edmundscocktails.co.uk — AI Site Grade

Analysis

Edmunds Cocktails — AI-Visibility Audit

The live site at edmundscocktails.co.uk is entirely invisible to every AI crawler — and every browser — because Cloudflare's JS challenge gate blocks all requests before any content is served.

Crawler Access

Every AI crawler tested — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User — receives a 403 with a Cloudflare JS challenge wall. No bot gets past it. The robots.txt and llms.txt endpoints also return 403 (Cloudflare challenge HTML, not actual files). The site runs on Shopify (hosted at 23.227.38.65, Cloudflare-proxied) and uses a "Managed Challenge" Cloudflare WAF rule that blocks all programmatic access. There is no robots.txt file to inspect — it cannot be fetched. There is no sitemap accessible to crawlers. The site is a black box to the entire AI ecosystem.

Cold-Knowledge Gap

A frontier LLM queried cold about "Edmunds Cocktails UK" describes the brand as a London-based premium RTD cocktail brand founded by a former bartender named Edmunds, serving the UK hospitality sector and selling direct to consumers. The actual site tells a different story: the brand was founded in 2018 (or 2019) by two brothers from Suffolk (not London, not a single founder named Edmunds), operates from Bury St Edmunds, and positions itself as a small-batch, hand-bottled operation. The LLM hallucinated a founder biography and a London origin — the site itself has no person named "Edmunds" as a founder. The cold model also incorrectly lists "Classic Margarita" as a core product; the site sells a Kumquat Margarita, not a classic one.

Schema Posture

The site uses only Organization and WebSite schema types across all pages. Product pages — including the Espresso Martini page with 750+ words of detailed content (ABV, tasting notes, ingredients, instructions, story) — carry zero Product schema (@type: "Product" is absent). No Product, Offer, BreadcrumbList, FAQPage, or Review structured data exists anywhere. The Organization schema has empty sameAs fields for all social links except Instagram, and the url field points to a Wayback Machine URL rather than the canonical live domain.

External Signals

The brand has near-zero discoverable external footprint. Searches for reviews, Reddit mentions, press coverage, and Trustpilot listings returned no results. The site itself quotes Trustpilot testimonials but no Trustpilot profile surfaced in search. The DNS TXT records show Klaviyo (email marketing) and Google Workspace (email) integrations, plus a Facebook domain verification token — indicating active marketing operations, but none of this is visible to AI crawlers. The Instagram handle @edmunds_uk is referenced in schema but the account could not be verified via search.

Content Quality Trapped Behind the Wall

The site's actual content — visible only through Wayback Machine snapshots — is well-written, detailed, and AI-friendly: product pages with ABV percentages, ingredient sourcing stories, cocktail history, preparation instructions, and sustainability claims (100,000 trees planted via Ecologi). A blog/journal section exists with substantive articles. A subscription model ("Subscribe (-10%)") is offered. None of this content reaches AI crawlers. The site is a strong candidate for AI visibility that is being actively blocked by its own infrastructure.

Findings

1. Cloudflare JS challenge blocks all AI crawlers High

   Every AI crawler tested receives a 403 with a Cloudflare JS challenge wall, preventing any content from being served. The site is a black box to the entire AI ecosystem.

   What to change: Remove the Cloudflare JS challenge gate for AI crawler user agents, or configure a WAF bypass for verified bots such as GPTBot and ClaudeBot.

2. robots.txt endpoint returns 403 High

   The robots.txt file cannot be fetched because the Cloudflare challenge blocks all requests. No crawl rules are available to AI bots.

   What to change: Ensure robots.txt is served without Cloudflare challenge, and include directives to allow AI crawlers.

3. Sitemap inaccessible to crawlers High

   The sitemap endpoint returns a 403, preventing crawlers from discovering site URLs.

   What to change: Serve the sitemap without Cloudflare challenge and submit it to search engines.

4. llms.txt endpoint returns 403 Medium

   The llms.txt file cannot be fetched, preventing AI crawlers from discovering AI-friendly content guidance.

   What to change: Create and serve an llms.txt file without Cloudflare challenge.

5. Product pages lack Product schema High

   Product pages, such as the Espresso Martini page with 750+ words of detailed content, carry zero Product structured data. No Offer, BreadcrumbList, FAQPage, or Review schema exists anywhere.

   What to change: Add Product schema with name, description, offers, and aggregateRating to all product pages.

6. Organization schema has empty sameAs fields Medium

   The Organization schema's sameAs fields are empty except for Instagram, and the url field points to a Wayback Machine URL instead of the canonical live domain.

   What to change: Populate sameAs with all official social profiles and correct the url to the live domain.

7. LLM cold knowledge hallucinates brand details High

   A frontier LLM queried cold about the brand incorrectly describes it as a London-based brand founded by a former bartender named Edmunds, and lists a Classic Margarita as a core product. The actual brand is founded by two brothers from Suffolk and sells a Kumquat Margarita.

   What to change: Publish accurate brand information on the site and ensure it is crawlable by AI bots to correct the knowledge gap.

8. Near-zero external discoverable footprint Medium

   Searches for reviews, Reddit mentions, press coverage, and Trustpilot listings returned no results. The brand has minimal external signals to support AI visibility.

   What to change: Build external presence through press coverage, review platforms, and social media engagement.

9. No web search results for the domain Medium

   Web searches for the domain and brand name returned zero results, indicating poor indexing and visibility.

   What to change: Improve SEO and ensure the site is crawlable to get indexed by search engines.

10. High-quality content trapped behind Cloudflare wall High

    The site contains well-written, detailed product pages, blog articles, and sustainability claims that are AI-friendly but completely inaccessible to AI crawlers due to the Cloudflare challenge.

    What to change: Allow AI crawlers through the Cloudflare challenge to make this content visible.

What's working

• Detailed product content with ABV, tasting notes, and instructions — Product pages contain rich, well-written content including ABV percentages, ingredient sourcing stories, cocktail history, and preparation instructions, which is ideal for AI consumption if accessible.
• Sustainability page with 100,000 trees planted claim — The sustainability page details environmental efforts including planting 100,000 trees via Ecologi, which can be a positive signal for AI-driven brand analysis.
• Organization and WebSite schema present on all pages — The site uses Organization and WebSite schema types across all pages, providing basic structured data for brand identity.
• Active marketing integrations (Klaviyo, Google Workspace, Facebook) — DNS TXT records show integrations with Klaviyo for email marketing, Google Workspace, and Facebook domain verification, indicating active marketing operations.
• Subscription model with discount offered — The site offers a subscription model with a 10% discount, which can be highlighted in AI-generated product descriptions.