fdry.com — AI Site Grade

Analysis

FDRY's AI-visibility posture is technically open but strategically hollow — every AI crawler gets a 200 with full content, yet the brand has near-zero external footprint and the cold LLM knowledge about it is completely wrong.

Crawler Access

All eleven AI crawlers tested (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User) receive a 200 with identical byte-size content as a browser visit. No UA-based blocking, no Cloudflare challenge, no JS shell. The robots.txt is a bare Yoast default with no AI-bot-specific rules — it disallows /?s=, /wp-json/, and /search/ for * but does not restrict any AI crawler. The llms.txt URL redirects to the homepage (effectively a 404 for the convention). The site runs on Cloudflare (NS: marlowe/piotr) with WordPress/PHP 8.1 and PleskLin, serving 141 KB HTML pages with full text content.

Cold-Knowledge Gap

Asked cold, a frontier LLM describes FDRY as a "London-based digital agency working with luxury and lifestyle brands" citing The Connaught hotel, Harrods, and premium hospitality names — none of which appear anywhere on the actual site. The real FDRY (Foundry Digital Limited) is an ecommerce-focused agency building Shopify/WooCommerce/Adobe Commerce stores for brands like NU:YU (clinical skincare), Comoletics, Hart Medical, A Weathered Penny (jewellery), and NAOS. The LLM hallucinated a luxury hospitality portfolio that does not exist. The gap is total: the model's prior describes a different agency entirely.

Schema Posture

JSON-LD is present on every page via Yoast SEO's @graph structure, including Organization, WebSite, WebPage, BreadcrumbList, and Article types. The Organization schema correctly names "Foundry Digital Limited" with alternateName "FDRY - Foundry". However, no LocalBusiness, Service, FAQPage, Product, Review, or AggregateRating schema exists anywhere — surprising for an agency that sells ecommerce builds and SEO services and claims "40-300% traffic increases." The SEO service page has FAQ content but no FAQPage markup. No ItemList schema on the work portfolio page despite listing ~50 client projects.

External Signals

FDRY has near-zero discoverable external footprint. Web searches for reviews, Reddit mentions, press coverage, or client references return no results. The only off-domain signals are a bare-bones corporate holding page at foundrydigital.co.uk (which redirects to fdry.com for services) and social profiles (Instagram, LinkedIn, Twitter/X, Facebook). No Clutch, GoodFirms, Trustpilot, or G2 profile appears. No industry publication mentions. The blog contains 80+ posts dating back to 2020, including recent 2025 content on AI and ecommerce trends, but these generate no visible backlink or citation signal.

Content & Structure

The homepage and service pages are text-rich (744-3,077 words) with clear H1/H2 hierarchy. The site explicitly positions around "GEO" (Generative Engine Optimisation) and AI search marketing — a forward-looking pitch that aligns with AI-visibility strategy. Yet the work portfolio page (/work/) returns only 15 words of visible text — a JS-heavy image grid with no alt text or descriptive content for crawlers. Individual case study pages (e.g., NU:YU) are text-rich but lack Product or CreativeWork schema. The blog covers timely topics (AI in 2025, SMS vs email, GEO marketing) but has no author schema beyond a single managing director.

Findings

1. LLM cold knowledge describes a different agency with luxury hospitality clients High

   A frontier LLM describes FDRY as a London agency working with The Connaught hotel, Harrods, and premium hospitality brands — none of which appear on the actual site. The real FDRY is an ecommerce-focused agency building Shopify/WooCommerce stores for brands like NU:YU and Comoletics.

   What to change: Publish authoritative content (case studies, client logos, testimonials) with structured data to correct the LLM's knowledge. Submit the site to LLM training data sources and ensure schema markup accurately reflects the agency's actual clients and services.

2. Near-zero discoverable external footprint across reviews, press, and social signals High

   Web searches for reviews, Reddit mentions, press coverage, or client references return no results. No Clutch, GoodFirms, Trustpilot, or G2 profile appears. No industry publication mentions. The blog generates no visible backlink or citation signal.

   What to change: Build external signals by claiming profiles on Clutch, GoodFirms, and Trustpilot; encourage client reviews; pitch case studies to industry publications; and engage in relevant online communities.

3. llms.txt URL redirects to homepage, effectively a 404 for the convention Medium

   The llms.txt URL (https://fdry.com/llms.txt) redirects to the homepage instead of serving a plain-text file. This breaks the llms.txt convention that AI crawlers use to discover allowed content.

   What to change: Serve a valid llms.txt file at the root URL listing key pages and summaries for AI crawlers.

4. Work portfolio page renders as a JS-heavy image grid with only 15 words of visible text High

   The /work/ page returns only 15 words of visible text — a JavaScript-dependent image grid with no alt text or descriptive content for crawlers. AI crawlers that do not execute JS will see an empty page.

   What to change: Add server-rendered text descriptions for each project, include alt text on images, and ensure the page is crawlable without JavaScript.

5. Insights page renders as a JS-heavy grid with only 13 words of visible text High

   The /insights/ page returns only 13 words of visible text — a JavaScript-dependent grid with no descriptive content for crawlers. AI crawlers that do not execute JS will see an empty page.

   What to change: Add server-rendered text summaries for each blog post, include alt text on images, and ensure the page is crawlable without JavaScript.

6. SEO service page has FAQ content but no FAQPage schema markup Medium

   The SEO agency service page contains FAQ content but lacks FAQPage structured data. This prevents AI crawlers from extracting Q&A pairs for featured snippets or AI answers.

   What to change: Add FAQPage schema markup to the FAQ section on the SEO service page.

7. Work portfolio page lacks ItemList schema for client projects Medium

   The /work/ page lists approximately 50 client projects but has no ItemList or CreativeWork schema markup. This reduces the chance that AI crawlers understand the portfolio as a structured collection.

   What to change: Add ItemList schema to the work portfolio page, with each project as a ListItem containing name, description, and URL.

8. No LocalBusiness schema despite being a physical agency in London Medium

   The site has Organization schema but no LocalBusiness schema, which is appropriate for a physical agency with a London address. This limits local AI visibility.

   What to change: Add LocalBusiness schema with address, phone, and opening hours to the homepage and contact page.

9. No Review or AggregateRating schema despite claiming traffic increases Low

   The site claims '40-300% traffic increases' but has no Review or AggregateRating schema to substantiate these claims with structured data. This reduces credibility for AI crawlers.

   What to change: Add AggregateRating schema on service pages if client reviews are available, or add Review schema for testimonials.

10. Blog posts lack author schema beyond a single managing director Low

    The blog covers timely topics but has no author schema beyond a single managing director. This limits the ability of AI crawlers to attribute content to specific authors.

    What to change: Add author schema (Person type) to each blog post with name and URL.

11. Robots.txt has no AI-bot-specific rules, leaving default access Low

    The robots.txt is a bare Yoast default with no AI-bot-specific rules. While this allows access, it misses the opportunity to guide AI crawlers to the most important pages.

    What to change: Add specific rules for AI crawlers to allow all content, and optionally add a sitemap directive.

12. Blog generates no visible backlink or citation signal despite 80+ posts Medium

    The blog contains 80+ posts dating back to 2020, including recent 2025 content on AI and ecommerce trends, but these generate no visible backlink or citation signal in web searches.

    What to change: Promote blog content through social media, guest posting, and outreach to industry publications to build backlinks.

What's working

• All 11 tested AI crawlers receive full content with no blocking — Every AI crawler tested (GPTBot, ClaudeBot, PerplexityBot, etc.) receives a 200 with identical byte-size content as a browser visit. No UA-based blocking, no Cloudflare challenge, no JS shell.
• JSON-LD schema present on every page via Yoast SEO — Every page includes Yoast SEO's @graph structure with Organization, WebSite, WebPage, BreadcrumbList, and Article types. The Organization schema correctly names 'Foundry Digital Limited' with alternateName 'FDRY - Foundry'.
• Homepage and service pages are text-rich with clear heading hierarchy — The homepage and service pages contain 744-3,077 words of text with clear H1/H2 hierarchy, providing substantial content for AI crawlers to index.
• Site explicitly positions around Generative Engine Optimisation and AI search marketing — The site explicitly positions around 'GEO' (Generative Engine Optimisation) and AI search marketing — a forward-looking pitch that aligns with AI-visibility strategy.
• Individual case study pages are text-rich with detailed descriptions — Individual case study pages (e.g., NU:YU) are text-rich with detailed descriptions, providing substantial content for AI crawlers.
• Blog covers timely AI and ecommerce topics with recent 2025 posts — The blog covers timely topics (AI in 2025, SMS vs email, GEO marketing) with recent 2025 content, demonstrating relevance to current trends.
• Sitemap index available with 80 URLs — The sitemap index at /sitemap_index.xml is accessible and contains 80 URLs, helping crawlers discover all pages.
• Cloudflare protection without blocking AI crawlers — The site uses Cloudflare for protection but does not challenge or block AI crawlers, maintaining accessibility.