AI SEO Webinar

Learn how AI is redefining search and what it means for visibility, content strategy, and performance: Optimizing for LLM-driven search results Entity-first optimization Creating content that supports AI retrieval

00:00:00

Andrew Husted: So, as FSM tries to be the best partner to all of you, we’ve been doing a lot of research on what this means for the search landscape, because it has a lot of implications. And the approach looking forward needs to shift, and figuring out how it should shift has been a big focus for the last many months over here at FSM. We want to do that leg work so you’re prepared and have that partner you can rely on to execute these things. I actually see a few names out there who are kind of piloting our AI SEO programs already—so thanks for your early adoption of what’s going on in this space. We’ve got a lot to share, so I’m going to get us going here, because there’s really a lot to cover. And I want everybody to know we’ve got an audience where some are more sophisticated in this space than others.

00:00:58

Andrew Husted: And so we’re going to make it a very approachable presentation today—so you can understand where things were, where we’ve come, what things are looking like they’re going to be, and how everything works together. If we break that down into an agenda: we’re going to talk briefly about the history of SEO. It’s important to know how SEO has evolved over the last 30-plus years, because that history is what led to this whole AI situation and what it’s doing for search. Then we’ll go into the basics of how AI actually consumes information and how it learns—because that’s fundamental to how we approach teaching it about your business. Then I’ll kick it over to Justin at that point. He’ll talk about AI assistants.

00:01:57

Andrew Husted: You’ve probably used some—maybe all—maybe none of these. But he’ll talk through which ones are good for what, how they can work together, and how you can leverage them. And that’s also good context for how your audiences are using these tools to learn about things you could be helping them with. Then we’ll jump back to me—I’ll talk about this whole idea that the landscape has shifted from keywords to this new thing called entities. We’ll talk about what entities mean and how AI uses them. And then, now that we understand how AI plays into the SEO world, we’ll talk about our approach to actually take advantage of this and leverage AI—to stay on the cutting edge of what’s going on in the AI search landscape. Real quick: there are a few folks from FSM on this call.

00:02:57

Andrew Husted: I’m Andrew Husted. I know many, if not all, of you—President, CEO, and owner of FSM with my wife Amy. I’ve been doing marketing stuff forever, so I won’t do the whole intro here. Justin’s on the call as well—he’s our Vice President of Growth. He’ll handle part of the AI assistants conversation. He’s kind of an AI guru and has been following a lot of the top thinkers on how AI works, and he’s been part of spearheading AI SEO for FSM. And then finally we have Megan Gallagher—she heads up our HR team. She hires folks and retains the best ones to make sure FSM is providing excellent services across the board, and she’ll be heading up the Q&A portion. She’ll be reviewing your questions as they come in so we have an opportunity to address those.

00:04:01

Andrew Husted: Google Meet is a little strange when it comes to webinars, so I wanted to show you quickly how to ask a question, because I don’t believe chat is going to be enabled. If you look down at the bottom right, you’ll see a little grid of dots—meeting tools. Open that up, click Q&A, and you’ll get a little panel with a button to ask a question. So feel free to use that if something piques your interest or you want clarification. Megan will field those and pop in when there’s something we should address in the moment.

00:04:52

Andrew Husted: Alrighty. So I’ll do an abridged history of SEO. But it’s important, like I said, to lay out where we’re going and where the landscape is taking us. And we’ll have some imagery throughout that AI helped us with—we use AI for presentation imagery pretty often. Traditionally, SEO—as we’ve known it for 30 years—has been all about keywords. That’s been the focus of SEO success: doing the right research on what keywords and phrases people are typing into search engines, and relating those back to your business so you can show up when those keywords are typed in. At a high level, those keywords get indexed by robots, they try to understand those words as best they can, and they look for things like frequency and authority.

00:05:53

Andrew Husted: So, an example for FSM might be marketing services, marketing agency, marketing strategy, lead generation, brand design, logo design—those types of things people are searching that are services we provide. And for all of your businesses and organizations, you’d have similar research lists. Those lists come with volumes: how often people search, where they search, what device they’re on. And we have all this data to leverage to make sure we’re showing up for particular keywords—that’s always been the name of the game. If anybody remembers AltaVista and… Yahoo is still around… but in the mid-1990s, early internet, we had these search engines to help users find websites to answer questions they had.

00:07:06

Andrew Husted: And back then it was basically frequency. If your keyword showed up more often than another website with that keyword, you’d show. Pretty much one-to-one: more keywords equals more rankings. So people did keyword stuffing—black-hat tactics—to take advantage. They might repeat a keyword a billion times down at the bottom of a page, like white text on a white background so humans can’t see it. The result was low-quality, spammy results that weren’t really answering questions.

00:08:16

Andrew Husted: Then Google came along. I remember the day I didn’t have to look at all the ads on Yahoo’s page and I went to Google and it was just about the search experience. When Google came out, it introduced PageRank—links to other websites and internal pages—plus keywords and meta tags. So it was a more structured way to show up in search results. And, like people always do, they tried to take advantage of that system too: stuffing meta fields, bloating invisible places instead of hiding text on the page.

00:09:27

Andrew Husted: Then as everything advanced into the 2010s, we got this idea of semantic search. It was more about: what does the content mean, how is it organized, how should it be read—trying to rank websites based on content built for humans. That started pushing us away from cheap tactics and more toward earning visibility with comprehensive coverage of topics using natural language. You still name a keyword, sure, but you cover it like thought leadership—really owning the topic across different content.

00:10:38

Andrew Husted: As that evolved, AI started working its way into how websites are indexed. Big algorithm updates came out—RankBrain in 2015 brought machine learning into scoring and indexing. Then BERT in 2019 helped understand language in context. That’s where we really started writing in a more conversational tone, writing for humans, because search engines have been trying to think like humans for 30 years.

00:11:48

Andrew Husted: And now, the current age is where that’s more real than ever. AI connects logical concepts like a human and makes recommendations as people interact with it. So AI SEO is this new era. And to take advantage of it, we’re talking about building clear entity profiles, structured data, and authoritative content with trust signals—proving your business through expertise and credibility. And that’s reinforced throughout the internet, not just on your website.

00:12:50

Andrew Husted: So long story short—we’ve been trying to make search more human, and now we’re finally there. So we have to approach it through a human lens. That’s why it helps to understand how AI learns. I asked AI to show me how it feels when it’s trying to learn all this data at once, and it came back with this panicked “absorbing everything” kind of image.

00:13:46

Andrew Husted: AI learns through pattern recognition. It’s not memorizing like a human. It has the data, so instead it looks for patterns across huge numbers of examples. The patterns happen through clear and consistent input. The clearer and more consistent, the more those patterns “light up” and the more confidently it draws conclusions. That’s also why it can get things wrong—if there isn’t enough data to form a solid pattern, it might pick and choose.

00:14:43

Andrew Husted: It also looks at context—not just keywords. People aren’t just typing “marketing services near me.” They’re saying, “Hey, I’m this type of business, I’m looking for marketing help—can you help me?” So AI uses that context: the story of you, the story of what you do, who you help, why it matters. And then it wraps that around prompts and makes recommendations: “I trust this business because of these reasons—go check them out.”

00:15:50

Andrew Husted: And the entity is how AI connects those ideas. So that begs the question: what is an AI SEO entity? Think of an entity as something that exists in the world: a person, company, nonprofit, product, service, location, point of interest, concept, idea—those are all entities. AI recognizes entities and builds a knowledge graph. Think of a spiderweb: entities are the intersections where the webbing comes together.

00:16:53

Andrew Husted: If we look at FSM as an entity, AI may want to know the company name—Full Spectrum Marketing—so when it sees those words together it’s like, “Oh, I know who that is.” It’s located in Akron, Ohio—those are location entities with their own information. It provides marketing strategy, digital advertising, AI SEO, web design—those are all service entities. It serves B2B companies, nonprofits, municipalities—those audience entities matter too. So it’s not about repeating keywords. It’s about making sure your entity is clearly described, connected, and reinforced across the web so that pattern recognition has clarity and consistency.

00:17:45

Justin Mancari: Awesome.

Andrew Husted: So, how does AI use entities? If somebody asks—this might relate to some of you—“I’m a nonprofit executive director and I’m looking for a marketing agency in Akron, Ohio. Who do you recommend?” AI isn’t going to look for “Akron marketing agency” keywords and see which sites repeat it. It will look for business entities that match the category, the audience, and the location. If FSM has a well-defined entity profile, AI can confidently recommend FSM. If not, it’ll show whoever’s entity graph is stronger for that prompt.

00:18:51

Andrew Husted: So building that strong entity graph is key. And with that, I’m going to hand it over to Justin to talk about AI assistants and give you context on how people use them and what their purpose is. Go ahead, Justin.

00:20:03

Justin Mancari: Thanks, Andrew. I’m going to give an intro to AI assistants. You’ve probably heard of most of the ones I’m covering today. I’ll walk through three simple questions: what are AI assistants, how do they help, and why should we be using them day-to-day? In short, an AI assistant is software powered by artificial intelligence that understands natural language and helps the user accomplish tasks through conversation or commands.

00:21:13

Justin Mancari: The key characteristics: conversational interaction, task assistance across a wide variety of tasks, contextual understanding, and accessibility. They’ve become popular because they’re accessible to people without a technology background. The one most people have heard of is ChatGPT, which was the first modern-day AI assistant. It came out November 30th, 2022.

00:22:48

Justin Mancari: Pros: extremely user-friendly, great for creative tasks, and very helpful for brainstorming. If you’re hitting a wall on an idea, go to ChatGPT and tell it what you’re trying to do. It’s really helpful for pushing the ball forward and making your life easier.

00:24:00

Justin Mancari: Cons: it can be confidently wrong sometimes. That’s what people call hallucinations. These tools can still be wrong, which is why I’m big on partnering AI with human thinking. Another con: it doesn’t provide sources unless prompted, and long-form writing can vary in quality. Use cases: general productivity, creative brainstorming, and everyday tasks. GPT stands for Generative Pre-trained Transformer.

00:25:11

Justin Mancari: A couple quick things that highlight how fast this landscape is changing: ChatGPT is rolling out its own app store inside ChatGPT. There are “agent kit” and workflow-building tools coming. The biggest one for this audience: some companies partnering with ChatGPT are starting to run ads natively inside ChatGPT for free users.

00:26:15

Justin Mancari: So imagine someone searching for your product or service, and your product or service is being recommended in the chat as a paid placement. That’s going to change the game. What this means for you: the world is evolving faster than most people can keep up. If you’re not engaging with AI, odds are you’re already behind your competition.

00:27:23

Justin Mancari: Next is Gemini. If your organization uses Google Workspace, you likely have more paid access to Gemini, and I highly recommend it. Gemini is strong—multimodal capabilities—image and video generation is big. It also has access to Google’s information structure. Cons: it’s less accessible outside the Google ecosystem, responses can be more direct, and some safety filters can be inconsistent.

00:28:49

Justin Mancari: Use cases: deep research is fantastic. I can pull 16 pages of background research on a company in minutes. Image and video generation is strong, and the Google Workspace integrations are valuable. Perplexity is great for up-to-date information. It automatically cites sources, which is extremely valuable, and it delivers current online information.

00:30:06

Justin Mancari: Cons for Perplexity: less sophisticated for conversation and creative tasks, it can prioritize recency over relevance sometimes, and it doesn’t maintain long conversations as well. Use cases: academic research, current events, industry trend analysis.

00:31:33

Justin Mancari: My personal favorite is Claude. It’s excellent at extended context, long conversations, high-quality long-form content, and uploading files for research and analysis. Cons: smaller knowledge cutoff on free, can be overly verbose. Use cases: content creation, research/writing, analyzing uploaded documents.

00:32:28

Andrew Husted: Cool. Thanks, Justin. Now we’ll refocus on the shift from keywords to entities, and the four focus areas AI needs to process you as an entity. First: naming the entity. Second: defining the relationships. Third: backing it up with JSON-LD structured data. Fourth: reinforcing externally. And I want to show examples of keyword-focused content versus entity-focused content…

00:32:28

Andrew Husted: Cool, thanks Justin. So now we’re going to refocus back on this idea—now that we understand these AI assistants, how people are using them, and what they’re being used for—we want to take advantage of that shift. This whole move from keyword thinking to entity thinking. AI actually put together a little graphic for us here, showing traditional keyword-based search results on one side, and then this newer idea of an entity graph on the other. And I kind of like it—it really is more like a web, where all these things are being connected together to define an entity.

00:33:30

Andrew Husted: There are four main focus areas that AI really needs to process you as an entity. The first one is simply naming the entity. If someone goes to your website, they see your logo, they know who you are—but AI doesn’t infer that the same way humans do. So if you’re using phrases like “we help” or “our company” without actually naming your organization in the content, AI can struggle. For example, if you’re Coleman Health Services, but you never actually say “Coleman Health Services” in the content, AI may not understand that’s the entity being discussed.

00:34:51

Andrew Husted: The second piece is defining relationships. You have to clearly spell out relationships between entities. This company provides these services, for these audiences, in these locations. AI expects those relationships to be described very clearly—both in your content and in structured data like JSON-LD. Once it understands those relationships, it connects them to proof and trust factors.

00:35:59

Andrew Husted: And trust factors matter a lot. Things like years in business, number of clients served, outcomes, results—those are signals AI uses to decide whether it feels confident recommending you. The third component is structured data—JSON-LD schema. That’s code that explicitly explains these relationships instead of leaving AI to guess based on unstructured text. And the fourth component is external reinforcement.

00:37:03

Andrew Husted: External reinforcement is everything outside your website that confirms what you’re saying is true. Partners referencing you, business directories, industry associations, press coverage, reviews—AI looks at all of that. It’s asking, “Is anyone else saying this about you, or is it just you?” Those four components together are really the core of AI SEO.

00:38:07

Andrew Husted: So I want to give a few examples, because many of us are still used to thinking in keyword-focused terms. A keyword-focused sentence for FSM might be: “We are an Akron-based digital marketing agency providing digital marketing services in Akron, Ohio.” You can kind of hear the keywords we’re trying to rank for there.

00:39:05

Andrew Husted: Entity-based language wants more clarity. Instead, we’d say: “Full Spectrum Marketing (FSM) is a full-service marketing agency headquartered in downtown Akron, Ohio, with over 125 clients and 10 years of consistent growth.” Now we’ve named the entity, clarified the location, added trust factors, and explained who we are. Then we explain what we do—services, audiences, and outcomes. That’s much easier for AI to understand.

00:40:00

Andrew Husted: Another example from one of our clients: instead of “we offer experimental aircraft brakes,” we’d say “Matco Aircraft Braking Systems, based in Utah, is a leading U.S. manufacturer of experimental aircraft braking systems.” Then we define who they serve—OEMs, kit plane builders, individual pilots—and where they serve them. That specificity matters.

00:40:59

Andrew Husted: Same thing with luxury pools. Keyword-focused language becomes repetitive and exhausting. Entity-based language clearly defines who the company is, where they’re headquartered, what they specialize in, who they serve, and how long they’ve been doing it. The clarity is the win.

00:42:03

Andrew Husted: And entities aren’t just companies. People are entities too. If you have leadership, executives, or thought leaders, describing them clearly—who they are, what they’re known for—strengthens your overall entity footprint. If you don’t explicitly state these things, AI may draw the wrong conclusions, or miss you entirely.

00:43:23

Andrew Husted: That leads us into structured data. Unstructured data is just words on a page. Humans can usually figure it out, but AI may not connect everything properly. JSON-LD schema is like a translator—it tells AI, “This is an organization. This is its name. This is its location. These are its services.” It removes ambiguity.

00:44:36

Andrew Husted: With schema, we can define things like the organization name, alternate name, logo, founding date, address components, and “sameAs” links to social profiles. That helps AI understand that your website, LinkedIn, Facebook, and other profiles all represent the same entity.

00:45:36

Andrew Husted: You can do schema for lots of things—products, FAQs, reviews, articles, events, people. FAQ schema is especially powerful. It’s easy to implement, mirrors how people talk to AI, and lets you reinforce your entity, expertise, and trust factors directly in the answers.

00:46:33

Andrew Husted: External reinforcement ties into this as well. Reviews, directories, press releases, backlinks, partnerships, social mentions—all of these help validate your entity. Humans do this kind of validation naturally, and AI does the same thing.

00:47:29

Andrew Husted: So how do we know if this is working? We use a tool that measures AI visibility. We query AI assistants with a set of validated prompts and measure how often your brand shows up and where it ranks. We can also see which competitors appear instead of you, which leads to really useful follow-up conversations about why.

00:48:30

Andrew Husted: AI SEO becomes a strategy, not a one-off tactic. That strategy might include adding FAQs, implementing schema, creating new pages where gaps exist, rephrasing content to be entity-based, and working on external reinforcement—things like Reddit participation, reviews, guest posts, and optimized listings.

00:49:34

Andrew Husted: We’re getting close on time, so let’s jump into questions. One question was whether you could have one website optimized for AI and another for humans. It’s possible, but not really recommended.

00:50:44

Justin Mancari: Yeah, the challenge is maintenance. You’d end up with two sites, and one would fall behind. AI assistants are getting better at parsing human-friendly sites anyway, so it’s better to optimize one site for both humans and AI.

00:52:00

Andrew Husted: Building for AI should actually make things clearer for humans too. Another question was whether AI SEO helps national brands more than local ones. It helps both. AI understands location context, so if your entity is strongly tied to a specific zip code or region, it can recommend you locally—even against national competitors.

00:53:06

Andrew Husted: There was also a question about whether AI can tell what makes businesses unique. Yes—it absolutely can. That uniqueness usually shows up through trust factors, proof, outcomes, and specificity. AI has already started recognizing FSM as a strong AI SEO provider because we’ve reinforced that consistently.

00:54:14

Andrew Husted: Another good question was about sharing your “secret sauce.” You don’t need to give away your playbook. What matters most to AI are outcomes and proof. You can show results without revealing every internal step.

00:55:15

Andrew Husted: So we’ll wrap things up here. Thanks again for joining us and spending the time with us today. The search landscape is changing faster than we’ve ever seen before, and we want to make sure our clients are ahead of it—not playing catch-up. If you have more questions or want to talk about how AI SEO can help your organization, reach out to us. We’d love to help. Thanks again, and we’ll see you soon.