GEO to Relevance Engineering: Redefining the Future of Search with Garrett Sussman (Part 2)

In this must-listen follow-up episode, Crystal Waddell is joined again by Marketing Director Garrett Sussman to unpack the fast-evolving world of SEO, GEO (Generative Engine Optimization), and the iPull Rank conception of "Relevance Engineering." 

From the power of Retrieval-Augmented Generation (RAG) to how AI is rewriting what it means to be "searchable," this episode dives deep into the shifts that matter most for small businesses and marketers. 

Discover how to test AI’s understanding of your brand, stay competitive in fractured discovery environments, and leverage “happy accidents” for product innovation.

🔑 Key Takeaways:

• GEO vs SEO: Why Generative Engine Optimization may not replace SEO—but instead become a specialized subset.
• Relevance Engineering: A new framework to ensure brand visibility across all platforms, from TikTok to Google.
• The Rise of RAG: How Retrieval-Augmented Generation changes how AI answers questions and includes your content.
• AI's Hallucinations: Why small businesses must audit AI outputs and the potential hidden product ideas they reveal.
• Testing for Visibility: How to explore how AI models “see” your brand—despite increasing personalization.

🔥 Episode Highlights:

"GEO is kind of a subset of SEO... but it’s also a different discipline because LLMs don't behave like traditional search engines." — Garrett Sussman

"Don’t try to be everything to everyone. Know your ideal customer. Make content for them." — Garrett Sussman

✅ Listener Action Items:

• Audit AI Search Results: Ask ChatGPT, Perplexity, or Claude what they know about your business.
• Review Your Visibility: Explore how your content performs across platforms, not just on Google.
• Track Referrals from AI: Start noting AI tools in your referral sources or “How did you hear about us?” forms.
• Experiment with Product Ideas: Pay attention to fictional features AI suggests—they might be your next big seller!
• Check Out the AI Search Manual: Linked by Garrett for deeper insights into the future of search.

Connect With Garrett Sussman:

Linkedin

Download The

Does SEO feel confusing, overwhelming, or just plain impossible to figure out? You’re not alone. That’s why I created the AI SEO Foundations course, powered by Crystal GPT: your personal AI SEO coach designed for busy, creative business owners like you.

Ditch the overwhelm and discover what SEO can do for your business! Head to SEOin7days.com (with the number 7!) and get started today—let’s make your brand easy to find and impossible to ignore.

Chapters

• 0:00 Garrett Part 2
• 0:36 The Rise of GEO: Generative Engine Optimization
• 1:41 Retrieval Augmented Generation Explained
• 2:11 SEO vs GEO: The Debate
• 3:29 Relevance Engineering: A New Framework
• 5:51 Challenges in AI Content Accuracy
• 9:52 Testing AI's Perception of Your Products
• 10:06 The Future of Search and Personalization
• 15:14 Rapid Fire Questions with Garrett Sussman

Transcript

[00:00:00] Garrett Sussman: Discovery is in a weird space. It's very fractured. And so in terms of your own research and your own investments. Double down on what works first. And then expand.

Unless you are like a massive brand, don't try to be everything to everyone and all at once. Know your, your ideal customer and make content for them. And then as you grow, then you can expand. 

[00:00:22] Crystal Waddell: Welcome to the Simple and Smart SEO show podcast.

I'm your host, Crystal Waddell, here to bridge the gap between SEO strategy and real world business success.

So grab a coffee or your favorite tea. and let's dive into Smarter SEO for your business. 

[00:00:36] The Rise of GEO: Generative Engine Optimization

[00:00:36] Crystal: let's shift into a little bit of this conversation of SEO versus GEO.

What is GEO? How's it different from SEO? Do you have a fun analogy for us on that one?

[00:00:47] Garrett Sussman: I think it's a very polarizing, controversial topic in the world of SEO.

Because there's a lot of nuance to it.

In the sense that GEO: generative engine optimization. [00:01:00] Seems to be this new discipline. For visibility in AI search platforms. Like chat, GPT in perplexity in AI mode.

What's problematic about it is. Before these tools were connected to the internet, like the, the first version of ChatGPT gave you answers just on training data.

So it basically copied the entire internet. Like at a certain point of time and would give all of its answers based on that. And so there's very limited ways that you could show up.

In an LLM that wasn't connected to the internet. And all you would basically be doing is like, what can I do now? For the next time that it scrapes the internet? 

[00:01:41] Retrieval Augmented Generation Explained

[00:01:41] Garrett Sussman: But then they started using this, this technology called Retrieval Augmented Generation. Which is, going to the internet. Getting information about your website that's relevant to what someone's searching for.

Bringing it back into the chat bot. And including it.

And so now you can start to see where it becomes a [00:02:00] gray area. Because the going to the internet part, the search engine part. Is SEO.

The just optimizing for a chat bot is pure GEO. And everything else is a hybrid. 

[00:02:11] SEO vs GEO: The Debate

[00:02:11] Garrett Sussman: And so the way a lot of SEOs are saying, the argument is either everything is SEO. There's no specific tactics that you can do that's different than SEO.

It's ultimately library sciences, information retrieval.

Like what I was saying, making sure your content can be discovered. And crawled. And found and indexed. Like put into the results.

And so GEO, the argument is like, yes, it's a lot of SEO tactics.

But because the way the LLMs work, that they don't always give you the same answers.

They are predicting the next words. It's a different type of platform. It requires a different type of discipline.

So, in a lot of ways, I think GEO is kind of a subset of SEO.

What makes it more volatile is the [00:03:00] fact that at a social level, you have CEOs and CMOs of major companies. Gartner coming out and saying that the future is AI. And the future is GEO.

And so when we're talking like purse strings and how people wanna spend money.

If they're talking about, oh, we'll spend money on GEO. Versus SEO, even though it's a much more nuanced conversation.

It leads to a much hotly debated topic about what do we call it. And what it, what is the definitive right answer.

So. 

[00:03:29] Relevance Engineering: A New Framework

[00:03:29] Garrett Sussman: at iPull Rank, we, we do have a framework. We look at a level above that.

Our founder, Mike King, has titled this Relevance Engineering.

Which is ultimately this idea, it's a discipline of becoming visible across all the different platforms.

Making sure, wherever people search, depending on all the different strategies. Because all the platforms have nuances to them.

You're improving your relevance and your visibility. It's a long-winded way of saying that there's no definitive right answer.

A lot of people [00:04:00] are calling it what they want to serve their own agenda. But ultimately, I think it's about visibility across the internet.

[00:04:06] Crystal: It changes from the perspective of really being able to commoditize SEO. And being able to check a box like, oh, I did this and this and this.

And I can hand that sheet back to somebody and say, these are the things that I did specifically.

Whereas, like you said, there's more than just your website. you've gotta do it across platforms.

I love the term relevance engineering because it's so true.

You can't just be relevant in one place unless that's the only place that someone would actually look for you.

[00:04:35] Garrett Sussman: Yeah, it's a, it's a semantic thing because it's like, technically, you know, optimizing for search on TikTok or optimizing for search on Reddit is SEO. 

But when people are talking about it, they don't, they're usually referring just to Google.

And so it's like, technically you're right.

It's all SEO. But that's not what we're used to.

So it's really just how is the way that we're [00:05:00] naming the discipline, serving the person who's talking about it?

Because as a business owner, it's confusing. Do I hire an SEO agency? Do I hire a GEO agency?

Do I hire, you know, a digital marketing agency? It, it actually has practical implications.

The words we use matter. And it matters in different contexts.

[00:05:19] Crystal: Wow, that's so interesting. And you explained something.

RAG, the whatever, augmented, retrieval.

[00:05:25] Garrett Sussman: Retrieve augmented generation.

[00:05:27] Crystal: explains to me something that I'd wondered for a while because I was an early adopter of Jasper. And

[00:05:32] Garrett Sussman: Hmm.

[00:05:33] Crystal: their big thing was always like, we've crawled the internet up to this point.

And they were using that model and then all of a sudden chat, GPT came out and pretty soon it was, no, this is up to date.

And so I was like, what changed about the technology that all of a sudden, we

can crawl now?

Because that's what I really felt like was missing.

[00:05:51] Challenges in AI Content Accuracy

[00:05:51] Garrett Sussman: Yeah, I mean, well, so it's this whole, like the whole kind of field of machine learning and artificial intelligence. Continues to advance.

[00:06:00] And people are still figuring out, like we are still dealing with, you'll, you'll see issues with hallucinations. The fact that, you know, AI just makes stuff up a lot of times.

And that's really problematic and that's not something that's been solved.

And so. Because AI wants to give you a confident answer. It never really tells you. I don't know. It'll always make up something.

So being able to have retrieval, augmented generation, the RAG pipeline.

Being able to go to the internet and get information, gives it a better chance of being able to give you a realistic answer.

Like eventually we would hope the technology advances where it can say, I don't have this information.

I'm not just gonna make stuff up for you. But that, that's not the case yet.

[00:06:42] Crystal: I've heard people who are like, you know, requesting sources and then even sometimes the sources are made up.

And I know even with when I'm creating copy for my product pages or something like that.

I have a brand guide. I have different reviews and different things that I've fed into the system.

And I'm like, okay, create this type of [00:07:00] page with this information and I, it makes up something I know didn't exist because this is my business.

I know what happens here. And so I have to say,

hey.

Don't make up a quote. Use a review that's true, and could be attributed to somebody who actually purchased something from my store.

And they're like, oh, okay, I can do that. But it's like, why didn't you do that in the first place?

[00:07:19] Garrett Sussman: Businesses have to deal about this. You're hearing ridiculous stories right now. Like a SaaS example.

Where people were asking about this music software and ChatGPT was telling people that it had this feature that it did not have. And so their customer service team got really confused. Because everyone was asking for this feature and like.

We don't have this. This doesn't exist. Now, the smart owners of that business took that information, identified it, and actually created the feature.

And people loved it.

But that could be an issue. People could ask, you know, like, why don't you have this in pink? Or, why don't you have this product that you can attach to your sink in the way that I want.

'cause ChatGPT said so. and so, you have [00:08:00] to continually listen and test and experiment for yourself.

What are these AI search models telling people? That may or may not be accurate. And actually have a plan in place for, for how to handle that.

I feel like as we become more educated and understand the technology, people will be more understanding when that's the case.

[00:08:17] Crystal: I think that's where that human element of empathy comes in.

Because the customer service agents were like, Hey, wait a second, we don't have this.

But somebody was smart enough to say, well, why is it coming up?

And like kind of a logical progression would be, you know, maybe the search engines were aggregating complaints or concerns or questions from the internet about a feature that they wanted that they couldn't find.

And so for them to create. I'm like, what a great response by the company, you know, to have the humility to say, Hey.

Maybe this is wrong, but there's an opportunity here. So.

[00:08:50] Garrett Sussman: And there are gonna be so many of those going forward. I think the savvy business owner has an opportunity to thrive.

There are so many opportunities, depending on leveraging the [00:09:00] tech. And just the, the ecosystem that it creates.

[00:09:03] Crystal: I think I've had that experience myself.

I make senior night gifts, I make gifts for athletes, like like on their numbers.

Their parents will send me pictures of their sports careers or whatever.

And so I'll put them on their numbers and I ship it to 'em.

It's very, very cool keepsake, you know, for their athletic career.

But early on, when I'm writing product product descriptions, Chat is telling me, oh yeah,

and I've got this amazing keepsake box.

And I'm like, keepsake box. I don't make a keepsake box. And then I was. Why don't I make a keepsake box? You know? '

Cause now I've got an upsell because I'm a manufacturer, I can make anything.

And so it was great because sometimes it's like those mistakes are actually some of the best ideas that I wouldn't have come up with on my own.

[00:09:43] Garrett Sussman: Happy little accidents. who said that? It sounds like Bob Ross. And his painting Little trees, right? Happy accidents.

[00:09:48] Crystal: No mistakes. Just happy accidents.

Okay, so what about testing? 

[00:09:52] Testing AI's Perception of Your Products

[00:09:52] Crystal: what's

[00:09:53] Garrett Sussman: Yes.

[00:09:53] Crystal: way to test how AI sees our products? And can you do it in incognito mode? Does it matter?

[00:10:00] If you do it from your own browser, is that gonna impact your results? Like, those are the types of things I'm kind of wondering.

[00:10:05] Garrett Sussman: My goodness. 

[00:10:06] The Future of Search and Personalization

[00:10:06] Garrett Sussman: It's so much harder now. And it's only gonna get harder.

Because in May, Google, at Google io, their big conference, talked about this idea of using more and more personal context. In the output.

So think about when you do a local search, when you ask for a pizza place near me, it recommends.

Pizza places within a certain geography that you can actually get to. It wouldn't be helpful if it gave me like pizza places in Australia.

So we've had an element of personalization.

The AI capabilities, looking at your entire search history, looking at everything it knows about you from the v various data points that you've allowed Google to collect, or open AI to collect.

Is going to give you personalized answers.

So monitoring becomes much more tricky. The way that some software tools are trying to solve this is they're creating fake [00:11:00] prompts, like fake, fake queries, fake keywords.

And running them, you know, hundreds and hundreds of times. At such a scale that can give you like kind of an accurate, not precise. But like an accurate estimation of what are the recommended products that we're seeing over and over and over again.

Now I think that like points you in the right direction.

But for tracking the output of this, it's gonna be a lot more difficult.

Anything independent of clicks you literally see come from chat GPT.

Like in your referral Google Analytics data.

Or you know, things that when you have a, how did you hear about us on your form, and they actually say, oh, you know, AI mode or perplexity.

Those are gonna be the best ways to track the actual revenue and conversions.

But to be able to track whether or not it's working, it's gonna be messy.

It's gonna be a gray area.

There are tools that allow you to do it at a, a kind of big scale level that aren't precise. The other option that I would [00:12:00] recommend is starting to find people in your segment audience.

And having them do searches is a very inexact science.

But like seeing what they're seeing. Like if you know that your average customer is, a. You know, millennial, who works in, the financial industry.

Have them do some of the searches and see if you're being recommended. And kind of run these tests and experiments outside of your own account.

[00:12:24] Crystal: on one of your webinars where you were with a lady who was running those experiments. 

 And she was sharing like the insights of what people searched and what that really was like versus what we might've thought it was gonna be like. I thought that was really interesting.

[00:12:37] Garrett Sussman: We did this whole ux AI mode study which was usability study.

Ran through 23 different people of going through these five tasks. Of, you know, what's the best find the best credit card loyalty program.

Or what's the healthiest breakfast cereal for kids? Or find a health clinic near you, like local, commercial.

And it's so interesting the way that people [00:13:00] search is so different.

Like going back to early in the conversation. Yes. Like Google still has like 90% of the market share. But people do search on TikTok for products. They do wanna go straight to Amazon. They do wanna go straight to the brands that they're familiar with.

So. Discovery is, is a weird, is in a weird space. It's very fractured. And so in terms of your own research and your own investments. Double down on what works first. And then expand. Like, unless you are like a massive brand, don't try to be everything to everyone and all at once. Know your, your, your ideal customer and make content for them.

And then as you grow, then you can expand.

[00:13:42] Crystal: that's so true. And then you brought in another. Another term, which is ux. And I felt like, if you build websites.

And you actually are building websites to convert. You start to understand how these elements all fit together.

And so, even though GEO is a new term. To me it just kind of [00:14:00] encompasses like all of it, you know?

Because when

you do it all yourself,

you don't have like, product silo over here. You don't have an SEO team over here. You don't have usability studies coordinators over here. You like, are "the thing"

so like to see it all finally come under an umbrella as a conversation for everybody.

I feel like, you know, small business owners are the best position for this because we've been doing all of it.

You know, now there's just like a comprehensive umbrella that says this is what you're doing.

[00:14:31] Garrett Sussman: Yeah, it's, I mean, it is, everything is so holistic as a business owner. Especially in the digital space.

And just keep in mind things are changing. Like I hope everything that I said on this interview isn't outdated by the time this is published. Like what I say today. Could be different tomorrow.

The whole AI mode thing that we've done is available.

But it's not the default version of Google search.

I feel confident in the next six to 12 months that it will be the default version of search.

And [00:15:00] we'll all be treating Google like we do chat GPT. And that fundamentally changes how we interact with search.

And as business owners you have to like kind of be ready for that in the background.

[00:15:11] Crystal: so do you have time for about 90 seconds? 

[00:15:14] Rapid Fire Questions with Garrett Sussman

[00:15:14] Crystal: Of rapid fire? Okay.

SEO is dying, yes or no? Gemini, perplexity. Claude Chat GPT or something else.

[00:15:24] Garrett Sussman: Ooh. ChatGPT.

[00:15:25] Crystal: Best use case for AI and SEO.

[00:15:28] Garrett Sussman: Oh. Oh, it's hard to do that. Rapid fire. Best use case for, for AI and SEO classification.

[00:15:34] Crystal: AI agents. Hype or breakthrough?

[00:15:38] Garrett Sussman: Hype right now, but in five to 10 years...

[00:15:40] Crystal: okay, breakthrough.

[00:15:41] Garrett Sussman: Breakthrough-ish.

[00:15:42] Crystal: Google AI. Overviews good, bad, or ugly?

[00:15:45] Garrett Sussman: Good.

That's, that's gonna get me some hate that one.

[00:15:48] Crystal: Jobs in SEO that AI is most likely to take in 2026.

[00:15:53] Garrett Sussman: Ooh. Anything entry level.

[00:15:55] Crystal: Jobs least likely to be taken.

[00:15:57] Garrett Sussman: anything strategy focused.

[00:15:59] Crystal: And [00:16:00] best AI powered app for SEO that no one knows about.

[00:16:03] Garrett Sussman: I dunno if no one knows. But I love profound for monitoring performance in ai.

[00:16:09] Crystal: Awesome, cool. Well, Garrett Sussman.

Thank you so much for joining me today on the Simple and Smart SEO show podcast. This was amazing.

I think I got through about a page of the two and a half pages of questions I had. But it's a great start.

[00:16:22] Garrett Sussman: You gotta check out that AI search manual too.

[00:16:24] Crystal: Yes, I'll link to the AI search manual.

And then if people wanna follow you and connect with you, how can they do that?

[00:16:31] Garrett Sussman: Yeah, mostly on LinkedIn these days. I was a big Twitter person before it became X. 

I'm still on there. But Garrett Sussman on LinkedIn. Find me.

Happy to, you know, talk about whatever you want.

[00:16:41] Crystal: Well, thank you again so much for being here. This was an incredible conversation. I knew it was going to be.

And I'm excited for SEO week in 2026, so I will see you there.

[00:16:50] Garrett Sussman: Thanks for having me, crystal.