everybodyagency.com — AI Site Grade

Analysis

Everybody Agency — AI-Visibility Audit

The cold LLM knowledge about "Everybody Agency" describes a completely different company — a New York creative agency founded by ex-Droga5 executives that worked on Peloton and Liquid Death — while the actual site is a UK-headquartered health and wellness digital marketing agency with offices in two continents, a B Corp certification, and a growing library of AI-visibility thought leadership.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Applebot-Extended, Bytespider, anthropic-ai — receive a full 200 response with identical byte size (~75KB) to browser traffic. No UA-based blocking, no Cloudflare challenge, no JS shell. The site runs on nginx hosted at 165.232.101.150 (DigitalOcean) with no CDN or WAF layer detectable. The robots.txt is minimal: it only disallows /wp/wp-admin/ and has no AI-bot-specific rules — a neutral posture that lets all crawlers through but also misses the opportunity to guide them toward high-value content. /llms.txt returns a 404 (WordPress error page), meaning no AI-friendly content map exists.

Cold-Knowledge Gap

The LLM prior is a complete identity mismatch. The model describes a NYC branding agency (Tim Leake, Mike Abbaticchio, Peloton refresh, Liquid Death) that has zero overlap with the actual business. The real Everybody Agency is a healthcare and wellness digital marketing agency with a CEO named Ben Myall, a leadership team including John Jobst, Courtney Cesari, and Courtney Maletic, and a Certified B Corp status. This is a critical gap: any AI engine answering "what is Everybody Agency?" without live retrieval will describe the wrong company entirely.

Content & Schema Posture

The site uses Yoast-generated JSON-LD with only WebPage, BreadcrumbList, WebSite, and SearchAction schema types. No Organization, LocalBusiness, ProfessionalService, FAQPage, HowTo, or Article schema exists anywhere on the site — including on insight articles and project case studies where Article schema would be expected. The homepage has only 189 words of visible text with a single H1 and six H2s. The insight articles are substantive (1,000–2,000 words each) and cover AI visibility, LLM retrieval, and healthcare content strategy with genuine depth, but they lack FAQ schema even when answering direct questions. The site has a dual-language structure (en-GB and en-US) with mirrored content paths, but the US site uses /en-us/ subdirectory URLs while the UK site uses root — a clean hreflang setup but no hreflang tags detected in the fetched pages.

External Signals

Web searches returned zero indexed results for the brand name, its domain, or its service keywords across multiple queries. No reviews, no Reddit threads, no press mentions, no Clutch/G2/DesignRush listings surfaced. The only external presence detected is a LinkedIn company page, an Instagram account, and a YouTube channel — all linked from the footer. The B Corp certification links to a bcorporation.net profile. This near-total absence of external citations means AI engines have almost no third-party signal to corroborate the brand's positioning, making the cold-knowledge confusion even more damaging.

Surprising Findings

The site publishes authoritative, well-researched content on AI visibility for healthcare brands (including an article by CEO Ben Myall on structuring content for LLM retrieval, published May 2026), yet the site itself has no llms.txt, no AI-bot guidance in robots.txt, no Organization schema, and no structured data that would help LLMs understand who the company is. The dateModified field shows "2026-01-29" and "2026-05-21" — dates in the future relative to typical knowledge cutoffs — suggesting either a WordPress date bug or intentional forward-dating. The /careers/ page exists in the sitemap but the about page states "We have no current openings." The /insights-2/ slug in the navigation suggests a renamed or duplicated insights section with the old slug still linked.

Findings

1. LLM prior describes a different company entirely High

   The cold LLM knowledge for 'Everybody Agency' describes a New York creative agency (ex-Droga5, Peloton, Liquid Death) with no relation to the actual UK-based healthcare marketing agency. Any AI answering without live retrieval will misrepresent the brand.

   What to change: Publish an Organization schema with the correct name, description, founding date, CEO, B Corp status, and location. Ensure the homepage and about page contain clear, unambiguous text that distinguishes the agency from the similarly-named New York firm.

2. No Organization or LocalBusiness schema on any page High

   The site uses only WebPage, BreadcrumbList, WebSite, and SearchAction schema. No Organization, LocalBusiness, or ProfessionalService schema exists, preventing LLMs from extracting the company's name, address, industry, or certifications.

   What to change: Add Organization schema (with sameAs for LinkedIn, Instagram, YouTube, B Corp profile) to the homepage and about page. Add LocalBusiness schema if a physical address is listed.

3. No /llms.txt file for AI crawler guidance Medium

   The /llms.txt endpoint returns a 404 WordPress error page. The site provides no AI-friendly content map, missing an opportunity to guide LLMs to high-value pages like insight articles and case studies.

   What to change: Create an /llms.txt file listing key pages (about, services, insights, projects) with brief descriptions to help AI crawlers prioritize content.

4. No indexed external citations for the brand High

   Web searches for the brand name, domain, and service keywords returned zero results. No reviews, press mentions, or directory listings were found, leaving AI engines without third-party signals to corroborate the brand's positioning.

   What to change: Build citations on industry directories (Clutch, DesignRush, G2), secure press mentions, and encourage client reviews on Google and LinkedIn to create a web of external references.

5. Insight articles lack Article schema Medium

   The site's substantive insight articles (1,000–2,000 words) do not include Article or NewsArticle schema, reducing their chance of being surfaced as rich results in AI-generated answers.

   What to change: Add Article schema with headline, datePublished, dateModified, author, and image to all insight articles.

6. No FAQ schema on pages answering direct questions Medium

   Insight articles that answer specific questions (e.g., 'How to measure LLM visibility') do not use FAQPage schema, missing a chance to appear in AI-generated FAQ snippets.

   What to change: Identify sections that pose and answer questions, and wrap them in FAQPage schema with Question and Answer properties.

7. Content shows future dates (2026) in dateModified Low

   Several pages display dateModified values of '2026-01-29' and '2026-05-21', which are in the future relative to typical knowledge cutoffs. This may confuse AI crawlers or indicate a WordPress date bug.

   What to change: Correct the dateModified fields to actual publication or last-updated dates. Investigate the WordPress date configuration to prevent future dates.

8. Navigation links to /insights-2/ slug Low

   The site's navigation includes a link to /insights-2/, suggesting a renamed or duplicated insights section with the old slug still active. This can cause confusion for crawlers and users.

   What to change: Remove the /insights-2/ slug from navigation and set up a redirect to the correct /insights/ section.

9. Dual-language site missing hreflang tags Medium

   The site has en-GB and en-US versions with mirrored content, but no hreflang tags were detected in the fetched pages. This may cause LLMs to serve the wrong language variant.

   What to change: Add hreflang tags to all pages linking to the corresponding en-GB and en-US versions.

10. Robots.txt lacks AI-bot guidance Low

    The robots.txt only disallows /wp/wp-admin/ and has no directives for AI crawlers. While this allows full access, it misses the opportunity to prioritize high-value content or prevent crawling of low-value pages.

    What to change: Add crawl-delay and allow/disallow rules for AI bots, and consider listing high-value paths like /insights/ and /projects/.

What's working

• All major AI crawlers receive full 200 responses — GPTBot, ClaudeBot, PerplexityBot, and other AI crawlers all receive the same content as browser users, with no blocking or JS shell. This ensures the site's content is fully indexable by LLMs.
• Publishes authoritative content on AI visibility for healthcare — The site features in-depth articles on LLM visibility and content structuring, written by the CEO, demonstrating thought leadership that can be cited by AI engines.
• Clean dual-language URL structure with subdirectory separation — The en-GB and en-US versions use distinct URL paths (root vs /en-us/), making it easy for crawlers to distinguish language variants.
• B Corp certification linked from footer — The site links to its B Corp profile on bcorporation.net, providing a verifiable third-party credential that can be used by LLMs to establish trust.
• Sitemap index available with 71 URLs — The sitemap index at /sitemap_index.xml lists 71 URLs, helping crawlers discover all pages efficiently.