fullycharged.com — AI Site Grade

Analysis

Fully Charged — AI-Visibility Audit

The most critical finding is a brand-identity collision: the domain fullycharged.com hosts a premium eBike retailer, but the leading frontier LLM (cold, no retrieval) describes it as the *Fully Charged EV media channel* — a completely different entity founded by Robert Llewellyn. The site has no disambiguation, no structured data, and no llms.txt to correct this confusion.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended — receive a full 200 response with ~289KB of content from the homepage. No UA-based blocking exists. The site runs on BigCommerce (Stencil framework) behind Cloudflare, with X-Frame-Options: deny and Strict-Transport-Security: max-age=300. The robots.txt has a single User-agent: * rule with no AI-specific directives — no GPTBot, no ClaudeBot, no PerplexityBot sections. The llms.txt returns a 404 (BigCommerce 404 page). The sitemap.xml also returns 404. No sitemap means AI crawlers must discover product and category pages purely through internal links, which is suboptimal for a site with 100+ product SKUs across 8+ brand subdirectories.

Cold-Knowledge Gap

When asked cold about "fullycharged.com", the LLM describes a YouTube channel and EV news site founded by Robert Llewellyn (Red Dwarf actor), with 500k+ subscribers, covering EV reviews and industry trends since 2013. The actual site is a London-based eBike retailer stocking Riese & Muller, Moustache, Tern, Gocycle, Super73, Urban Arrow, and Benno — with six physical stores, a 12,000 sq/ft London Bridge HQ, and the "UK's largest eBike service centre." The LLM's secondary knowledge (when prompted with "Fully Charged eBike retailer London") is more accurate but still mentions "Cowboy" (a brand not stocked) and "Shoreditch, Clapham" showrooms (the site lists London Bridge, Brighton, Guildford, Silverstone, Cornwall, New Forest). The gap is severe: an AI engine retrieving the site cold would likely conflate it with the EV media brand, or fail to surface it for eBike purchase intent queries.

Schema Posture

The homepage and most category pages contain zero JSON-LD schema beyond a BreadcrumbList on sub-pages. No Organization, LocalBusiness, Product, FAQPage, or Review schema exists anywhere on the pages inspected. The homepage has an FAQ section (rendered as visible text) but no FAQPage markup. Product listing pages show prices (£3,499.00, £8,079.00) and stock status ("In stock", "Pre-order") in visible HTML but without Product schema. This is a significant missed signal for AI answer engines and Google Shopping graph integration.

Content Quality

The site has strong, original copy — detailed brand pages (Moustache, Riese & Muller), a family cargo bike guide, FAQ sections, and downloadable buyer's guides (PDF). The homepage word count is ~682 words of visible text, which is adequate. However, the reviews page is extremely thin (~140 words) with only four rider story headings and no embedded review widgets or review schema. The site claims "over 700 5-star Google reviews" but does not display or schema-markup any of them. Internal linking is thorough (25+ links per page), but the absence of a sitemap undermines crawl efficiency for the long tail of product pages.

Findings

1. Brand-identity collision with Fully Charged EV media channel High

   Cold LLM knowledge describes fullycharged.com as a YouTube channel and EV news site founded by Robert Llewellyn, not the London-based eBike retailer it actually is. The site has no disambiguation, structured data, or llms.txt to correct this confusion.

   What to change: Add an Organization schema with LocalBusiness subtype specifying eBike retail, and publish an llms.txt file that explicitly states the site is an eBike retailer, not a media channel.

2. Missing llms.txt file High

   The llms.txt endpoint returns a 404 (BigCommerce default page). This file would allow the site to provide explicit guidance to AI crawlers about its identity and content.

   What to change: Create an llms.txt file at the root that describes the site as a London-based eBike retailer and lists key pages.

3. Missing sitemap.xml High

   The sitemap.xml returns a 404. Without a sitemap, AI crawlers must discover product and category pages purely through internal links, which is suboptimal for a site with 100+ product SKUs.

   What to change: Generate and submit a sitemap.xml listing all product, category, and informational pages.

4. Missing Organization and LocalBusiness schema High

   No Organization or LocalBusiness JSON-LD schema exists on the homepage or any inspected page. This deprives AI engines of clear signals about the site's identity, location, and business type.

   What to change: Add Organization and LocalBusiness schema to the homepage with name, address, telephone, opening hours, and business type (eBike retailer).

5. Missing Product schema on listing pages High

   Product listing pages display prices and stock status in visible HTML but lack Product JSON-LD schema. This is a missed signal for AI answer engines and Google Shopping graph integration.

   What to change: Add Product schema to each product listing page with name, description, price, currency, availability, and brand.

6. Missing FAQPage schema on FAQ sections Medium

   The homepage and other pages contain FAQ sections rendered as visible text but without FAQPage markup. This prevents AI assistants from directly surfacing answers.

   What to change: Add FAQPage schema to pages with FAQ content, wrapping each question-answer pair.

7. Missing Review schema despite claiming 700+ Google reviews Medium

   The site claims 'over 700 5-star Google reviews' but does not display or schema-markup any reviews. The reviews page is thin (~140 words) with no embedded review widgets or review schema.

   What to change: Embed Google reviews or add Review schema with aggregate rating on the reviews page and homepage.

8. No AI-specific directives in robots.txt Low

   The robots.txt has a single User-agent: * rule with no sections for GPTBot, ClaudeBot, PerplexityBot, or other AI crawlers. While not blocking, this misses an opportunity to guide AI crawlers to important pages.

   What to change: Add specific sections for AI crawlers (e.g., GPTBot, ClaudeBot) with Allow directives for key pages.

9. Cold LLM knowledge inaccurately describes the site as EV media High

   When queried cold, the LLM describes fullycharged.com as a YouTube channel and EV news site, not an eBike retailer. Secondary knowledge is more accurate but still contains errors (e.g., mentions Cowboy brand not stocked, incorrect showroom locations).

   What to change: Implement structured data, llms.txt, and external citations (e.g., Wikipedia, Google Business Profile) to correct the LLM's knowledge.

10. Thin reviews page with no review content Medium

    The reviews page contains only ~140 words with four rider story headings and no actual review text, ratings, or schema. This is a missed trust signal for AI assistants.

    What to change: Expand the reviews page with full review text, ratings, and embed Google reviews or add Review schema.

What's working

• All major AI crawlers allowed access — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and other AI crawlers receive a full 200 response with no UA-based blocking.
• Strong original copy on brand and category pages — Detailed brand pages (Moustache, Riese & Muller), a family cargo bike guide, and downloadable buyer's guides provide rich, original content for AI crawlers to index.
• Adequate homepage word count — The homepage contains ~682 words of visible text, providing sufficient context for AI crawlers to understand the site's purpose.
• Thorough internal linking — Pages contain 25+ internal links, helping AI crawlers discover related content despite the missing sitemap.
• BreadcrumbList schema on subpages — Category and product pages include BreadcrumbList JSON-LD schema, aiding AI crawlers in understanding site structure.
• Test ride page with clear call-to-action — A dedicated test ride page provides detailed information and a form, which can be surfaced by AI assistants for purchase intent queries.