Ranking in the Age of AI: The Playbook for 2026 and Beyond

So I believe we're going live, guys. This is me quarterbacking Restream. This is our live streaming tool. So I hope it all works. Drop it in the chat right now where you're coming in from, what part of the world you're coming in from, what time is it over there? We've got a good one for you today. We're going to start in about a minute or so with my friend Bernard. I actually did a webinar with him last week. I was like, hey, just come do a webinar with us this week. And we operate in the same circles and we've been Kind of just talking about how, before we start, just before we kick this off, how the future of work kind of seems like a video game. So maybe we can get into that. But we're going to be talking about ranking in the age of AI. Okay. The old stuff that's new again, the new stuff that is new. Software as we see it, is software in trouble? And software as it helps SEO and marketing and kind of maybe contrarian takes we have on things too. And so we'll Last week when we did this, it was much more of a conversation. And Bernard, last week I did 3 webinars or 3 presentations or 3— I did a couple speaking things. I'm just like, you know what? I'm going to try going completely raw with no slides at all. And I think it worked out. Anyway, all that to say, Bernard. So Bernard, I'm going to do a quick intro for you and you can intro yourself as well.

Cool.

He's a friend who runs Clearscope, awesome SEO software. And I would say they've been ahead of the competitors for a while. Actually, someone that we are— oh, do I need to add you to the team? Looks like you disappeared. Why did Bernard disappear? What the heck? Add to stream. Okay. Yeah. Well—

Oh, I'm back.

Yeah. Someone moved you backstage. Sorry about that. So anyway, all that to say, I've known Bernard for a while. And they've been a good player in the space. And even someone we've interviewed recently that we're bringing on in about, I would say in a month or so, she actually cited that she loves using Clearscope. So all that being said, Bernard, do you want to introduce yourself? Anything you want to add to that?

No, just been in the SEO game for longer than I would like to admit, probably around the 2013 era. So coming up on 12, 13 years. Practicing, living, breathing SEO, just seen the whole landscape really shift from links to PBNs to quality content to user engagement signal. And now we're entering the foray of AI, and I think it's one of the most fascinating fields that's so fast changing, right? If you think about any Of the OG marketing channels. It's like email marketing's about the same. Social media is about the same, but SEO is like every other year some pretty crazy stuff happens. Yeah.

So I mean, with this conversation, I mean, maybe we can start, cuz you talked about PBNs, we talked about links, we talked about quality content, right? And so maybe we can just start off with talking about what's old that tends to work really well today. Cuz there's a lot of people coming into this AEO game, this GEO game, and they're saying they They feel like they've discovered something new with listicles. Oh my God, you got to do listicles. Oh my God, you got to do guest blog posts. And so you and I, we've been playing this game for a little bit. So I think there's a lot of confusion today. So let's just kind of get clear on what actually, what's the old stuff that works. That's a good starting point.

Yeah. I mean, links, PR, this stuff is really old. What PageRank was originally built on still works for the era of artificial intelligence, Because we need to trust the AI models to recommend good sources, good citations, and have high-quality content. And what better way than links PR? What do you think?

I think, okay, listicles work. Listicles meaning that a lot of people a year ago or so, everyone was spamming listicles, right? The best SEO software company, best digital marketing agency, and they're They're just spamming their own websites and leveraging their domain authority for that. And they thought they had discovered something new, not new. Now the LLMs are like, oh, okay, it's pretty easy to see that you're publishing on your own website and you're favoring yourself. And Google's always known this is the game, right? It's like, oh, now people are like, oh, you got to do PR. You got to get links to add to your domain rating. But also at the same time, you have to publish on other people's blogs. So now I have agencies and I do this the same with other agencies where we exchange with each other on these listicles, right? Link building is still going to be around. I think links are still a very strong signal even in 2026. And I think the old, more importantly, you and I as SEOs in the very beginning, it's like, oh, I should go reach out to your website, Bernard, and I should go try to get a link. But no, no, no, that's the very beginning. You should try to build a relationship because what Bernard and I have developed is actually a relationship over time where we actually have met in person. There's multiple angles to work with each other. And then you actually learn that that's actually how you do business.

It's true. It's true. And as you were talking, the other thing that's, what is that still exists that is still relevant is technical SEO. I don't know about the validity of the rumor of llms.txt. Maybe you have some thoughts on that, but it's just really simple, right? If the AI agents cannot crawl, render, or understand your content in a timely manner, then yeah, you're not gonna be surfaced. You're not gonna be visible on these different AI responses.

Yeah. Who do you think, I mean, maybe if we can give some people some practical examples here, maybe stuff you're doing at Clearscope or what, maybe what clients are doing too, but who do you think is doing AEO well? 'Cause last week I gave a couple examples of like, hey, CNBC does this over here, or so-and-so does FAQs, right? Like what are you guys doing? Maybe some customers that you can share are doing that.

Yeah. I mean, I think the nuts and bolts, right? Kind of the, the key takeaway sections at, at the top, too long, don't read, you know, kind of providing a few sentences about what the article is about. We've been seeing just correlation. I'm not gonna say it's like, okay, do that for everything.

Mm-hmm.

But it seems to help with, you know, getting your content cited within, you know, AI models. I think that what we've been testing as well is video, like native original video, right? Whether that's the author of the post that's just sitting down and saying, hey, you know, I wrote this and, you know, here's what to get out of that. Embedding a short, like less than 1-minute clip.

Mm-hmm.

You know, the reasoning behind that is again, to basically prove that the piece of content that you've created is original, unique, you know, human, at least assisted stuff. So I think those things come to mind. I've heard of a variety of other interesting tactics with varying degrees of validity. But this is kind of, I think the Metehan example comes to mind. He was the one that popularized the idea of replacing the normal share buttons. Like, you click on share to Facebook or share to X with AI buttons where it's like, summarize this on ChatGPT or summarize this on Perplexity. And he's claimed that he's found some reasonable Although I'm skeptical about the longevity of that kind of a tactic.

Yep. Yep. I mean, if I were to simplify this, I mean, you're trying to make things easier for the search engines or the answer engines to scan, right? So the key takeaway sections, the FAQs, maybe you're talking about the bullets that you have on the top, or even clear subheadings, right? That type of stuff.

Yeah, totally. I mean, lots of people do that now to the point where And, you know, I guess, you know, one of the themes that we're talking about here is, you know, what is old is new and what is new is old. And it's, it's really a tale as old as time, right? Somebody finds out some, you know, loophole in the system and then they, they exploit it and then everybody starts doing it and then models suppress the validity of that signal. And then you get kind of this full circle of what's old is new and what's new is old.

Yeah.

And so I think we're getting back to that whole like, okay, well, answer the question As quickly as possible. You know, that's best for users, it's best for agents. And, you know, providing summarizations of key takeaways is also good for chunking and passage similarity. And, you know, just fill that with as many, like, entities as you can, and that's going to be good. And I wonder, you know, like, what the models are going to do once everybody starts doing that, you know. Yeah.

Yeah, usually. So actually, maybe if you can describe what chunking is for, for those that don't know.

Yeah. So chunking is the fact that when models respond to users for prompts, what they have to do is they have to look at A, their training data. So their fundamental knowledge of what they believe to be the topic. And then they do grounding with Google search, which is to say that they went then The models want to make sure that the content that they're about to give to the end user is factual. It's correct. It's up to date. Throughout all of this, what the model is doing is that it's selectively looking at pieces of content, more specifically paragraphs or sentences that contain the right entities that will build your particular answer. So chunking then is this idea that models are essentially looking at not your entire content, but passages, paragraphs, and sentences within your content that will then help them produce the most relevant and accurate answer to the end user.

Got it. Thanks for that. What do you think now? I mean, okay, I started in SEO. I'm assuming you started in SEO, but We've learned always over time that you can't just rely on SEO as a channel, but it seems like the people who are scared the most right now are people who have just relied on SEO. So how are you diversifying or should you even be diversifying right now away from SEO? How are you thinking about it?

I mean, I think the moral of the story is that if you have a channel that produces 30, 40, 50% of your overall customer acquisition, it's always safer to diversify from that. So SEO has conveniently been a very strong channel for a variety of different industries, business models, and that's why it's been such a lucrative channel to pursue. But now more than ever, I think we're seeing the cracks form in terms of at least the traffic component of what SEO offers to brands. So I would say, I think what we see a lot of people turning towards is their own communities, like walled garden communities. I think this is like private Slack channels. You had Traffic Think Tank, which got bought by Semrush and HubSpot bought like The Hustle, right? Basically a lot of these different software companies are recognizing that communities are very important because communities will then trade notes, discuss best practices, and in theory recommend their platforms, tools, and services. So I think communities is probably one of the more emerging channels. And you know that as Reddit as the go-to biggest community, but we're seeing a lot more fragmentation, whether it's Circle.so or these different private Slack channels, private WhatsApp groups, even just like iMessages between, you know, a network of, you know, really smart people. And I think that's kind of in a lot of ways what people are defending themselves against because we're seeing just this huge inundation of AI slop that is just pervading, you know, social media. It's pervading written content. And the most, the biggest problem with it is that you just don't know what's true. factual and if the original author even had anything to do with that piece of content.

So when I look at our marketing strategy right now, YouTube's a part of it, right? Webinars work. To your point, live events work, whether it's speaking at a conference and then hosting a little dinner afterwards or workshops and things like that. And obviously email works as well. And so there's a handful of things in our stack. And I think there's building products and splintering things off and giving the features away for free. What's working well for you right now?

Yeah, I think to just summarize what I feel like I've heard you say, it's honestly, it's credibility. It's things that exude credibility, trust, and respect from the people that you're trying to get in front of, right? YouTube works for us because we're here live streaming or recording, And the audience knows that this is not some AI slop that's deeply wrong or whatever. Email works because at the point you've subscribed to our newsletter, that means you trust us and you want to receive information from us. And so I think what we're seeing work is just a lot of the high trust, high credible things, live events, of course. You cannot have robotics replace that, at least not anytime in the short run.

Yeah. And I guess I'm curious too, like I'll share a story with you. The question, the prompt here is what happens to SEO work in general? And I'm not talking about individuals, but just work in general. So last Friday I was talking to one of our internal SEOs about our internal marketing strategy. So we have pretty good domain rating or domain authority on our website, Single Grain. And I was like, okay, well, I want to make more of these programmatic pages, especially because OpenAI is coming out with ChatGPT ads. I was like, for over a year, I've wanted these pages on my sites, like ChatGPT ads, ChatGPT ads agency, ChatGPT services, ad services, all these permutations that you can make. We hadn't gotten it up. These are simple programmatic pages. I went to him. I was like, hey, how long is it going to take to get these pages up? We want to make sure that we strike when the iron's hot and these pages are up. We're capturing that value. Ideally, we would've done this months ago. And he's like, well, this is Friday. This is already late Friday. He's like, it's going to take me till Monday to do the keyword research and then until Friday to get the articles up, which is already pretty fast, right? Maybe in the old world, but I was like, you know what? F that. So I opened up Manus. I didn't even use any tool or just the general agent. I opened up Manus. I was like, hey Manus, I want to rank for, I want to attack this ChatGPT ads agency type permutations for my website, singlegrain.com. That was the prompt, the BS prompt, okay? Whatever prompt, okay? And then what it did was it actually did the research for me. And I said, hey, I want these. I have plural template pages like services, agencies, whatever. Those are more listicles. And then I have singular like agency. So I have 2 templates. I was like, I want it to follow these 2 templates over here. I want you to give me a couple keyword examples and then show me 2 examples, okay? It literally did that. And then it made 150, a parallel process, 150 of them. I spot checked a couple. I'm like, dude, I just did $45,000 of work in less than 30 minutes. And so my point of saying this is my team's on top of it. This guy is like, he's got other stuff to do. So there's no shade on him at all. And in fact, he's seen the light on this now. But what do you think happens to SEO work now? You have a product that helps people with the work. What do you think's going to happen?

Yeah, I think I think what you're talking about is very interesting because I think analyst-level work, right? This is the, okay, let's export all the Google Search Console queries and then let's look at keyword cannibalization, or let's prune underperforming pages, or let's detect which pages have technical errors and how important it is to fix whatever it is on those specific pages. I think all of that work just out of the box, like you are mentioning is really easily done with your typical AI agent. I think that component is definitely being replaced, automated the most. However, I am not sure, at least currently, if the models are good at the A, the strategy, right? They can't, you can't just dump a domain in and then just be like, fix my SEO problems, right? It's, they're gonna make a little bit too many assumptions on You know what that means. And then B, it's also not really, at least currently replacing the work, like the, the, the actually going through and changing and updating and, you know, fixing all of the issues and the problems, right? I mean, yes, it will give you like outputs, but if you're working within any CMS, right, it's not so easy to just copy and paste in an updated article. You gotta copy and paste it. You gotta do the internal linking. Yeah. Make sure the imagery. is up to date. It's a lot of manual work to still go back into your site and make all of the changes. But the analyst-level stuff, right? This is like Screaming Frog for technical stuff, Keyword Insights for topic clustering, all these different permutations. I think we're seeing agents and AI really shine in that particular ecosystem. And I I have a feeling that it's gonna be the CMS, the content management system, like the Webflow and the HubSpot and the WordPress that is really going to be where the value like happens because that's like within the workflows, right? I'm imagining this future world where you just interact with your WordPress, right? And you're just like, keep these 10 pages up to date. And you have read, write, edit permissions to do whatever you want to the page. I think that that's actually like the real gain. And I'm sure tons of people are working on this, including WordPress and Webflow and all these different things. But I think that that's the future state of like, okay, this is now 10x better. I think replacing analyst work, being able to do things faster, at least from an analysis perspective, is good. I think the real game changer comes when CMSs, whether it's through plugins or natively, can just run things on their own. What do you think?

Yeah, I think it's— we like to call this the execution layer. So we have single grain, but I have the domain single brain. And single brain is like, it'll come up with the insights and then that's nice and all that, but can you just do it for me? And I think a lot of software is probably going in that direction. I also don't think I don't think that the high agency people are going to get replaced. I think what's going to happen is these people are going to, to your point earlier, we were talking about playing this like a video game where you have all your agents bringing the work to you and saying, this is done. It becomes more virtual, more virtual, like you're wearing VR glasses. I think that's going to happen. I think this is only going to give us more time to do high leverage work. The example I just gave you with Manus, which is again, like a general agent, and you can use whatever tool you want. For those of you listening, I think is just the tip of the iceberg. And that's why at our company, which I mentioned to you last week, but our mandate is that everyone needs to be proficient with Claude Code within the next 6 months or so, or we just can't move on. And that is a big ask, but I think we just talked about the ability to do a lot more than you're actually capable of right now. I mean, before we were talking here, maybe 20 minutes before, I was in Claude Code. So yeah.

So I guess my question to you is how are you defining proficiency in Claude Code?

Yep. So we actually, I have, I actually have a complete table on the different levels on, you have capable, you have, oh, sorry. You have unacceptable, capable, and then you have adapted and then you have transformative. So transformative would be like you are actually building tools. Like we have a tool called the Deal Reviver, which will scan all of our lost deals in HubSpot over the last whatever time period, and it'll actually craft messages custom to that person and also look at recent funding news or anything that might've happened to that company and have a draft email ready for the human to review and hit send. That to me is not just high leverage. It's high leverage because it adds ROI. It's not AI theater. And so every team needs it. We're doing these hackathons right now, tomorrow and Friday where we're locking down the company. And so all that to say is for each team, it's defined out and then we're going to help provide the resources to get there. But at the same time, you can bring the horse to water, but you can't make him drink, right? And there's a lot, I can sense resistance with people. Oh, I'm so busy right now. I'm so overwhelmed. But I'll tell you what, I don't know if you've seen this, Bernard, but I have a lot of founders that message me and they're sick and tired of the overwhelmed. And they're just like, dude, you cannot be overwhelmed anymore, right? And my CTO was telling me that the other day. And so I know a lot of people listening to this don't want to hear this, maybe do, but this is the reality. And so my whole thing is like, guys, I'd rather us not have to go through this round of cutting a bunch of people. I'm going to do what I can to bring all the people with us. And we're going to— that's why it's like weekly office hours mandatory, hackathons mandatory. We're going to stretch them out. So obviously I'm very passionate about this stuff right now, but I think it's very important that we bring our people with us, Bernard.

No, I absolutely agree. I mean, I think we're— I like to make the metaphor, right? It feels like we were basically plowing fields with horse and bulls and stuff. And now all of a sudden, right, this tractor comes along. Yeah, it's scary and people die because they jump in front of the tractor and it's this weird machine. But at the same time, it kind of feels like we are at this intersection of, okay, well either you get in the tractor and you 10x, 50x your productivity, or you keep plowing the fields by hand and you get left behind. So I—

It was funny. It's not really funny, but so the Shopify founder was saying this, right? And I was like, oh yeah, this is actually true. If you went through 51 weeks of 2025, okay, 51 weeks, the year's basically done, right? But week 52 was probably the most important one. That's when everyone started jumping on this stuff. And that was when people are on holiday break, right? And so in the world that we're in now, things are just changing every single day. And I don't know about you, but I'm more excited than I've ever been, I think, in a very long time.

Yeah. I mean, with that excitement, I'm also scared and nervous. You know, personally, I think that we've talked about this when I was visiting Miami that one time, but I think there's a lot of implications on traditional white-collar work. And I'm feeling like there's going to be a lot of labor shortage, you know, as a result of people adopting AI becoming a bunch more like productive. And I think I'm nervous and scared for, you know, the parts of the world that, you know, aren't able to adapt. And I don't know, you know, what that means for them. And yeah, it's just kind of, I think, an increasingly more dystopian world that it feels like we're moving towards.

Yeah.

Where, you know, everything is— I think, like, you know, throughout the era of the internet, For the most part, everybody abided by the, it is true until proven false. But given, you know, the release of general artificial intelligence models across video, text, imagery, so on and so forth, we're now moved into a world where everything is, you know, false until proven true. And I just feel like there's a lot of implications of, you know, all of this that just does not feel great for society.

Yeah.

And so that's how I feel.

Well, I'll tell you what. So let me, there's actually the founder of Groq with the Q, not the K, the chip company that sold to NVIDIA for $20 billion. So I want to get your take on this because I was actually, we're doing a, I was, Neil and I were doing a podcast yesterday on this. And I think it's actually a contrarian view on the future. And then maybe after that, we can kind of move into software as we see it. Because a lot of software stocks are collapsing right now and we can talk about how we look at software helping SEO. and also marketing in the future and kind of just, we can shape the conversation around that. So did you see what the Grok founder said? You didn't, right?

I did not. No.

So the context here is this. So again, sold this company for $20 billion, sold to Nvidia, did well. And he actually thinks that what's going to happen is obviously this little, this cup over here, okay? This is going to become, it's deflationary. The cost of buying this cup is going to go lower and lower over time because you're going to have robots building everything for you. Now, at the same time, because that's happening, working is going to become optional, right? So working's going to be optional and people are going to opt out of working for others because they have a lot of abundance already, right? And maybe everything you need, you're going to be able to get. Now, which then leads to people are thinking, okay, a lot of people are going to be fired, which might be true in the short term, but in the long term, there's not going to be enough people to work on things because Elon actually said this piece. He thinks in maybe the next 5, 10 years or so, I think there's going to be job displacement for sure. I think there's going to be social unrest.

Okay.

I think as we go through this adjustment, but I think afterwards is hopefully we get to afterwards, it's going to be prosperity. Right. And I think what's going to happen, because you and I, Bernard, like, I don't know about you, but when I was younger, I was like, I don't want to work. I just want to sit on the beach and do nothing. Right. But then you quickly realize, like, after you've kind of crossed a certain threshold, you're like, no, it was never about that. It was always about being useful and doing something that you wanted to do. So I think we're actually going to create an abundance of high-skilled people, entrepreneurs doing the things that they love doing. And that's what actually gets them to go all in, not doing jobs that they don't like doing. So that's a thought for you. What do you think?

Yeah, I think there's a lot of assumptions that that model builds on top of, but I would say for a sliver of people with high agency, with the setup. And what I mean the setup, it's the hand that you got dealt for life. And there's this Y Combinator post that lives rent-free in my mind. You might have read it, but it's the one where it's like carnival. So it's like if you got dealt a good hand, grew up in a middle class or upper middle class environment, Then life is a lot like going to the carnival. You go, you pay the price of admission, and then there's those games that you play, right? And for upper middle class people, they can afford to basically throw the dart and try to hit the bullseye, you know, like 3 or 4 times before, you know, they have to give up their dreams and, you know, go and work a normal 9 to 5, whatever, right? And then, you know, he makes the analogy to say that, like, You know, if you're born into a very great hand, you know, ultra high net worth, so on and so forth, you basically get to just sit there and throw the dart until you hit the bullseye. And, you know, that's just like the setup that you got. And then his last point was basically like, oh, and guess what? You know, everybody who got dealt the most like unfortunate hands, whether, you know, just kind of being, you know, dealt into like a bad family dynamic or poverty or so on and so forth, like They don't even get to play, right? They're the ones working the carnival. And so anyways, I say that because when you say, you know, what that is like, what you see, or at least what the founder of the Grok sees, it's like, yeah, I see that the people who get to play the carnival, yeah, they got— they have a lot more chances to play to hit the bullseye, but there is this huge percentage of people that don't even like Get there because of the hand that they were dealt. Yeah.

I think what's going to happen is the people that didn't get to participate are now going to have more shots on goal to participate. I think I'm a big proponent of equal opportunity, not equal outcome. And we don't need to go into politics or anything, but that's how I see it. I think we're going to have more opportunity. I think more people are going to be able to see the light that it was always about being useful and building something that you wanted to build to help other people. I think the world didn't allow for more entrepreneurship. I mean, everything around that is built by an entrepreneur, right?

Yeah.

And so that's what's exciting to me. So let's talk about software because a lot of entrepreneurs built these software companies. You look at Snowflake, you look at Workday, for example, you look at HubSpot, Intuit, they're all tanking right now. Okay. And so you and I, I mean, we're in software, right? And then maybe we can talk about how it relates to marketing, but it's just like a lot of people are tweeting now. It's like the death of software. Oh my God. SaaS is over, blah, blah, blah. What's your take on that and what's your positioning going into this new world? Because what was hard to build before, you actually have more abilities to build faster.

Yeah. So I think it's a great point and I think the devil lies in the definition of software, right? I think that you have software and I'll just say Salesforce on the right-hand side and HubSpot and these different really robust Platforms with very fleshed out ecosystems. And then you have software on the left where it's Screaming Frog or an email delivery checker. I use Neverbounce or a sequencer, which is like outreach.io or whatever, which is software, but more of a Okay, this just does one thing and it's more of a feature or, or an add-on or, you know, whatever it is that you want to call it. So I say that in the sense that in the pure definition of, okay, it was just a feature or, you know, it did one thing really well. I do think a lot of that software is at risk because, like you've experienced yourself, Claude Code and these new AI like Lovable, so on and so forth, they're really good at building that. thing that just does one thing and you can basically just build it, use it, and then throw it away as needed. And so I think that class of software is at risk just because of how displaceable it is. But I think software at the Salesforce and platform level, I think is actually not like inherently at risk. I think that a lot of AI applications are going to be layered on top of the Webflow, right? The CMS or the HubSpot or Salesforce, like the CRM level, right? Because there's actually a lot of meaningful data and like interaction that these platforms have accrued over time. You know, Salesforce owns Slack, it owns, you know, all the Atlassian, you know, those, those kinds of things. And I think that class of software. stands to benefit a lot from artificial intelligence, even though perhaps the markets are not clearly reflecting that. What do you think?

I think to your point, systems of record are gonna get more valuable. So like Slack, for example, I pull a lot of information from Slack or HubSpot we use or Salesforce, whatever it is. And that's why Salesforce is protecting a lot of their data that they have, 'cause they know it's gold. I would just say that I think, sure, a lot of people are like, you can just build Salesforce, you can clone it immediately. Think of marketers are making a bunch of product slop right They're just using this to vibe code a bunch of BS. And that's fine. Experiment with it, but just understand that a lot of stuff that you build, it requires maintenance and you're going to have to keep compounding on that software too. And the people who use Salesforce at an enterprise level, they can't afford to have any downtime or any issue because that can cost them billions of dollars. And so I think there's going to be room for— always going to be room for enterprise software. I think people are always going to also want to pay for customer success as well, like a managed service, that type of thing. I think we're— I like where we're at right now at Single Grain because we have the agency part. People are always going to want people to do the work. But I think from a marketing angle, I think the ability, for example, you know what I did the other day, Bernard? One of my friends was talking to me about how he sources 8 and 9-figure deals right now. I'm like, how do you do it? He's like, I just do this and do this. I was like, oh, I have an idea. Okay. I went into Cloud Code and I have a deal Manufacturer now. So here's what it does. It pulls from all of my LinkedIn connections. It pulls from all my Gong calls, all of my HubSpot calls, all of my emails. It's hooked into my Google Calendar as well. So it knows everything. I'm like, hey, okay. So, you know, and it's also hooked into like a news source.

Okay.

So it sees who's raising funding, who are the hot startups right now, whatever. And I'm like, hey, who's connected to OpenAI right now? Like, who am I connected to at OpenAI? It lists 4 people. It shows the levels and it shows the angle that I should be attacking with. And then it shows the LinkedIn message and also the email message. Why don't I just give that away for free and collect leads from that? I think as a marketer right now, if you're very creative, bringing this back to SEO, what are the things that you can create to give away in your niche, whatever it is exactly, and then just use it to collect emails, drive traffic towards it? That's how I look at it. I mean, I also look at Clearscope or I look at ClickFlow, for example. What are the features that we can splinter off and give away? To drive awareness, right? Because a lot of these things, everyone's just copying each other in terms of features. And so in my mind, it's like, how do you just make sure that you— I think the ultimate moat, people like to say distribution is the moat, but I like to say ultimately what matters is people like to say taste, right? So what is taste exactly? Taste is the things that you say or the things that you create. And in my mind, taste is a lot of it's designed, right?

Yeah. No, I think That's a very interesting point. I guess from a marketing perspective, I think of that as lead magnets and yeah, I mean engineering as a channel is always interesting. I think Ahrefs does this really well, right? They have the plagiarism checker, they have the AI detection checker.

Of course.

And this and that. And I think now more than ever, You don't need a growth engineer to make those things as much. But if you get on the Claude code, I think interesting things that people probably want to know are going to be insights. I think about the Semrush sensor, but how is volatility within, say, citations within the big sources? How's that playing out? Is Reddit on the up, down? Is Wikipedia on the up, down?

Yeah.

I think that's interesting. Query fanout. There's a lot of already vibe-coded to professionally coded lead magnets between Mike King and Dan Petrovic, where it just hits the API and it pulls back the fanout of queries that a prompt has for ChatGPT or Gemini if they were to exist. I think that these are good examples. You're gonna start thinking about vibe coding a lead magnet that does some processing. I'd say more or less like what I'm finding with at least vibe coding, and you've touched on this, is that it's really about the data, right? Like if you have API access to data sources that either you own and it's proprietary to you, right? Like Gongs, the HubSpots, the Slacks, or It's proprietary enough and not that accessible, whether it's behind a paywall like an X, which I think their API costs are really high right now, but that's value. You can take that and then you can have an edge over somebody who basically doesn't have access to valuable and insightful datasets.

Yep. That's the major point. So it's like, what is the moat ultimately? And so One, I think it's taste, but the other piece is data. If you were able to make these lead magnets and collect the data, so I'll give you another example. I'm messing around with the— we have the Gong MCP, Model Context Protocol. Just look at these things allow you to— these LLMs to talk to these tools. Gong, which is our sales call thing, and then we have HubSpot CRM, and then we have Ahrefs MCP as well. What I just did was I just queried it. I was like, hey, why don't you just look at the last 6 months of sales data and then just see what are people asking for the most? And then also cross-reference that with Ahrefs to see what keywords are trending upwards and then map out a bunch of keywords that we should be going after. So that in itself could be a tool, but here's the thing. If you're able to collect data around what people are looking for, you're going to see patterns in that. And I'm not saying that's the end-all be-all itself. I'm just giving you an example of a tool that you can make to collect data. That is going to be the thing that guides product development in the future. And that's going to kind of protect yourself and insulate yourself. And I think the key thing too has always been, It's the Bruce Lee quote, right? He fears the man that has practiced one kick, not 10,000 kicks, right?

Yep. Yeah. One kick 10,000 times rather than 10,000 kicks one time. Yeah, absolutely.

Yeah. So it still comes down to really hyper-focused. I can hype this up all we want. Bernard, we can hype this up all we want, but sure, Claude Code is powerful, but there's still a lot of blocking and tackling stuff you need to do within SEO. Within marketing in general?

Yeah, I mean, we touched on this, I think, during last week's webinar, but I think a lot of it is, is like, you know, how well are your, you know, how well-formed are your ideas and, you know, how, how good can you get the agent to execute on it is becoming more of the, like, the work that the humans are at least contributing to meaningfully because, you know, the agents are The ones going around doing what your bidding is, right? Yeah.

Maybe we can work towards closing off with this piece. I'm just wondering if you have any examples here. We go back to taste, right? What does taste mean exactly? So there's an example of a guy on— he has a YouTube channel and he is vibe coding till he hits a million. He's going to stream until he hits a million dollars. You see that one?

I have seen that one pop up on my social media. Yes.

So I was watching one of the videos yesterday and the guy's in his office dancing.

Yeah, yeah, yeah, yeah, yeah, yeah. He's actually vibe coding.

He's vibe coding and each of his streams is 4 to 6 hours or so. That is a way to drive distribution. That is a form of taste because it's a little different. If someone's going to do vibe coding to $100 million, I'll probably watch that too. I'm interested. And so to me, going back to SEO and AEO, SEO and AEO, it has The people that performed the best, it has been about taste. Even the spammers back in the day, they tastefully spammed their way to the top. They came up with new ways of doing these. The affiliate marketers were the most creative. I think more than ever now, that's going to become more and more important. And you gain that taste by experimenting with things and trying things that are outside the limits sometimes. But what do you think about Bernard? That Bernard, how does he taste?

Oh yeah. I think that's That's right. And it's back to that adage, right? What is old is new and what is new is old. And I think a lot of the tried and true is being spammed and replicated to the nth degree.

Yeah.

And I think as a result, what we're seeing the forefront of people pushing things in hyper unique ways that are interesting and Weird and in that way shareable and viral. And that gets you a link, it gets you publicity, it gets you eyeballs. And then, you know, I think one of the earlier points too is that I do think you have a great point in that, you know, now more than ever, the ability to create is being democratized in a way that we've never seen happen before. And so that kind of begs the question, right? If everybody can now create, how do you stand out? And taste is, I think, a very core factor. Data, if you have access to interesting data that nobody else does, that's another factor. And then how you take these building blocks and spin it into something truly interesting actually feels like it's out of left field, right? Like the guy who's dancing while vibe coding to A Million Dollars, that's like one example I've seen. This thing on, you know, where it's like a Claude instance that's like nurturing a tomato and it's like sold a tomato and there's like a meme coin like attached to it.

Oh, that's funny.

But like, but yeah, it like has like 12 different sensors, you know, humidity, soil, something like that. And then it will like add and like remove like sunlight and like air and like humidity. And it's like, I think it's crypto Twitter. Yeah, yeah. It's, you know, It's to say people are taking these things and, you know, like they're applying very novel ways to using this ability to create to their advantage.

Yeah.

And I think a lot of that originality and that creativity honestly comes from trying, right? Trying new things, experimenting. You'll see once you start, you know, dabbling with all of this how, you know, X could connect with Y or Y could connect with 4, and then you get to these strange and interesting outcomes. And then in that way you stand out and you're able to at least capture some awareness.

So last thought from you, are you bullish about the SEO industry in general over the next 5, 10 years or so? Are you bearish?

I guess it depends on what being bullish or bearish means. If we say being bullish is the fact that SEO will continue to be a large driver of value and leads. I think the answer is yes, but not in the way that you would expect it, right? Like in the past it was traffic, but in the future it's influence, right? And so at the end of the day, us as human beings, whether we're chatting with an agent or in the farther future, you just have an earpiece in your ear and you're chatting with your own private, like, instance of like an AI model that like knows a lot about you, you're still chatting with an AI and the AI still needs to A, have training data and B, ground their responses with some sort of factual or credibility check, right? And in those worlds, I think that search engine optimization based content is going to evolve into just being training data for agents or validating data for agents. And that is going to be important and it's going to influence what these models respond with. If you say SEO as we know it, sending traffic and all that stuff, then I'm very bearish. I think that traffic is going to be a thing of the past and you really have to think about SEO as a channel to influence AI models and not as a channel to get ranked number 1 and therefore get a lot of clicks onto your blog or product page. What do you think?

No, I think that's completely right. I think SEO has always been an evolution and I think the ones that don't want to evolve are actually the ones who complain the most on X. But the ones that have always evolved, they see it differently. They see the opportunities and they see that it's not But 15 years ago, maybe it was very, yes, Google was the only game in town. But now we talk about the everywhere game, right? Whether it's creating software or whether it's attacking YouTube or what, if it makes sense going on Reddit, there's all these different channels to attack because attention's so disparate. And so I'm excited about the evolution because if it had continued to be the same thing over and over, it would've been boring. It was too easy to game already, right? People knew it. And so the game will always evolve over time, and those of you who are adaptable who are always trying to learn, you're going to be successful. And so especially now that you have all these, I like to call these tools or weapons available to you, I think it's a very exciting time. So Bernard, I mean, other than people going to Clearscope, what else, where else should people go to find you online?

Yeah, I'm pretty active on LinkedIn. That's, you could just search Bernard Huang, add me, follow me there. And yeah, I mean, Clearscope, we do webinars. We had Eric on last week. And just lots of forward-thinking, really interesting conversations, especially as we find ourselves more and more in an AI-assisted world, whether that's personal or professional. It's definitely, I think, time to sharpen the tool, sharpen the tools and really go out there and see, you know, and try everything that's going on.

Love it, Bernard. Thanks so much for doing this. We'll definitely do this again. And someone asked in the chat too, if the replay will be available, you can just check out my YouTube Leveling Up with Eric Siu.