The short version: Your CEO's profile page is either structured as a knowledge graph node (something AI and search systems can recognize, connect, and cite) or it's just a page with their name on it. You can check in a few clicks, start improving it in a few hours, and use it as a foundation for your broader content strategy.

Start Here: The DIY Audit

This is easier to work through if you run the check first and read the rest of the article second. Two windows isn't a bad approach.

Step 1. Go to your company website. Navigate to your leadership or executive team page.

Step 2. Open any executive profile. Start with the CEO or named founder.

Step 3. Copy that profile URL. Paste it into schema.org/validator. Run the test.

Step 4. In Google, search "[CEO name]" site:[yoursite.com]. Look at every result. How many link to or tag that person as a named entity, not just mention their name?

For many organizations, what comes back from the validator is almost nothing useful. Maybe a generic WebPage entry. No Person. No name, jobTitle, or sameAs references. No structural connection between that human and the organization they lead.

Think about what that means. This person is often the face of your organization. They have experience, affiliations, memberships, certifications, articles they've written, boards they serve on, social profiles they actively use. Adding even a small amount of structure to that profile page creates a hub of interconnectedness. Done right, it helps your entire digital ecosystem, from your website on out.

This is the opportunity this article is focused on. And it exists for nearly every organization, from the small local shop to the regional nonprofit to the multinational.

A note on execution time: How long it takes to act on this depends on your platform. If you or your team control the schema layer directly, a first pass can take 10-30 minutes. If schema requires a developer or IT approval, you're looking at a ticket queue. The planning phase described below is worth doing regardless. Knowing what the technical opportunity looks like can also guide content decisions before the ticket gets written.

Why This Matters Right Now

When someone asks ChatGPT, Perplexity, or Google's AI Overview "who does AI search consulting in Chicago?", the system isn't ranking pages. It's inferring an answer from a knowledge graph, a web of connected entities assembled from structured data across the web. For more traditional search, authority signals work similarly. This is the foundation of what's known as EEAT (Experience, Expertise, Authoritativeness, Trustworthiness).

Your organization is either a recognized node in that graph, with a named, credentialed human attached to it, or it isn't.

What makes something a node versus a page:

A node has a canonical identifier: a URL that definitively represents a specific entity. It has typed properties that tell machines what the entity is: name, jobTitle, worksFor, knowsAbout, sameAs. It has connections: the Organization references the Person; the Person references external authorities like LinkedIn and professional publications.

An executive profile with a photo, a two-paragraph bio, and an unlinked LinkedIn icon is not a node. It's a page with a name on it. That's a real distinction, and one you just checked for in about 15 minutes.

What SEO and AI Systems Actually Do When Someone Searches Your Space

When someone searches "who handles AI marketing strategy for mid-size manufacturers in Chicago," no single answer lives on a single page. The system fans out, triggering multiple layered searches, cross-referencing claims across sources before assembling a response.

Think of it as triangulation. The AI isn't just looking for a match; it's looking for confirmation. Does the name on this firm's website match the person cited in that industry publication? Does their LinkedIn show the same title and specialization? Does the structure on their profile page connect them to the organization they claim to lead?

When those signals are inconsistent, or absent, confidence drops. The organization gets passed over for one the system can validate more cleanly.

Here's what makes this different from traditional SEO: many of those cross-referencing queries have very little or no conventional search volume. There's no keyword to rank for. You can't optimize your way into that validation process. Either your entity is structured clearly enough to be confirmed across sources, or it isn't.

Google stated at Search Central events in late 2025 that structured data is "critical for modern search features" and "very precise, and good for AI systems." That's the actual input format the system is built around, not a search ranking trick.

Recent data confirms the effect. Ahrefs examined 4 million AI Overview citations and found only 38% came from pages ranking in the top 10 for the same query. Nearly a third came from pages beyond position 100 entirely. (Ahrefs notes their parsing methodology improved since an earlier study, so the numbers aren't directly comparable over time, but the directional finding is clear.) Ranking well still matters. It doesn't determine citation.

What "Fixed" Looks Like, and What I Built

I use the 2 Find Marketing website to practice what I tell clients to do. It's a small site, which actually makes it a useful test case. Most recently I've been working through the entity architecture piece. Here's what the updated version looks like in practice. The technical implementation (JSON-LD, code examples) is covered in the follow-up post, or you can view source on the site directly.

The components:

Organization schema at the root level, with explicit founder and employee references pointing to named Person entities
Person schema at a canonical profile URL: a page that definitively represents me as an entity, with jobTitle, worksFor, knowsAbout, alumniOf, and sameAs links to LinkedIn and other authoritative profiles
ProfilePage schema wrapping that URL, establishing the structural relationship between the page and the entity it describes
Article schema with author attribution linked to the Person entity, so every piece of authored content connects back to the author as a recognized node

The content layer on the site serves as a single source of truth that feeds both the schema output and the UI. One source, one edit point, consistent representation everywhere.

This article is itself the first Article schema instance on the site. It's linked to my Person entity. That's not incidental. It's the architecture doing what it's supposed to do.

One honest note: I'm not reporting significant AI citation numbers from a single-practitioner site. I'll document directional before/after results as the site matures. What I have right now is a clear baseline and a sense of what I'm measuring against.

The next layer is ensuring consistent entity representation across every page on your site where your leadership team is mentioned. That's what creates the triangulation effect that reinforces AI confidence in citing your organization, and it's the subject of the next post in this series.

The AI Prompt to Plan Your Own Implementation

Or: what you can do today.

Use this prompt in ChatGPT, Claude, or Gemini. It produces a Person schema property plan in about 15 minutes, regardless of your technical background. Bring the output to your developer, or evaluate it against your CMS capabilities yourself.

"I'm planning to add Person schema markup to my executive's profile page. Here is the profile URL: [URL]. Their name is [Name], title is [Title], they studied at [University], their expertise areas are [list], and their LinkedIn/professional profiles are [links]. Please recommend which Person schema properties to implement and explain why each one supports SEO and AI discoverability and EEAT signals. Do not write the JSON-LD, just the property plan. Also provide updated content architecture for the page, considering the human aspect of the profile. Output should include a short brief for copywriters to update the content as needed."

Ask for the property plan first, not the code. The plan is what you need to evaluate effort and scope. The code follows once you've confirmed the approach. That code post is next in this series.

Measurement: What You Can (and Can't) Track

Measurement here deserves its own treatment, and that's a future article. For now, the honest short version.

This isn't a "my exec page ranking improved" kind of change. You're building foundational entity infrastructure. The effects are distributed across your content ecosystem, not concentrated in any single metric.

What you can measure:

Schema validation is the fastest, most binary check. Your Person entity either validates and is internally consistent, or it doesn't. schema.org/validator and Google's Rich Results Test both give you a clear yes/no. Start here. It's immediate and free.

Bing Webmaster Tools AI Performance Dashboard is worth a baseline snapshot before and after you make changes. Bing has been more transparent than Google about AI-influenced traffic signals. Treat it as directional, not definitive.

GA4 AI referral traffic. Set up a segment for traffic from AI platforms (ChatGPT, Perplexity, etc.) before you change anything. Volume will be thin for most sites. What you're watching for is trend, not absolute numbers.

Google Search Console with advanced filtering. RegEx queries can surface how entity-related terms are performing in traditional search. Not a direct AI visibility metric, but entity strengthening has measurable downstream effects on regular search as well.

Manual baseline testing has value as a qualitative check: ask ChatGPT, Perplexity, and Gemini about your space before and after, and screenshot the results. Session and model variability makes it hard to draw firm conclusions, but a meaningful shift will show up if it happens.

What you can't measure yet:

There is no Google Search Console for AI citation. Tools claiming to measure AI visibility (Semrush, Profound, Conductor, BrightEdge) sample a narrow slice of queries against a narrow slice of platforms. They're directional, not definitive.

No controlled study exists showing "we added Person schema and AI citation frequency increased by X%." The performance data on schema's broader impact doesn't isolate Person schema specifically. The reasoning is sound. The chain of evidence has a gap. Anyone claiming precise attribution here should be treated skeptically.

Saying that clearly is itself an EEAT signal.

This is roughly where link building measurement sat in 2008: directional signals are real, causality isn't yet provable. That's not a reason to skip a 30-minute check when the downside risk is near zero.

The Honest Frame

This is foundational work, not a quick win. It's also a bit of toe-dipping into content that's often under-optimized and surprisingly low-effort to update.

The gap between implementing entity architecture correctly and seeing it pay off in search and AI citations is real. Some changes surface quickly, others take weeks or months, and for a smaller or newer site it can take longer still.

What doesn't change: machines need structure to understand context. The cost of doing it right is low. The cost of not doing it compounds as search and AI systems mature and the knowledge graph becomes the default index for generated answers.

The audit costs you 30 minutes. The gap it reveals is almost certainly there. Addressing it doesn't require a new platform, a new vendor, or a new strategy. It requires structured data about the people behind your organization, and most organizations haven't done it yet.

That's why you start here.

Ted Fay is a marketing consultant and founder of 2 Find Marketing. He writes about AI search, structured data, and practical implementation for marketing leaders. Away from work, he rides with Chicago Rando.