structured data.
• HTML sitemap lists 80+ pages — An HTML sitemap at /sitemap/ provides a human-readable list of pages, aiding navigation and discovery.
• DNS TXT records verify integrations with major platforms — The domain has TXT records for HubSpot, Salesforce, Zendesk, and others, indicating established business infrastructure.
• AI in mental health dev diary provides transparency — A detailed blog post about AI in clinical mental health is publicly available, demonstrating thought leadership.