Answer engine optimization is not a rebrand of SEO. It is a different problem with some overlapping tools. The audience for a talk on AEO is usually a mix of people who have heard the term and are not sure what to do with it, and people who have been trying things and are not sure if any of it is working. The session is designed to be useful to both.

What AEO Actually Means

When someone asks ChatGPT a question about your industry, or runs a search that triggers a Google AI Overview, or checks Perplexity for a quick answer, they are not reading ten blue links. They are reading a synthesized response. The system pulling that response is making decisions about what sources to trust, what content to excerpt, and what entities to cite.

AEO is the practice of making your content — and your entity presence online — easier for those systems to use accurately. It is partly a content problem, partly a technical problem, and partly a trust and authority problem. The keynote covers all three layers, and I’m explicit with audiences about which layer any given tactic actually belongs to, because a lot of the confusion I run into comes from people applying a content fix to what is actually a trust problem, or vice versa.

What the Talk Covers

  • How AI search systems read content. The difference between keyword matching and semantic retrieval. How these systems identify authoritative sources versus content that looks authoritative but lacks substance.
  • Entity authority and why it matters more than domain authority. Being clearly identifiable as an entity — a person, a business, a topic — with consistent signals across the web is a core input to AI citation behavior. I cover what that looks like in practice: consistent naming, consistent facts about who you are across every platform that mentions you, and a clear knowledge graph presence rather than conflicting bios scattered across a dozen profiles.
  • Structured data as a trust signal. Schema markup is not just for rich results anymore. It is one of the primary signals AI systems use to understand what a page is about, who created it, and whether the information can be attributed. I tie this directly into the AEO conversation because most audiences have not connected those dots yet — they think of schema as an SEO tactic from five years ago rather than the backbone of how modern systems verify a claim.
  • Content structure for AI readability. How to write and format content so that an AI system can extract a coherent answer from it. This is different from writing for human readers, though the best approach serves both. I show real before-and-after examples of a page restructured so the direct answer sits in an extractable block near the top, with supporting nuance and caveats afterward instead of buried in the middle of a paragraph.
  • What you can measure and what you cannot. AEO attribution is imperfect right now. I give an honest picture of what is trackable — referral traffic from AI platforms where it exists, branded search lift, direct citation checks by manually querying the tools — what requires inference, and what to ignore in the noise. I would rather a room leave with realistic expectations than a dashboard promise I can’t back up.

What This Is Not

This is not a talk about gaming AI systems or prompt engineering your way into citations. I am skeptical of that framing and I say so plainly. What actually works is a longer game: being genuinely authoritative on a topic, having consistent entity signals, and structuring content so that an AI system can confidently attribute information to you rather than paraphrase around you.

If you are looking for quick tricks, this is not the session for that. If you are trying to understand what is actually shifting in search and what to prioritize over the next twelve to eighteen months, it is a useful hour. I’m also candid about the tactics that circulated in 2024 and 2025 that have already stopped working — stuffing FAQ schema with unrelated questions, or publishing thin ‘definitive guide’ pages purely to trigger citation, both of which AI systems have gotten measurably better at discounting.

Who the Talk Is Built For

The AEO keynote works well for marketing teams, SEO practitioners, and content strategists who are trying to understand what to do about AI search without getting swept up in hype cycles. It also works for executives who need a ground-level explanation of why the traffic patterns they are seeing in analytics have started behaving differently — a real, common scenario now is a leadership team staring at a quarter of declining organic sessions with no obvious cause, when the actual answer is that a meaningful share of what used to be a click is now an AI-generated answer that never sends the visit at all.

I calibrate the technical depth to the audience. For a room with developers, I go further into the structured data and entity graph side, including specific JSON-LD patterns. For a marketing leadership room, I focus more on content strategy and measurement framing, and spend less time in the weeds of markup syntax.

A Practical Example I Use in the Room

One example that consistently lands: two businesses in the same industry, similar domain authority, similar content volume. One gets cited regularly when someone asks an AI assistant a question in that category. The other almost never does, despite arguably better-written content. The difference, when you dig in, usually isn’t the writing quality — it’s that the cited business has a consistent, well-documented entity presence: a Person or Organization schema that actually matches their bio everywhere else online, clear authorship on their content, and a site structure where the direct answer to a likely question sits in one identifiable place rather than scattered across three pages. The uncited business has good content buried inside a page structure that gives an AI system nothing clean to extract and attribute. I walk the room through that comparison side by side, because it makes the abstract idea of ‘entity trust’ concrete in a way that a slide of bullet points doesn’t.

Questions I Get Asked Most Often

The most common question after this keynote is some version of ‘how do I know if I’m being cited at all.’ The honest answer is that it takes manual checking right now — running your own likely customer questions through ChatGPT, Perplexity, and Google’s AI Overview periodically and noting whether you show up, and if so, how accurately. There isn’t yet a mature analytics tool that gives you this the way Search Console gives you organic ranking data, though that’s changing. The second most common question is whether AEO work cannibalizes traditional SEO effort. It largely doesn’t — the entity and structured data work that improves AEO citation also tends to improve traditional search performance, because both systems are ultimately trying to identify trustworthy, well-organized sources.

How This Connects to My Work

Most of my applied work at Salterra over the past couple of years has involved schema markup and entity optimization, which turns out to be the technical foundation of AEO. I also cover this material in the live cohorts at SEO University, so it is actively tested against real-world feedback, not just assembled from articles. Cohort students bring in their own sites, we check what’s actually getting cited and what isn’t, and the keynote content gets revised based on what that testing shows rather than staying static from one year to the next.

For a look at the broader speaker context and how to get a session on the calendar, see the SEO speaker page. The post on a schema markup keynote goes deeper on the structured data side if that is the area your audience needs most.