creamos.com — AI Site Grade

Analysis

Cold-Knowledge Gap

A frontier LLM queried cold about "Creamos creative agency" returned zero specific knowledge — no awareness of the brand, its services, clients, or market position. The model stated it could not confirm the agency's existence. This is the central finding: a site that brands itself as "The Anglo-Hispanic Creative Powerhouse" is entirely invisible to the AI knowledge layer that powers chat-based discovery, AI search, and agentic retrieval.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended — receive a 200 status with full content (101 KB, identical to browser baseline). No UA-based blocking exists. The robots.txt contains only HubSpot CMS boilerplate disallows (/_hcms/preview/, /hs/manage-preferences/) and mentions zero AI bots by name. No llms.txt exists (404). The site runs on HubSpot CMS behind Cloudflare DNS (NS: jeremy.ns.cloudflare.com, may.ns.cloudflare.com). The technical gate is wide open — AI crawlers can ingest everything — but there is nothing guiding them to the most important content.

Content & Schema Posture

The homepage (332 words) and service pages (Evolution, Acceleration) are thin — each under 250 words of body text. The blog is a ghost town: /creamos-blog returns "No blogs found", /en-us/blog contains Lorem Ipsum placeholder text, and /es-ar/blog is also Lorem Ipsum. A stale test page (/thomas-test-blog) with "Test blog for test" is live and indexed in the sitemap. The sitemap also includes /portfolio-testing and /orne-test-2 which 404. JSON-LD schema is present as Organization on every page but is static and identical — same description, same contact point, same single LinkedIn sameAs URL. The FAQ page uses proper FAQPage schema with 12 questions, which is the strongest structured data on the site. Work detail pages use BlogPosting schema (misapplied — these are portfolio case studies, not blog posts). No Product, Service, ItemList, or CollectionPage schema exists for the portfolio or service offerings.

External Signals

Web searches for "Creamos agency London creative reviews", "Creamos creative agency clients MTV Lloyds", and site:creamos.com all returned zero results. No press coverage, no review sites, no Reddit threads, no external mentions were discoverable. The only external link on the entire site is a single LinkedIn profile (ar.linkedin.com/company/cr34m05). The FAQ page mentions clients including MTV and Lloyds Bank — but these names appear nowhere else on the site, have no dedicated case studies, and are not verifiable through any external source. The site claims a London office (86-90 Paul Street) and a New York presence, but the LocalBusiness schema on the contact page only lists the London address.

Surprising Findings

The site's most differentiated claim — the "Anglo-Hispanic" model of UK/US strategy teams paired with an Argentina-based creative studio for cost efficiency — is mentioned only in the homepage H1 and the FAQ. This is the single strongest positioning signal for AI engines to latch onto, yet it appears nowhere in meta descriptions, schema descriptions, or dedicated landing copy. The "Good add Ventures" venture studio (where all profits are reinvested) is another distinctive fact buried in the FAQ and absent from the homepage. The site has no case study PDFs, no downloadable credentials deck, no team page, no about page — the /about URL was not found in the sitemap. AI crawlers can access everything but will find a thin, schema-static site with an empty blog, placeholder pages in the sitemap, and a brand story that the cold-knowledge layer has never heard of.

Findings

1. Cold LLM query returns zero knowledge of Creamos High

   A frontier LLM queried about 'Creamos creative agency' could not confirm the agency's existence, indicating the brand is entirely absent from AI training data and knowledge bases.

   What to change: Build authoritative external references (press releases, guest articles, listings) and publish rich, unique content on the site to increase the likelihood of inclusion in AI training corpora.

2. No llms.txt file to guide AI crawlers Medium

   The site returns a 404 for llms.txt, missing an opportunity to explicitly direct AI crawlers to the most important pages and provide a concise brand summary.

   What to change: Create an llms.txt file listing key pages (home, work, FAQ, services) and a brief brand description.

3. Homepage and service pages have thin body text Medium

   The homepage (332 words) and service pages Evolution (243 words) and Acceleration (206 words) contain very little substantive content, limiting the information AI crawlers can extract.

   What to change: Expand each service page to at least 500 words of unique, descriptive copy covering methodology, benefits, and differentiators.

4. Blog contains no real content, only placeholder text High

   The blog pages (/creamos-blog, /en-us/blog, /es-ar/blog) show 'No blogs found' or Lorem Ipsum placeholder text, providing no value to AI crawlers or human visitors.

   What to change: Remove placeholder blog pages or populate them with genuine articles about the agency's work, industry insights, and thought leadership.

5. Sitemap includes stale test pages and 404 URLs Medium

   The sitemap contains /thomas-test-blog (live test content), /portfolio-testing (404), and /orne-test-2 (404), wasting crawl budget and signaling poor site maintenance.

   What to change: Remove all test and 404 URLs from the sitemap; ensure only production-ready pages are included.

6. Organization schema is static and identical across all pages Medium

   Every page uses the same Organization JSON-LD with identical description, contact point, and single LinkedIn sameAs URL, missing opportunities to provide page-specific context.

   What to change: Customize the schema on each page to reflect the specific content (e.g., service schema on service pages, FAQPage on FAQ, BlogPosting on actual blog posts).

7. Work case studies use BlogPosting schema instead of more appropriate types Low

   Portfolio pages like /work/choc-factory and /work/napp are marked up as BlogPosting, which is semantically incorrect for case studies; this can confuse AI parsers.

   What to change: Replace BlogPosting schema with CreativeWork or Article schema, and add properties like datePublished, author, and description.

8. No Service or Product schema for service offerings Medium

   The site offers services like 'Evolution' and 'Acceleration' but does not use Service schema to describe them, reducing the chance of appearing in AI-generated service comparisons.

   What to change: Add Service schema to each service page with name, description, provider, and areaServed.

9. No external mentions, backlinks, or search presence detected High

   Web searches for the agency name, client names, and site:creamos.com returned zero results, indicating no press coverage, reviews, or indexed pages in Google.

   What to change: Launch a PR and content marketing campaign to earn mentions on industry blogs, directories, and social media; ensure the site is indexed by Google.

10. Major client names (MTV, Lloyds) are unverifiable externally Medium

    The FAQ mentions MTV and Lloyds Bank as clients, but no case studies, logos, or external references support these claims, which may reduce trustworthiness for AI systems.

    What to change: Create dedicated case study pages for each major client with permission, including project details, results, and client testimonials.

11. No about page or team page exists Medium

    The site lacks an about page and team page, which are critical for establishing credibility and providing background information to AI crawlers and visitors.

    What to change: Create an about page detailing the agency's history, mission, and the Anglo-Hispanic model, and a team page with bios and photos.

12. Key differentiators (Anglo-Hispanic model, Good add Ventures) are only in FAQ Medium

    The unique Anglo-Hispanic model and the venture studio are mentioned only in the FAQ, not on the homepage or service pages, missing opportunities for AI to surface them.

    What to change: Prominently feature the Anglo-Hispanic model and Good add Ventures on the homepage and dedicated service pages with supporting copy.

What's working

• All major AI crawlers are allowed and served full content — GPTBot, ClaudeBot, PerplexityBot, and others receive 200 status with full HTML, ensuring no technical blocking prevents AI ingestion.
• FAQ page uses correct FAQPage schema with 12 questions — The /frequently-asked-questions page is marked up with FAQPage schema, which is well-structured and can be used by AI for direct answers.
• Sitemap is accessible and contains 80 URLs — The sitemap at /sitemap.xml returns 200 and lists 80 URLs, providing a clear crawl path for search engines and AI crawlers.
• Work pages contain substantive case study content — Pages like /work/choc-factory (325 words) and /work/napp (359 words) provide detailed project descriptions, which are valuable for AI to understand the agency's capabilities.
• Contact page includes LocalBusiness schema with address — The contact page has LocalBusiness schema listing the London office address, which helps AI verify physical presence.