gosuperscript.com — AI Site Grade

Analysis

The single most non-obvious finding: Superscript's entire site is a Vue.js (Gridsome) SPA that delivers under 60 words of visible text per page via plain HTTP GET — yet every AI crawler is allowed full access with no blocks.

JS-Rendered Content Shell

Every page fetched — homepage (56 words), about-us (48 words), blog post (9 words), broker page (58 words), public liability page (53 words) — returns a near-empty HTML shell. The site is built on Gridsome v0.7.23, a Vue.js static-site generator, but the server-rendered output contains only navigation chrome and a single sentence of body text. The actual content (FAQ answers, product descriptions, blog body) is injected client-side by JavaScript. AI crawlers that do not execute JS — including GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, and ChatGPT-User — all receive a 200 status with the same empty shell. compare_bot_access confirmed every bot gets the same 117,994-byte response as a browser, but the visible text payload is negligible.

Crawler Access and robots.txt

The robots.txt explicitly names OAI-SearchBot and ChatGPT-User with permissive rules (disallow only /admin/ and /assets/gated/). No rules exist for GPTBot, ClaudeBot, Google-Extended, PerplexityBot, Bytespider, or Applebot-Extended, meaning they fall under the wildcard User-agent: * which allows all. No bot is blocked at the server level — Netlify returns 200 for all UAs. The /llms.txt returns a 404 (rendered as the site's Vue 404 page). The sitemap.xml contains 949 URLs covering deep industry-specific pages (wine-bar insurance, tattoo-artist insurance, etc.), but those pages suffer the same JS-rendering problem.

Cold-Knowledge Gap

The LLM's cold knowledge describes Superscript as a UK insurtech founded in 2015 by Cameron Shearer and Ben Rose, focused on SMEs/tech/startups, with flexible "insurance as a service" policies underwritten by Lloyd's. It also claims Superscript was acquired by Zego in 2022 — a fact that no public source on the site or in search results corroborates. The site itself makes no mention of Zego anywhere. The schema on the homepage lists the founding date as 2014-04-04 (not 2015), and the site presents itself as an independent FCA-authorised broker. The cold knowledge also omits that Superscript is a B Corp (the blog covers B Corp impact areas), that it offers a broker service for complex risks, and that it claims a 4.8 Trustpilot rating with 2,131 reviews (embedded in JSON-LD but not visible in rendered text).

Schema and Content Signals

The homepage carries rich JSON-LD: InsuranceAgency with aggregate rating (4.8/5 from 2,131 reviews), contact info, geo coordinates, opening hours, and sameAs links. Product pages like public-liability and cyber-insurance include FAQPage schema with detailed Q&A content — but this structured data is the only place AI crawlers can access the actual product information, since the visible text is empty. The FAQ content is well-formed and answer-rich, making it the primary vector for AI comprehension. Blog posts use Article schema with datePublished (one post dated 2026-03-17). No HowTo, Product, or Comparison schema types were detected.

External Signals

The site links to Trustpilot (uk.trustpilot.com/review/gosuperscript.com) and claims a 4.8 rating, but the Trustpilot page blocks automated fetches (403). DNS records show multiple verification TXT records (Google, Facebook, Stripe, Atlassian, Miro, Postman) indicating a broad SaaS integration stack. The site is hosted on Netlify with a permissive CSP (default-src * data: 'unsafe-inline' 'unsafe-eval'). No Wayback Machine snapshots were found, suggesting the domain may have been reconfigured or is relatively recent in its current form.

Findings

1. All pages return near-empty HTML shells to AI crawlers High

   Every page fetched returns under 60 words of visible text because content is injected client-side by Vue.js. AI crawlers that do not execute JavaScript receive a 200 status with the same empty shell.

   What to change: Implement server-side rendering (SSR) or pre-rendering for all pages so that AI crawlers receive fully rendered HTML with all content.

2. /llms.txt returns 404 Medium

   The /llms.txt endpoint returns a 404 error, missing an opportunity to provide AI crawlers with a curated summary of the site's content.

   What to change: Create an /llms.txt file that lists key pages and a brief description of the site's offerings for AI crawlers.

3. Cold knowledge falsely claims acquisition by Zego High

   LLM cold knowledge states Superscript was acquired by Zego in 2022, but no public source or site content corroborates this. The site presents itself as an independent FCA-authorised broker.

   What to change: Add a clear statement on the site about ownership and independence, and ensure schema markup reflects the correct legal entity.

4. Cold knowledge omits B Corp status and broker service Medium

   The LLM's cold knowledge does not mention that Superscript is a B Corp or that it offers a broker service for complex risks, missing key differentiators.

   What to change: Ensure B Corp and broker service information is prominently featured in visible text and schema markup on relevant pages.

5. Trustpilot review page blocks automated fetches Medium

   The Trustpilot page at uk.trustpilot.com/review/gosuperscript.com returns a 403 error, preventing AI crawlers from verifying the claimed 4.8 rating.

   What to change: Consider embedding Trustpilot reviews directly on the site or using a widget that is accessible to crawlers.

6. No Wayback Machine snapshots found Low

   The Wayback Machine has no snapshots of gosuperscript.com, which may indicate the domain was recently reconfigured or is relatively new in its current form, reducing historical credibility signals.

7. Sitemap has index=False directive Low

   The sitemap.xml contains an index=False directive, which may confuse some crawlers and reduce the likelihood of all pages being indexed.

   What to change: Remove the index=False directive from the sitemap or ensure it is correctly formatted.

What's working

• Homepage includes rich JSON-LD with aggregate rating and contact info — The homepage carries InsuranceAgency schema with aggregate rating (4.8/5 from 2,131 reviews), contact info, geo coordinates, and sameAs links, providing structured data for AI crawlers.
• Product pages include FAQPage schema with detailed Q&A — Pages like public-liability and cyber-insurance include FAQPage schema with well-formed Q&A content, making product information accessible to AI crawlers despite empty visible text.
• Robots.txt allows all AI crawlers full access — The robots.txt explicitly allows OAI-SearchBot and ChatGPT-User, and no rules block GPTBot, ClaudeBot, Google-Extended, or other major AI crawlers, ensuring no access restrictions.
• Sitemap contains 949 URLs covering deep industry-specific pages — The sitemap.xml lists 949 URLs including niche insurance categories (e.g., wine-bar, tattoo-artist), providing broad coverage for crawlers to discover content.
• Blog posts use Article schema with datePublished — Blog posts include Article schema with datePublished, helping AI crawlers understand content freshness and structure.