Schema markup is one of those topics that sounds dry on a conference agenda and ends up being one of the more practically useful sessions attendees sit through. That is because it is one of the few areas in SEO where there is a direct, observable line between implementation and result.

Why Schema Markup Is Worth a Keynote Slot

Most SEO tactics operate on a delay. You make a change, you wait weeks or months, you try to attribute a ranking movement to something you did. Schema is different. Deploy it correctly and you can see rich results, knowledge panel changes, and AI citation behavior shift within days. That makes it a productive topic for an audience that wants to understand cause and effect, not just accumulate abstract best practices.

It is also the area where most teams are leaving the most value on the table. Basic implementations are common. Custom, entity-level, relationship-aware schema that actually communicates what a business does, who it serves, and how it connects to authoritative entities is rare. That gap is where a well-structured keynote does its best work.

What the Keynote Covers

The session I typically deliver on schema markup moves through a few distinct layers.

  • What schema actually is and what it is not. A lot of teams have been told schema is for rich snippets. That is true but incomplete. Schema is a communication layer between your site and the systems — search engines, AI models, knowledge graphs — that are trying to understand what your content means, not just what it says.
  • The types that move results. Not every schema type is worth the same investment. The session covers which types are highest-leverage for most businesses and why: Article, FAQPage, HowTo, Product, LocalBusiness, Person, Organization, and the relationships between them.
  • Entity-based implementation. I cover how to treat schema as entity markup, not just page decoration. That means thinking about who you are, what you do, who you serve, and how those relationships are expressed in structured data.
  • Common implementation errors. Incomplete markup, missing required fields, contradictions between schema and visible content, and misuse of types. I walk through real examples.
  • Validation and monitoring. How to use Google’s Rich Results Test, Schema Markup Validator, and Search Console to confirm what is working and catch what is not.

The Entity Graph Problem Most Teams Ignore

The part of schema implementation that most teams skip is relationship markup. They will put an Organization type on the homepage, an Article type on blog posts, and stop there. What they miss is that Google and other systems are trying to build a graph of entities and relationships — who published this, what organization do they belong to, what topics do they cover, what other authoritative entities are they connected to.

A Person schema that includes sameAs references to a verified LinkedIn profile, a Wikidata entry, or a Google Knowledge Panel anchor does something that basic Article schema does not: it tells the system this is a real, identifiable entity with an established presence. That connection between entity disambiguation and schema implementation is something I spend significant time on in this session because it is where the ROI is, and it is where most implementations stop well short of what is possible.

The relationship between Organization and Person schema — how they reference each other, how they express the connection between a brand and the people who run it — is a specific area where I walk through before-and-after examples. The difference in knowledge panel behavior and AI citation frequency between a site with flat schema and a site with properly nested, related entities is not subtle.

Who This Session Is For

The schema keynote works best for audiences that include SEO practitioners, content strategists, and web developers who are responsible for site performance. It can be calibrated for a beginner-friendly overview or a technical deep dive depending on the room. For a mixed audience, I usually split the session: conceptual framing for the first half, implementation specifics for the second, with clear signposting so people know which parts are most relevant to their role.

What the Audience Leaves With

Concretely: a clearer mental model of how search engines and AI systems read structured data, an understanding of which schema types to prioritize, and a practical checklist for auditing what their current implementation is missing. I do not hand out slides and call it done. The goal is that someone in the room goes back to their desk and does something different the following week.

What Schema Cannot Do

I am also direct about limitations in this session, because schema is one of those topics that gets oversold at conferences. Schema markup does not create ranking signals the way links or content quality do. Deploying FAQPage schema on a page that does not answer questions clearly will not produce rich results and will not improve rankings. Schema communicates what your content means — it does not substitute for the content itself.

There is also a ceiling on what rich results can do for click-through rate. A site ranking fifth with FAQ rich results may outperform a site ranking third without them in some queries, but it will not overcome a position-one result with a strong title and compelling meta description. I include this because it helps audiences allocate effort accurately. Schema is high-leverage in the right context, and lower-leverage when other fundamentals are not in place.

How This Connects to AI Search

One reason schema has gotten more relevant, not less, is the rise of AI-powered search. The systems behind AI Overviews, ChatGPT, and Perplexity do not just read text. They interpret entities and relationships. A well-structured schema implementation helps those systems understand who you are and what you know, which feeds directly into whether you get cited as a source. I cover this connection explicitly in the session because it is where a lot of audiences have questions they did not realize they had.

The practical implication: if you want your business to appear as a source in AI-generated answers — not just in the blue-link results — entity-level schema is part of the foundation. It is not the only factor, but it is one of the few you can control directly. Most sites I audit have not touched this. The keynote gives a concrete picture of what a stronger implementation looks like and what it would take to deploy it.

I have been building and teaching custom schema strategy through Salterra and SEO University for years. The keynote draws directly from that applied work, not from aggregated case studies.

If you are considering this session for your event, the SEO speaker page covers logistics, format options, and how to start a conversation. If your event leans technical, the post on a technical SEO talk may also be worth a look.