AI Site Grade
cheil.uk — AI Site Grade
Cheil UK's subdomain redirect, missing sitemap, and zero structured data leave AI crawlers with no authoritative content to index, while cold LLM knowledge relies on outdated third-party data.
Cheil UK's AI visibility is critically limited by a redirect to a global domain with no sitemap, no structured data, and duplicate page metadata, causing AI engines to rely on stale external information.
- Findings
- 12
- Evidence checks
- 31
- Completed
- 30 May 2026
Analysis
Cheil UK: A global agency subdomain with no AI surface
The most consequential finding is that cheil.uk is not a standalone site — it immediately 301-redirects to www.cheil.com/uk, a subfolder of the global Cheil Worldwide domain, and the global domain has no sitemap, no llms.txt, and no structured data on any page tested.
Crawler Access
All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended, anthropic-ai — receive a 200 with full HTML content on every page tested. No UA-based blocking exists. The site runs on AWS (CloudFront via 18.165.x.x IPs) with HSTS and X-Frame-Options: DENY but no CDN/WAF that discriminates by user-agent. The global robots.txt at www.cheil.com/robots.txt is a bare User-agent: * Allow: / — permissive but also completely silent on AI-specific directives. The llms.txt endpoint does not exist (redirects to homepage). No sitemap is published anywhere on the domain (returns 404).
Content & Schema
Every page on the site — homepage, about, work, services, news, career — shares the identical <title> and <meta description>: "Cheil UK | Creative Agency Experts" / "Cheil UK is a creative solutions agency..." No page has unique title tags. Zero JSON-LD schema of any type was found on any page. No Organization, WebSite, BreadcrumbList, FAQPage, or CreativeWork schema exists. The homepage delivers only ~160 words of visible text. Work detail pages contain substantive case-study content (e.g., Try Galaxy with 73.63M downloads, 17 awards) but lack any structured markup to help AI engines parse the results, metrics, or client relationships. The news section has blog-style articles with dates and external links to Campaign and The Drum, but no article schema.
Cold-Knowledge Gap
A frontier LLM queried cold about Cheil UK recalls it as the London arm of Cheil Worldwide (Samsung-founded, 1973), names Samsung as the flagship client, and cites specific work: Samsung KX in King's Cross, a 2018 Cannes Lions Grand Prix for "Samsung O!U", and a 2023 Campaign "Creative Agency of the Year" shortlist. The model also lists Coca-Cola, Adidas, and Nestlé as clients. The actual site does not mention Coca-Cola, Adidas, or Nestlé anywhere — the client list visible on the site is Samsung, Priority Pass, Philip Morris International, BVI, NIVEA, and VR Therapies. The cold model knows about a Cannes Grand Prix win; the site does not mention it. The model knows about Samsung KX; the site does not mention it. The site's own positioning ("+1 strategy", "Be Bold. Be Brave. Be First.") is entirely absent from the model's prior. This gap means AI engines answering queries about Cheil UK will draw on stale or incorrect third-party data rather than the agency's own narrative.
External Signals
DuckDuckGo web search returned zero results for multiple queries targeting Cheil UK — an anomaly that may indicate low domain authority or indexing issues. The site does link externally to Campaign, The Drum, and LBB Online for press coverage, but those articles were not independently verified in this audit. The career page links to a LinkedIn company page. No reviews, Reddit threads, or third-party directory listings were surfaced.
Surprising Details
The cheil.uk domain has a Wayback snapshot from March 2026 (future-dated), suggesting the redirect may be recent or the snapshot is misdated. Every page includes Korean-language UI elements ("공유하기 닫기 URL복사") in the rendered HTML, indicating the site template was built for a Korean-speaking parent company and not fully localised. The work portfolio is heavily Samsung-dominated (6 of 9 featured projects), which aligns with the cold model's prior but creates a single-client dependency risk for AI-generated brand summaries.
Findings
No sitemap published on www.cheil.com High
The global domain www.cheil.com returns a 404 for its sitemap, preventing AI crawlers from discovering all pages efficiently.
What to change: Publish a valid sitemap at www.cheil.com/sitemap.xml listing all subfolder pages including /uk/.
No llms.txt file available Medium
The llms.txt endpoint does not exist on either cheil.uk or www.cheil.com, missing an opportunity to guide AI crawlers to key content.
What to change: Create an llms.txt file at www.cheil.com/llms.txt listing important pages like /uk/about/, /uk/work/, /uk/news/.
All pages share identical title and meta description High
Every page on the site uses the same <title> and <meta description>, preventing AI crawlers from distinguishing page topics and reducing search relevance.
What to change: Assign unique, descriptive title tags and meta descriptions to each page reflecting its specific content.
No JSON-LD schema on any page High
Zero structured data of any type was found on the site, meaning AI engines cannot extract entity relationships, organization details, or content types.
What to change: Add Organization, WebSite, BreadcrumbList, and CreativeWork schema to relevant pages, especially work case studies.
Cold LLM knowledge contradicts site content High
A frontier LLM recalls Cheil UK as having clients like Coca-Cola, Adidas, and Nestlé, and mentions a Cannes Grand Prix win and Samsung KX, none of which appear on the actual site. The site's own narrative is absent from AI knowledge.
What to change: Publish authoritative content on the site covering key achievements, client list, and unique positioning, and use structured data to reinforce it.
Zero web search results for Cheil UK queries High
Multiple DuckDuckGo searches for Cheil UK returned no results, indicating very low domain authority or indexing issues that limit AI visibility.
What to change: Improve SEO fundamentals, build backlinks from reputable industry sites, and ensure the site is properly indexed.
cheil.uk redirects to www.cheil.com/uk Medium
The cheil.uk domain immediately 301-redirects to a subfolder of the global domain, which may confuse crawlers and dilute authority.
What to change: Ensure the redirect is properly handled and that the global domain's sitemap includes all /uk/ pages.
Korean-language UI elements present in HTML Low
The rendered HTML includes Korean phrases like '공유하기 닫기 URL복사', indicating incomplete localization that may confuse AI crawlers parsing content.
What to change: Remove or localize all non-English UI elements in the English version of the site.
Homepage delivers only 160 words of text Medium
The homepage has very little textual content, limiting the information AI crawlers can extract about the agency's services and value proposition.
What to change: Expand homepage copy to at least 500 words covering services, differentiators, and key achievements.
News articles lack Article schema Medium
News pages contain blog-style content with dates and external links but no Article or NewsArticle structured data, reducing their discoverability in AI-driven news features.
What to change: Add Article or NewsArticle schema to all news detail pages.
No BreadcrumbList schema on any page Low
The site does not implement breadcrumb structured data, making it harder for AI to understand site hierarchy and page relationships.
What to change: Add BreadcrumbList schema to all pages reflecting the site's navigation structure.
Work portfolio heavily dominated by Samsung Low
6 of 9 featured work projects are Samsung-related, creating a single-client dependency risk for AI-generated brand summaries that may overemphasize Samsung.
What to change: Diversify the featured work portfolio to showcase a broader client base.
What's working
- All major AI crawlers allowed with full HTML content — Every tested AI crawler receives a 200 response with full HTML, ensuring no bot is blocked from accessing the site.
- Global robots.txt allows all crawlers — The robots.txt at www.cheil.com/robots.txt has a permissive Allow: / rule, ensuring no accidental blocking.
- Work detail pages contain rich case study content — Pages like the Try Galaxy case study include detailed metrics (73.63M downloads, 17 awards) that provide valuable content for AI extraction.
- Active news section with recent articles — The news section contains multiple articles with dates and external links, indicating regular content updates.
- Career page links to LinkedIn company page — The career page provides a link to the Cheil UK LinkedIn page, helping AI engines connect the site to social profiles.
- HSTS and security headers in place — The site uses HSTS and X-Frame-Options: DENY, providing a secure connection that AI crawlers prefer.
Track cheil.uk across AI search
This is one snapshot. Open the interactive report to inspect evidence, or grade another site free.