jacobbailey.com — AI Site Grade

Analysis

The cold-knowledge model conflates jacobbailey.com with a completely different agency — a London-founded 1978 firm with Sydney and Singapore offices, Rolls-Royce and Harrods clients, and a "Brand as a Business" methodology. The actual site describes a Suffolk-headquartered agency with 32 staff, three offices (Ipswich, London, New York), and zero external search footprint.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, anthropic-ai, Applebot-Extended, Bytespider, ChatGPT-User — receive a 200 status with the same full HTML content as a browser (35,501 bytes). No UA-based blocking, no JS shell, no Cloudflare challenge. The robots.txt contains no AI-bot-specific rules; only a catch-all User-agent: * disallowing /cpresources/, /vendor/, /.env, and /cache/. The llms.txt returns a 404 (an HTML error page, not a plain-text 404). The site runs on Craft CMS behind a Kubernetes pod (via: 1.1 jacob-bailey-production-6976dcc48f-6244s:8080) with no CDN or WAF visible in headers.

Cold-Knowledge Gap

The LLM's prior describes a completely different entity: a London-founded 1978 agency with Sydney/Singapore offices, Rolls-Royce and Harrods clients, and a "Brand as a Business" approach. The actual jacobbailey.com is a Suffolk-headquartered agency (HQ: 1 Woodbridge Road, Ipswich, IP4 2EA, company number 3619009) with 32 crew members, offices in Ipswich, London, and New York. Clients include Otis, Mastercard, Durex, Scholl, BT, Sodexo, Greater Anglia, and Countryside — not Rolls-Royce or Harrods. The site's methodology is "Connecting Relevance through Data, Technology & Creativity" with an "i5" process, not "Brand as a Business." The model knows nothing about the actual domain when queried directly by URL.

Schema Posture

Every page carries the same minimal JSON-LD: a WebPage with Organization identity and a BreadcrumbList. The Organization schema is stub-level — it contains only @type: Organization and name: "Jacob Bailey" with no url, logo, sameAs, address, telephone, description, or foundingDate properties. No LocalBusiness, ProfessionalService, or CreativeWork schemas appear anywhere. The homepage has no FAQPage, HowTo, Product, or ItemList schema despite featuring a portfolio grid and service categories. The blog uses Article schema only implicitly via WebPage — no BlogPosting or NewsArticle type.

External Signals

The site has a vanishingly small external footprint. DuckDuckGo searches for the agency name combined with "Ipswich," "Suffolk," "creative agency," "marketing," and specific client names (Otis, Mastercard) return zero results. A site:jacobbailey.com search also returns zero. The Wayback Machine has a single snapshot from March 2026 — no historical archive before that date. The only external links on the site point to LinkedIn, Instagram, Craft CMS partner page, and three unrelated domains (032design.co.uk, destinationcore.com, infotex.uk). The site has two Google Search Console verification TXT records and two Microsoft verification records, suggesting it was submitted to search engines, but no organic search presence is detectable.

Content Freshness

The homepage dateModified is 2025-10-02 and the copyright footer reads "2026," but the blog's most recent post is March 2024 (the Borderlands book launch). The "2023 Round-up" post is the second-most-recent. No content published in 2025 or 2026 exists on the blog. The vacancies page states "We're not actively recruiting at the moment" with no open roles listed. The team page shows 20 staff members — a small operation for an agency claiming international reach.

Findings

1. LLM prior describes a different agency entirely High

   The cold-knowledge model conflates jacobbailey.com with a London-founded 1978 agency with Sydney/Singapore offices and Rolls-Royce clients. The actual site is a Suffolk-headquartered agency with 32 staff, offices in Ipswich, London, and New York, and clients like Otis and Mastercard.

   What to change: Publish a detailed Organization schema with address, founding date, logo, sameAs, and description. Submit the site to major AI training data sources and directories to correct the model's prior.

2. No organic search presence detected High

   DuckDuckGo searches for the agency name, location, and clients return zero results. A site:jacobbailey.com search also returns zero. The Wayback Machine has only one snapshot from March 2026.

   What to change: Implement a comprehensive SEO strategy including backlink building, local citations, and content marketing. Ensure the site is indexed by Google and Bing.

3. Organization schema is stub-level with minimal properties High

   Every page carries the same minimal JSON-LD with only @type: Organization and name: 'Jacob Bailey'. Missing url, logo, sameAs, address, telephone, description, and foundingDate.

   What to change: Expand the Organization schema to include url, logo, sameAs (LinkedIn, Instagram), address, telephone, description, and foundingDate. Add LocalBusiness or ProfessionalService schema as appropriate.

4. Blog uses WebPage schema instead of BlogPosting or Article Medium

   The blog page and individual posts use only WebPage schema. No BlogPosting, NewsArticle, or Article type is present, reducing AI understanding of content structure.

   What to change: Add BlogPosting schema to blog posts with headline, datePublished, author, and image properties.

5. Homepage portfolio and service categories lack structured data Medium

   The homepage features a portfolio grid and service categories but has no FAQPage, HowTo, Product, or ItemList schema. This limits AI extraction of key offerings.

   What to change: Add ItemList schema for portfolio items and service categories. Consider FAQPage schema for common client questions.

6. llms.txt returns 404 HTML error page Medium

   The llms.txt file is missing, returning a 404 HTML page instead of a plain-text 404 or a valid file. This prevents AI crawlers from discovering structured site information.

   What to change: Create a valid llms.txt file with a summary of the site, key pages, and links to sitemap and schema.

7. Blog has no content published since March 2024 Medium

   The most recent blog post is from March 2024, and the second-most-recent is a 2023 round-up. No content from 2025 or 2026 exists, indicating the blog is inactive.

   What to change: Publish fresh content regularly to demonstrate site activity and provide material for AI crawlers to index.

8. Vacancies page states no active recruitment Low

   The vacancies page indicates the agency is not actively recruiting, which may signal a lack of growth or resources.

   What to change: Remove or update the vacancies page if recruitment status changes, or add a careers section with company culture information.

9. Site lacks CDN and WAF, served directly from Kubernetes pod Low

   The site is served directly from a Kubernetes pod without a CDN or Web Application Firewall visible in headers. This may affect performance and security.

   What to change: Consider using a CDN for improved global performance and a WAF for security.

What's working

• All major AI crawlers receive full HTML content — All tested AI crawlers (GPTBot, ClaudeBot, PerplexityBot, etc.) receive a 200 status with the same full HTML content as a browser. No UA-based blocking or JS shell is present.
• Robots.txt has no AI-bot-specific disallow rules — The robots.txt file contains only a catch-all rule disallowing non-content directories. No AI crawlers are blocked.
• Sitemap is accessible and contains 80 URLs — The sitemap is available and lists 80 URLs, providing a clear structure for crawlers.
• BreadcrumbList schema is present on every page — Each page includes a BreadcrumbList JSON-LD, helping AI understand site navigation hierarchy.
• Google and Bing Search Console verification records exist — DNS TXT records indicate the site has been verified with Google Search Console and Microsoft Bing, suggesting prior search engine submission.
• Site runs on Craft CMS, a modern and flexible platform — Craft CMS provides good content management capabilities and is generally SEO-friendly.