SEO in the Age of AI - Gaetano DiNardi

When I first got into marketing, SEO was by far the most hipster of all the sub-disciplines. It promised massive amounts of free traffic. If you could just figure out how to game the algorithm, there was this David and Goliath energy to it. Scrappy startups out maneuvering these big slow moving incumbents who hadn't figured out digital yet, and it had this air of mystery.

A lot of SEO experts came off like these underground. Punk rock types like Rebel Alliance versus Galactic Empire Vibes, because so much of it was hard to verify, the space was also full of mythology, half-truths, misinformation. That was well over a decade ago and a lot has changed since then. Not necessarily the misinformation part, that's probably still around, but the game itself has changed massively because of ai.

So to help unpack it all, I brought on the coolest SEO expert I know, Gaetano DiNardi. We're gonna talk about how SEO has evolved, how AI is reshaping it, and what to focus on in 2025. Gaetano so happy to have you on the show. Well, thank you.

That was an awesome intro. Happy to be here. You know, back in the day, SEO. Had kind of this like, set of rules, you know, if you follow the experts, it was all about, all right, you gotta do your meta tags and your page headings and mention the keywords so many times and build your backlinks and structure your pages like this.

I'm just curious from your point of view, and I haven't been as clued into SEO over the last little while, like what has changed over the past 10 years? What conventional wisdom just doesn't apply anymore. Well, you, you could do all that stuff and still not perform well. So, that, that is just like the foundational building blocks of like table stakes now.

So all, all those things, I think people viewed them as, oh, you know, if we do all this, then we'll be far more likely to succeed. Now that is table stakes. So that used to be true because nobody was doing it. Once everybody figured out, these are the, these are the ways you need to structure pages and content in order to perform well.

Uh, then it became less effective. So in, in a, in a summary, the law of shitty clickthroughs, where as more widely adopted a technique becomes the less effective it is, that has basically happened over and over again. Lather, rinse, repeat. everybody figured out, you know, once I do the basics and then drive tons of back links and then, you know, repeat, lather, rinse, you can grow traffic.

but that's, uh, no longer the case. In the new world that we're in. I mean, I remember like people used to obsess over every Google update. You know, it's like it's panda and it's this and it's that.

I don't even remember the names of all of them. we zoom out from that, did a lot actually really change in terms of the game, like over that time, or is it really the, the fundamentals are the same and maybe it's just getting harder because of competition? Yeah. In, in a sense, it's getting tougher because of competition, but fundamentals still are important.

you, you know, I can take the same exact strategy I. And apply it to two different brands. One being a powerful, successful, well-known brand, and the other being a scrappy, unknown startup, and the scrappy, unknown startup is not gonna perform. Meanwhile, the big.

Well-known brand is gonna perform same exact strategy, same exact everything. Um, and that, that's just the way it goes because bigger batter brands have, they, you know, they get a preference, right? And so there, there's that aspect of it. which I think now AI is, uh, democratizing a little bit where you can be a smaller scrappier challenger brand and still.

Get a lot of coverage in AI search. you know, that's just the way that I see things unfolding, but when it was just a Google game, yeah. You know, the biggest, baddest brands generally win. I.

What, what is that about? Like, is, is Google just on the backend, kind of using some, uh, signal of brand strength or company prestige or something like that and, and kind of putting their finger on the scale or what, how does Google kind of tell who the biggest, broadest brands are from that point of view? Yeah, there's, uh, site quality ratings and the Google leaks that came out. clearly, you know, brand mentions are, are a factor now that came out.

so the more brand search you have, that is a factor I. And in fact, uh, I think it was a guy named Tom Capper, on Moz. He came out with a really interesting study that showed if you're one of those sites that gets a lot of SEO traffic that's non-branded, but nobody's checking for your brand. That's kind of weird.

It just doesn't seem to really add up. And so those kinds of sites are gonna have a really low overall site quality score. Meanwhile, if there's a correlation between high amounts of branded traffic and non-branded traffic, that's generally a good sign because it's more normal. And if you think about the kinds of sites that may have no brand, but a lot of non-brand, what is that usually?

Sites that are gaming it somehow, you know, casino, online gambling, maybe some low grade publishers doing a bunch of concept farm stuff, right? affiliates. So they're, they're using that to fight back against the spammers of the web essentially. You know, on, on that note, one of the things that I have really noticed and been reflecting on when it comes to search in my own experience as a user is just, just how bad it is.

Like how you, you search for something and every single. Article seems to regurgitate the same like stale, generic, lifeless insights, and it's probably why I use chat GPT for coming close to a hundred percent, at least as a starting point of queries. I will, I will go and type in chat GPT instead of Google. You typed about this too.

Has, has, SEO kind of killed Google or obviously there are many, many people still using it, but have they kind of become victims of these practices where the content isn't even worth reading anymore? I. There's, there's been some steps to, to sort of fight back against that with, the, the new ways that search is working. but the trade off is that there's less traffic going out.

So if you think about, uh, the way AI mode works, there's a concept called query fan out. Query fan out means, let's just say you start with a seed keyword. Like, um, identity theft protection, let's just say that's the keyword. The query fan out generates what's called synthetic queries around it, which basically tries to cover the whole concept of identity theft protection in like one conglomeration Frankenstein answer.

So that diversity of answers aims to some extent. fight back against like the lifeless, soulless. It, you know, it, it just one type of page down the, down the line of the 10 blue links, they're all being the same lifeless, soulless page. The AI mode tends to, to fight back against that with the query fan out.

Because it's gonna give you perspectives. It's gonna give you, you know, pros and cons, pricing models, what people are saying online. It's gonna scrape together all these different, types of answers or fragments of answers. They actually call it FRAs, and then they, stitch it together.

And then there you have it. And, and so that's one way, you know, Google and other AI platforms can, can fight back against lifeless, solace content. But, but like, if I think about the root cause of the problem, it's because Google tends to reward in, in the classic model, it rewards consensus, meaning like you can't deviate too far off. From the main agreement of what something is, because if you do that, it's not gonna show up.

So it's really hard to inject what's called information gain or, you know, new twist or spin on something because if you're too far deviating away from the consensus answer, it's just not gonna get surfaced. And so that's, I think like the, the push and pull of it. It is an interesting, because, you know, on, on social, like let's say the, the LinkedIn. space, even though it is very much an echo chamber, people are seemingly rewarded for these like contrarian takes and for like poking the bear a little bit.

And then in, in Google it's like, well, let's just see what like the average of the average of the average is and give it that Yeah, you do not get rewarded for contrarian takes in, in SEO. you, you have to, I, I mean, if you can. Potentially, but it's tough. And user engagement signals would probably be the thing that, uh, dictates that.

So what you would kind of need to do to an extent is like reframe the overview in the consensus way, but then you have to change it. So I think an example would be, is identity theft protection worth it? Think of that query for an example. Historically, all the vendors and providers would say, yes, it's worth it.

Yes, it pays for itself 10 times over. Yes, blah, blah, blah. Everybody just saying, yes, you need it because they wanna sell it. But when I was working in ID theft protection, we were the only brand at the time that said, well, it's only worth it if.

You know, this, this, this, this, this. It's only worth it if you don't wanna manually track your credit score. It's only worth it if you don't wanna manually open your credit card statement every month and look through all the transactions and make sure nothing's fraudulent. Summary.

You know, if you wanna do all this manual work, then you don't need it. If you're good enough to be on top of it and you're digitally savvy and you have the time, then yeah, you don't really need it. But if you are not tech savvy, if you don't have the time, if you don't have the patience, if you don't have the know-how, then the tool of course just automates all that for you. So we were the only brand at the time just laying it out like that.

And that is a way to put a spin of maybe, you know, a new approach, a new take, a new thought on something that requires a lot of consensus. I mean, that sounds like a more useful article to me as a reader. Like I would rather, of course always very suspicious when it's like, should I buy the thing that I'm selling? Yes.

Buy it. You know, for me, Yeah, it's more honest. It, it, it's more honest, like, uh. I, I think that, that, that's why, um, newsletters and following people's sub stacks is becoming so popular because you're gonna get a lot of those contrarian takes.

Like if you want a guy that is doing a lot of contrarian takes in SEO right now, his name is Dwayne Forrester. Great guy to follow, very tech focused, very AI forward thinking, but he's thinking of stuff that everyone is so far behind on. But I follow him because if you were to write that stuff on Google, I'd never find it. So that's why I think the emergence of Substack and following people's personal brands and um, even social profiles to an extent, you know, this is really becoming a growing thing.

I totally agree. I mean, that is a hundred percent of where I educate myself, uh, these days. I'm curious to come back to what you were describing before about Query Fan out and this AI generated summary and is what you're referring about. Just to make sure all the listeners are on the same page, is this the, the AI generated text that appears at the top of a search result when you search for something in most engines these days?

Yeah, that's right. That's absolutely right. I mean, let's, let's talk about that. 'cause that's a fairly, uh, new maybe in the last year or two.

Uh, innovation, like how much is that radically changing the landscape from When we, before we had 10 blue links, and then we had 10 blue links and a sea of ads, and now we've got this like thing that you don't even need to click to see the answer anymore. Well, I mean, I could go off on this. I don't know how crazy and deep I could go, but maybe I'll just try and summarize the first batch of things that come to my mind. first of all, the device that you're on is tracked and all of your behavior history on your browser, like, let's just use Google, right?

It reads your Gmail. It has access to your calendar. Uh, it knows your clicking chrome, it knows your click preferences. It knows everything basically about you through your interaction with Google Properties, and it creates this.

Profile of who it thinks you are and what you like. Based on that reason alone, no two prompts are gonna have the same answer. If two people, two different people search the same thing or similar thing, it's gonna be highly, highly personalized to your taste. And so just based on that alone.

You have a, a, a really like advanced, advanced version of what used to be the 10 blue link personalization. Now it's gonna just feed way more into that. and, and, and um, from there, the next thing that I really think about is the SEO implication of like old school versus like kind of new school. If you're someone doing SEO rank tracking gonna be gone.

Because we're moving to a visibility model over a keyword position. Rank tracking model AI mode is probably gonna be the new default pretty soon. It hasn't happened yet, but like all the signs are there. The writing is on the wall.

10 blue links may be completely gone by, you know, 20, 28 or to get it, you may need to like. Toggle something that says, yeah, bring me back to classic Google. There may be some people who don't like the new ai, but I think AI will be the, the default. Search experience pretty soon.

So that kills rank tracking. that also changes the way you think about content structuring like you used to be, uh, able to rank for a massive amount of queries and get traffic by doing one mega skyscraper page. Remember the skyscraper technique that was like really popular from Brian Dean, uh, the back Lanco era that's kind of dead. So instead of now doing one mega page, or like everybody's used to say, like, content length, six, 7,000 word guides and keep re updating your content.

Uh, I think even Neil Patel and his like SEO presentation slides were showing like, Hey, look at how Wikipedia keeps updating their pages every year and making them longer and longer and longer. Uh, you should be doing the same that that's dead now. So the disassembly of the skyscraper. Is actually what's great now because rather than covering all these things jammed into one mega page, you're better off hitting the user intent straight on the head, going as long tail granular as possible.

To get that real deep topical coverage. And then ideally with practical, uh, internal linking, having, if you can do it URLs nested together into the same folder structure so the site knows that all the little clusters of these things are linked together in this little section, you know, that's better. So in some ways it's full circle because HubSpot was talking about hub and spoke modeling, you know, back in 2017 or whatever. But we got away from that somehow.

and what's bad now is like a million things jammed into blog that's like not great because blog has generally pagination. And so like, have you ever seen blog slash page slash like 37? That's the only way you'll find content that's bad. Because now it's like four or five, what's called seed positions away from the homepage in terms of crawl depth.

And it's just not good for, for for modern SEO. And so, I don't know, I feel like I could rant about this for a long time. Maybe I'll stop talking. But that's just what comes to mind.

a fascinating change. I mean, so I guess there's like two, two threads that we could explore. The first is where, are trying to rank for informational content where it's like, let's say what is identity theft protection? And maybe you're Acme identity theft protection.

So you're, you know, you wanna rank for that term, you're publishing your content and that's bringing you a lot of traffic at, at one point in time. And now all of a sudden somebody can search for that. Get an AI generated summary get their answer without necessarily any click at all to anywhere. And, and meanwhile it might be pulling from you might be pulling from your competitor, but it almost doesn't matter 'cause like I'm not leaving the Google front page.

How much traffic are, are people losing from that sort of thing in your experience? Uh, a ton. on the low end, 30 to 40%. Um, on the high end, 70 to 80% depending on how basic the query is.

I mean, if you wanna actually know something that, came out in the Claude AI leak that really summarizes this, I'll show you. So I have this post coming out, soon, but, essentially this is info for brands trying to optimize for AI search why you should avoid covering topics that are considered general knowledge or established information. Because the quad AI leak revealed, there is a command in their systems called never search, meaning never search the web, never cite any brands, never put any links to anything.

For any queries that represent, uh, what they're considering, timeless info, fundamental concepts or generalized knowledge. So you're, there's no point in really covering that stuff that has near 100% consensus, especially when AI mode is the future, when people are migrating to perplexity and clawed and HGBT and et cetera. There's just no point in doing it because you're never gonna get sourced, ever. You're never gonna get cited, ever.

So this really pretty much covers the what is X thing, unless it's like something that is nuanced and emerging and evolving. If it's like a timeless, classic thing, like what is, I don't know, demand gen marketing. I mean, very unlikely that that's gonna have new information. It might.

But the writing is on the wall here. Um, facts that are timeless or stable. this includes things that, um, are basic, established historical information. It's very unlikely that, um, you're ever gonna get cited in here.

And so I think that was a really good learning and validating to, to see, um, this is the direction we're going. Yeah. So does that just basically killed informational SEO as a strategy or it just means that people have to play that game a different way? Well, I think you'd have to ask yourself like, uh, is the informational questions people are asking stuff that is, is already been answered a million times and there's not much newness that you can add to to it.

Uh, I think in cyber tech there's probably a lot of that. Like, you know, how does phishing work? You know, uh, what is identity lifecycle management, stuff like that. It's just what is network security?

All those sort of cyber glossary of definitions that just never really change. Maybe, maybe some of the newer ones evolve, but for the most part, a lot of the classic stuff never changes. And so, um, you know, you have to ask yourself if you're a challenger brand, like why do it? I think this was the, the tried and true B2B cyber tech, SEO playbook, um, really for like a decade.

Uh, every single cyber tech brand did the same thing. Glossary of terms pages. That was just what they did. They all copied each other.

The first to ever do it was Palo Alto Networks. I think they did it in 2013 or 2014, maybe even sooner. And then everybody else followed suit and, and copied. So I guess today, if you're starting your your SEO strategy and you're doing that, man, good luck.

So the other thread, there's that informational thread, and then the other kind of path is uh, a search query that shows. Purchase intent of some kind. So like best identity protection, theft provider. And that's the case obviously where people can be, can be bidding as well for paid placement.

But people vie very strongly for organic placement as well. And then here you have the AI. Making its own selection of who to present. And, and so one of the questions I know that we're asking internally where I work and other people are obviously asking is like, how do you get into to be one of those recommended vendors in that prepared text?

Do we have any indication of what we might need to do? Let's show you an example. right. So here's a tool that I've been playing around with.

It's called Scrunch ai. Um, disclaimer, I'm not a affiliate. I have no association with this company. I just like the tool and I'm using it.

you can track AI visibility for any number of prompts associated with your brand and get, um, information on what third parties are ranking, how your brand's ranking, what you can potentially do to increase your visibility there. So look at this one. What is the best identity theft protection solution? Um, this particular client is called Aura, and if we look in the citations for this show, cited websites, what do you notice?

PC Mag Right. Radar. Tom's Guide. Cnet Investopedia.

There's not one citation here. For any actual brand page. So you see the answer is Aura LifeLock by Norton Identity Force, and you can even hover over the citations there and see in this particular case it was PC Mag, um, and it was TechRadar getting the, the, the, the credit for, for the source. But the brands are showing up in the, in the tech space answer, but they're not being cited.

Their actual websites are not being cited. And so this is a huge misconception that a lot of companies have. They think, yeah, we just need to optimize our pages, right? The thing we talked about in the beginning.

Make our, make our pages retrievable by ai. Make sure they have chunk ability factored in title tag, FAQs, structured data, right schema, all that's important, but you need that third party strategy. There's no amount of first party content on your own website strategy that you can really do. To guarantee that you're gonna get pulled into these answers for competitive queries.

and so it becomes really a third party game in, in this case, right? So this is a real live example, and it's not that different from old school where you wanted to do something called, I, I used to call it SERP Domination. SERP Domination Strategy, where you would search for a query, you would go through the 10, you have to do this manually, but you would go through like the top 10 ranking blue links. You would do an audit of, all right, you know, this is a high ranking page.

We're not mentioned. What do we have to do to work with that site? This is almost that same thing now, just expanded outside of Google. It's really applying it to, you know, chat, GBT, perplexity, cloud, et cetera.

So, I mean, first off, this is super cool. Uh, for those that are just listening audio only, we will include a link to scrunch in the show notes because what we're seeing here is it's showing all the kind of main assistance chat, GBT perplexity, Claude AI overviews, and looking at presence position. And sentiment and what, what I came away with from what you're just showing, Gaana was like these big mainstream publications like PC Mag or Toms Guide or whoever, they're like the, the LM is essentially kind of outsourcing its judgment to them to some degree, and using them as proxies of that sort of consensus view.

It seems. That's right. That's right. And what's great about this is you can see how important third parties are in the overall dashboard.

So for the set of prompts that I am monitoring, you can see the breakdown of. Citation by brand, competitor, and third party. And so again, it's debunking this huge misconception that executives have. They think that so much of this is reliant on your own website strategy.

It's actually the opposite. It's like a 85, 15 90 10 split, like for the competitive bottom of funnel, what I used to call money terms. You know, you can call 'em money prompts, whatever. Um, it's like a 90 10 game.

I. They pull this information off, especially for high volume stuff. They pull this information. Look, look what's getting the most amount of responses.

NerdWallet security.org. CNET security.org.

Again, Tom's guide. There's the first one You see that it's not a third party is Norton, but so much of this is third party affiliates. and I think that just might be, um, you know, uh, sort of a full circle moment for where we're going in SEO and I think it's also M'S mechanism for anti-spam antiga, um, you know, pushback, right? Because you can, you can quickly see how.

SEOs would figure out a way to manipulate this if it, if it was, you know, more of a, what you can do on your own website came if, if that was the case, everybody would just do top 10 best lists for every single thing possible, and then that would get pulled into LLMs, but that's not what we're seeing. Is it that different from like the big innovation of Google that set it apart from other search engines? As I understand it was like the introduced this idea of like domain authority and, and looking at back links and saying like, all right, if a site gets a lot of links from other.

High authority sites, then it's a more reputable site and people obviously learn to game that with link buying and and whatever. But this is almost just another type of backlink, like a citation. But still the trust and the authority of the site that's linking to you is what's really determining whether you appear, isn't it? Um, yes and no.

Um, I think what I've seen is that the relevance matters a lot more now. It just so happens that in this category there's mega domains. So in the, in the instance of identity theft, you know, it's such a valuable commercial category that you're gonna see the big players, right? So you're gonna see the Toms guides, the C Nets, the NerdWallet.

If NerdWallet is in it, you know, it's financially. Profitable, right? They're making bank off this. But you also see some smaller players too, like security.

org, a little more niche, but because it's a security focused domain, it actually has even more citations than Tom's Guide. And Tom's guide is like probably double the authority, overall authority. So you have this aspect of relevance, by topic. That I think matters a lot.

So what's the takeaway for CMOs and marketers? It's, do I need to be in NerdWallet? Maybe. but you're probably better off, uh, finding a more relevant, less authority source.

Like those can also be great. So like if I go to a different project, I'm just gonna go to this one in cyber tech that I'm working on. Um, if I go to sources. I go to third party now this is B2B cyber tech.

Think like enterprise, Mm-hmm. you're gonna see solutions. Review Sentinel One, you're gonna see Gartner. So right away you're seeing like a different caliber of, of websites in here.

Which, you know, it would be a mistake for a company like this to think that they should go and work with like the highest overall authority domains. Like, maybe that would help, but you'd probably be better off, you know, going to solutions review and seeing, is this a pay to play or, or not, you know, this, this looks like it potentially could be a pay to play. So you need to figure out, you know, how to get on these, these pages, build, build some kind of relationship with these brands and figure it out.

So that, that kind of gets to where my mind was going. Like if you're, a B2B CMO or SEO program manager and you're trying to figure out this new landscape, obviously your mind goes to like, how do I, how my presence in these kind of mainstream prestige review type sources? Is that what we should be thinking about? Yeah, that's right.

That's right. It's, I think it's um, you can call it Link Building 2.0. I think, uh, you know, if you're a bigger brand that has like PR and comms and broader mark comms, this is a skill they're gonna have to learn how to pick up because it's, it's gonna matter, right?

So if you can, if you can work on that with your, uh, comms team, I mean, that's gonna just help a ton. 'cause this is a new, new world we're in where, you know, mentions, um. You can, you can argue even now that they're almost equally as important as, as backlinks. If I think about that old world that I referred to where you could have, uh, a digitally savvy startup come in and kind of like find some way to game their way.

To the top of the, the pile and get a lot of traffic. it seems almost like, brand incumbency is a much bigger moat and it's gonna be very hard for a startup to like break into that without doing that brand work first. That's absolutely right. if you don't have that brand foundation.

It's gonna be really tough. Now, if you can somehow brute force your way onto all these lists, then yeah, you know, maybe you can get some quick traction, but it ain't easy getting on these lists, man. If you're a no name, why should they work with you? Why are they gonna add you to their list?

Like what's in it for them? Very little to nothing. So, you know, um, it's not as easy as everyone is, is, is thinking, you know, just, Hey, just get someone to do outreach and get on these lists. Yeah.

In theory, that makes sense. But, you know, I, I, I haven't seen anyone successfully do that. least not yet. And so I think that's gonna be probably the next, wave of things that's gonna come out as part of this new AI era that we're in.

I mean, you, you used the words pay to play and I, I'm not gonna point at anyone in particular, but I think we all sort of know. Some of these major, uh, sources that are considered very reputable, that enterprise buyers look at do have a definite pay-to-play component. So I guess two, two things for me, like number one, like how much is, is the game just rigged? And, and number two, like if you were trying to like advise or were part of like a gorilla startup today trying to break in, like what, what would your access point be?

What would you be thinking about? So to the question of, uh, how easily gameable is this? you know, what I could tell you is there's, there's a lot of lucky natural stuff that happens for you when you're automatically a big brand. So like, here's like an example.

I will go back to that, prompt on like identity access management tools, and we'll just look at this. So the, the most, um, cited page is this is larry.com page, so top 13 IAM tools. And this is not a, uh, affiliate site or a third party.

This is actually like a provider just doing the tear down. And, um, if you go to the list of tools, they of course no surprise, promote themselves at the top. And then if you look at who else they include on these lists, right? Okta Cyber Arc One login, Microsoft Enterra SailPoint, RSA, secure id.

Paying identity Semantic, ForgeRock, IBM, Oracle Duo, like did any of those brands work really hard or do outreach to get on this list? No, they just got on this list because they're a big brand. So that's part of the effect that I'm talking about where, you know, being a big brand just. Gets you more free stuff, and that's just what, what happens?

can, can a list like that be gamed? I mean, probably not, right? Because they're, they're just trying to promote themselves as the top ranking tool on that list. Maybe they, now you could look into the back links to that page and say, all right, maybe they're buying links, you know, to some extent that still could work.

And then what was the second half of the question again? gaming it and all right. Let's say you, you are just starting out. You're a new startup without any brand advantage, and you are Yeah.

into this world where, you know, Gartner isn't covering you, Tom Scott isn't covering you, whatever, whatever your sources of Yeah. What, what do you do for SEO? Do you even consider SEO or do you, do you focus your energy elsewhere? Um, I mean, it depends on, I think, how competitive the category is and where I might have like an advantage in, in doing content.

So, and, and it also depends like, am I at ground, ground zero with nothing, nothing, nothing. You know, if, if I'm at nothing, nothing, nothing, I'm not doing SEO. There's no chance I'm doing it from absolute zero. Nothing, man.

You've gotta go do some foundational work first. Like you've gotta drive customer reviews, you've gotta do probably LinkedIn, founder led marketing, maybe drive up a newsletter of some sort. And you also need to bring like a unique like position to, to like your overall content and why should people even listen to you. Um, I think that that matters too.

So you probably need to do like, you know, a podcast interview tour. You probably need to be a guest on webinars. You probably need to do some like small local events. You know, you gotta do some stuff to, like, you gotta grow your social following.

I think you need to do a lot of those foundational things before you just start doing SEO. so that's where, if you're like a founder that has say. A personal brand of some kind already has like a following that's pretty decent. Maybe you already have like a, a newsletter or a substack that people follow or like it.

What I see a lot of is, you know, X big wig, they get a little bit of a boost. 'cause if you are. You know, X head of SEO at Amazon and now you're doing a AI search tool. Yeah.

People are probably gonna be more willing to give you a look because, oh, this was the former XSEO at Amazon. Like, this person's gotta know their stuff, right? So I think you've gotta do at least a year or two of foundational work. You can't just do SEO at the gates anymore.

You're gonna fall flat on your face and you're gonna be disappointed. So it's maybe that point where you have found product market fit, and you're now trying to scale up your growth channels. that's where it Yeah. Not only do you need product market fit, you need like market channel fit.

You, you, you, you need market channel fit. Like is like, is this channel the way that you can actually, be successful in telling your story, promoting your brand, promoting your product. Does it like, does it even work? Somebody was asking me about like, so there's this guy, startup founder hit me up for advice.

Uh, is SEO worth it? Um, he's solving a, uh, a old problem with a way, a super new way. Of of doing things. Let me just see if I can summarize what this is.

Um, AI simulations for hiring. You can see candidates. Yeah, you can see candidates in action before you hire them. AI simulations that reveal what interviews can't get.

Real job performance and realistic scenarios cut bad hiring by 70%. This is a whole new way of solving an age old problem. Uh, and if you think about SEO, no one's searching for this. There's no demand.

So you would entirely have to base your strategy on consensus and then flip. I. It's so tough to get that to work consensus and flip. I, I don't know, I gotta maybe brand that or something.

It's tough, man to do it that way. Ideally people would just be searching for that thing that, that, you know, that they want to do. And I think, I think, uh, pep Pep was talking about this too, like the, one of the most important factors for like being a successful startup is like go after a big market with a lot of demand already and just disrupt it. You know, the, the, I don't know if this ai, uh, interview thing is like, uh, I don't know if it falls into that.

You know, I don't know if that, that, that's it. So yeah, basically I told this guy, no, don't do SEO, it's a waste of time. We've mentioned brand a few times, and one of the things that I find really interesting is just like how brand has had a bit of a renaissance as a concept. I.

in the, at least in the LinkedIn, uh, echo chamber that I inhabit. and you've talked about that a few times in your posts about like SEO, maybe how brand influences it, or SEO as brand in a time when, you know, you're not necessarily attributing the click anymore on a page. How do you, how do you think about that? Um, well, obviously, you know, if you have a steady stream of branded search demand and volume and people are.

Looking through to your site from branded search, that's a tremendous benefit, um, because it almost guarantees to an extent that you're gonna be able to maintain, traffic levels. Now, some of that branded traffic is not like buyer traffic. It might be investor traffic. You know, I need a job traffic.

But still, it, it's, it's good that you have that brand moat because. And, and a lot of sites that I audit and do reporting on, you know, we work on a lot of content. What we find is, let's say we did 50 net new pages in the course of a year, um, or it could be a mix of up content updates, refreshes and net new. What ends up happening is, let's say the site has a large repository of content.

Well, we only worked on 50 specific pages, but the site has hundreds of pages. Those 50 pages that we worked on where impressions and clicks actually grew, just as many that we didn't work on floundered. So it actually offset everything. It actually, so in a previous world, you wouldn't have so much floundering, you would have more consistency 'cause things didn't flounder as hard.

So then the 50 new pages that you built or refreshed would grow a lot. And then brand would ideally maintain its balance. And so now you have a story to tell where, look, traffic is wow, doing super well, but now what's happening is those other pages that are floundering because of AI overviews and such. So you're working twice as hard to stay where you were, you're, you're right where you were and you did all that work.

So now that leaves, if you don't, if you're not good at framing it, what's gonna happen? Executives are gonna say, damn. You know, we're investing, we did 50 new pages. Content traffic is exactly the same level.

Well, you have to look at the back end of that. What is happening on demo requests, uh, what is happening on overall site engagement. it's very likely that those pages that floundered were maybe top of funnel informational pages that weren't doing much anyway. And so basically what's the point on brand if you have that brand stability keeping you at at a basic level, it's really just gonna help you round things out overall, and so that that is really important.

So I had to like, summarize and tell me if this is accurate or not, what I'm hearing from you about SEO today? on page signals matter a lot less than, uh, off page signals, mentions, you know, references, citations, It's not, you know, the, the grow quickly that it might once have been, certainly not the place to start. And you're really fighting for a much smaller share for a piece of a much smaller pie, let's say. Because a lot of the things that used to be answered via search and a click to a website where maybe someone could say, oh, who's this now that I'm here is now answered directly, zero click in, uh, in the first page.

And, and so you're really just competing to appear in that consideration set for these lower, buying intent type of queries. Tell me if I got all that right and where I might be wrong. No, I, I think a lot of it is good. Um, y you know, I think I would summarize it for you like this.

let's think about top of funnel, middle Funnel. Bottom funnel. I. In the new world, this is gonna be the best way to summarize everything.

What's getting killed the most by far top of funnel that's just getting wiped out, right, for the most part. And so what you can think about then is previously all this traffic that websites were getting was really a lot of top of funnel, so it was tire kicker traffic anyway. Why do you think there's so many reports now of traffic down CTR down, but demo requests up, revenue up because AI is just wiping out traffic you never should have had in the first place. There's no business for you ranking for what is X and all this other top of funnel stuff, right?

This is a quick, remember the thing I showed you on Claude Leak? They don't want you gonna a website for that. They just wanna serve to you right there. Bang.

All right, so that actually makes me feel a lot better that the clicks we're getting are not gonna be as many tire kickers. That's good, but it's just so hard for everybody to get on board with this reality of less traffic. Everybody just punched in the gut over less traffic. It's like, who gives a shit?

You know? It's, it's better quality. All right, so top of Funnel Check. We've covered that.

We'll skip middle of funnel because I think there's a lot to be unpacked on that. Bottom of funnel as we saw with the Aura Example and many others that we talked about are really big on third party citations, reviews on Gartner, some of the affiliates that have survived, um, the pay to plays and then, just overall like brand pr, getting coverage, stuff like that. Of course you have the X best list if you can form some kind of collaboration and partnerships with companies that are doing a lot of content marketing.

That are ranking for those top X best tool reviews, get on there somehow. That's also good. But as I showed you, often that is just a factor of being a big brand. But anyway, at the bottom of the funnel, it's really becoming more oriented to third party, especially if you are in a hyper competitive high volume category.

and then you have to focus on Gartner G two, like those peer sourced review sites. They, they matter. No way around that. And so where does that leave you with a lot of opportunity?

Middle of funnel, that long tail query, fan out, nuanced, you know, is it worth, is X worth it? You know, bring those new aspects of content, to, to the answers, right? Like, don't just, uh, go with the consensus. Maybe you can, maybe you can flip it here.

I think there's a lot to be unpacked in that middle of funnel. To me, that's where there's gonna be a lot of SEO value. I think that's the way I would think about it. Top of funnel cut, bottom of funnel, really third party heavy.

Hit the middle of funnel hard find, middle of funnel and nail it. That to me is, is really where it's going. I. And middle of funnel being those people that they're like interested in the category, they're not yet at the point of like, which X should I buy?

But that place where they're, they're also just not like, what is x? If that's where you can come in Yeah, perspective yeah, yeah, I would, I would use something called the problem identification chart. That's what I use. To sell my SEO services.

I mean, I, I have this down by, it's like in my head, I don't need to look at a spreadsheet anymore, but I actually have a spreadsheet called SEO Problem Identification chart. this is actually a technique from a guy named Keenan. 'cause I, I read a lot of sales books and my life, um, and it helped me. But like, this is how I would think about, you know, what is middle of funnel for any business.

So I'll just show it to you. And if anybody wants to make their own, just go read Keenan's Gap selling book, and he breaks down. problem identification chart. It's genius.

All right, so you start with the problem. You, you don't even think about top of funnel, bottom of funnel or whatever. You just start with problem. So in the cases of SEO, and actually I need to refactor this for ai.

But in traditional SEO, these were some of the most common problems. Organic traffic is flat or not growing, or organic traffic is not producing revenue. It's not attracting the right people. This, I would say today is actually a good thing if this is happening, but a problem was in the past, you know, organic traffic is primarily branded search, and we don't have any non-branded visibility.

So what's the impact, you know, not being found for a non-brand commercial search. For attracting the wrong people, it would be sales team working irrelevant leads, wasted effort on bad content. If traffic is not producing revenue, then it's missed revenue targets not enough pipeline for sales team, but it also overlaps with, um, sales team potentially working irrelevant leads, irrelevant content, stuff like that. And then where the goal is for constant strategy is the root causes.

Why are these things happening? And this is what, like content marketers and stuff, they don't get deep enough into their business to know this. If you can map out all the root causes of all those problems, then you're gonna have content for days. Um, so this could be a content piece, right?

something like, uh, ineffective content process or content targets. At the top of funnel instead of commercial keywords. Then another idea, lack of proper keyword analysis or content misaligned to search intent content decay, keyword cannibalization. You know, uh, content team doesn't understand customers and their problems.

A content piece could be, you know, how to better understand customer jobs to be done for better SEO output. Right. So right off the bat, this thing is just fueling ideas and then you could, you know, with AI you could accelerate this a lot. You could pop it into Chad, GBT, have it expand.

So that's where AI can also be useful. But I would, for middle of funnel, I would start with problem impact root causes, build out a problem, identification chart, go crazy. It's not stuff that has volume usually, but that's okay. love that take and I think it's a, it's a great place to, to wrap us up on, because it's kind of hopeful.

The glass half full perspective, all that traffic you're getting that never converted. It doesn't matter Anyways, there's this new kind of blue ocean that you could, uh, play in around the middle of funnel. And at least the path is clear on the bottom. Thank you for, uh, being our guide to the wild world of SEO in the age of ai.

really appreciate you coming on. Hey. Pleasure. really appreciate it and it's been fun.

Thanks again.