ligo.co.uk — AI Site Grade

Analysis

Cloudflare JS Challenge Blocks All AI Crawlers From Any Content

The live site at ligo.co.uk returns HTTP 403 with a Cloudflare JS challenge to every single user-agent tested — including GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, and a plain browser. No bot receives any real content. The robots.txt and llms.txt endpoints also return the same 403 JS challenge wall, meaning neither file exists in a readable form. The site is hosted on Cloudflare (proxied A records to 104.26.12.48, 104.26.13.48, 172.67.75.90) with Shopify verification codes in DNS TXT records confirming a Shopify backend.

Cold-Knowledge Gap: AI Models Know Nothing About a 20-Year-Old Retailer

The LLM knowledge query returned zero verifiable information about ligo.co.uk — no awareness of its product categories, its 20-year history, its Glasgow base, or its brand name "liGo." The model explicitly stated it could not confirm what the site does. This is a complete knowledge vacuum for a business that, per its own About page, has been operating since 2003, sells across telephones, camping, binoculars, kitchen appliances, DAB radios, and baby monitors, and claims to be an antidote to big-box retailers. The gap between the site's actual substance and what AI models know cold is total.

Schema Posture: Zero Structured Data on Any Archived Page

The Wayback Machine snapshots of the homepage, telephones collection page, and About page all show zero JSON-LD schema types — no Organization, no Product, no BreadcrumbList, no FAQPage, no LocalBusiness. The site has FAQ content, product listings with prices and ratings, and a clear business identity, but none of this is exposed in machine-readable structured data. Heading structure is flat (H1s on category pages, H2s and H3s for products), but no semantic markup aids AI extraction.

External Signals: No Indexed Mentions Anywhere

Web searches for ligo.co.uk, "liGo.co.uk", and the brand name "liGo" combined with its product categories returned zero indexed results across general web search, Trustpilot, Reddit, or social media. The site has Google, Facebook, Klaviyo, Ahrefs, and Shopify verification TXT records, indicating active SEO and marketing tooling, yet none of that investment has produced discoverable external citations. The About page quotes customer testimonials ("Exceptional service", "Exceptionally slick service") but those reviews exist only on the site itself — no third-party review platform footprint was found.

Findings

1. Cloudflare JS challenge blocks all AI crawlers from accessing any content High

   The live site returns HTTP 403 with a Cloudflare JS challenge to every tested user-agent, including GPTBot, ClaudeBot, PerplexityBot, and Google-Extended. No bot receives any real content. The robots.txt and llms.txt endpoints also return the same 403 wall.

   What to change: Configure Cloudflare to allow AI crawler user-agents (e.g., GPTBot, ClaudeBot) through the JS challenge, or serve a static version of the site to bots. Alternatively, move the site behind a reverse proxy that permits known bot IP ranges.

2. Robots.txt and llms.txt are inaccessible due to Cloudflare JS challenge High

   Both /robots.txt and /llms.txt return HTTP 403 with a Cloudflare JS challenge, meaning crawlers cannot read any instructions or AI-specific guidance.

   What to change: Ensure /robots.txt and /llms.txt are served without JS challenge, or host them on a separate domain that is not Cloudflare-proxied.

3. AI models have zero knowledge of ligo.co.uk despite 20-year history High

   LLM knowledge queries returned no verifiable information about the site, its product categories, its 20-year history, or its Glasgow base. The model explicitly stated it could not confirm what the site does.

   What to change: Improve AI visibility by allowing crawlers, adding structured data, and building external citations (reviews, backlinks, social media presence).

4. Zero JSON-LD structured data on homepage, collection, or about pages High

   Archived snapshots of the homepage, telephones collection, and About page contain no JSON-LD schema types such as Organization, Product, BreadcrumbList, or LocalBusiness. This prevents AI models from extracting business identity, product details, or site structure.

   What to change: Add JSON-LD structured data for Organization, Product, BreadcrumbList, and LocalBusiness to all relevant pages.

5. No indexed external mentions on web search, Trustpilot, Reddit, or social media Medium

   Web searches for the domain, brand name, and product categories returned zero indexed results. No third-party review platform footprint was found despite customer testimonials on the site.

   What to change: Encourage customers to leave reviews on Trustpilot, Google, and other platforms. Build backlinks through partnerships, PR, or content marketing.

6. Flat heading structure with no semantic markup for AI extraction Low

   Archived pages show H1s on category pages and H2s/H3s for products, but no semantic markup (e.g., schema.org) aids AI extraction of product details or site hierarchy.

   What to change: Add structured data (Product, BreadcrumbList) and improve heading hierarchy with descriptive H2s and H3s.

What's working

• Shopify backend with SEO and marketing verification codes — DNS TXT records include verification codes for Google, Facebook, Klaviyo, Ahrefs, and Shopify, indicating active SEO and marketing tooling investment.
• Detailed About page with 20-year history and customer testimonials — The About page describes the company's 20-year history, Glasgow base, product range, and includes customer testimonials, providing rich content for AI if accessible.
• Product listings with prices and ratings on collection pages — The telephones collection page displays product names, prices, and star ratings, which are valuable for AI extraction if structured data were added.