The God of SEO: Charles Floate on AI, Black Hat SEO, and Beating Google

Charles Float, thanks for coming on to the podcast. Let me, let me, let me tell you something. For maybe, for the last like two years, I have been recording your tweets on like LinkedIn and X, just with my phone, and then turning them into TikToks and and Instagram Reels. And dude, half of them go viral because the stuff that you say is so interesting.

Thank you. I appreciate that. Yeah.

Like, I've just, just through like Instagram and TikTok and YouTube Shorts, I've literally had millions of views of me just recording my my screen with you, with your tweets and your LinkedIn posts on on the screen. So, like, it's, it's great to have you on the show, and lots of listeners and and viewers of the podcast wanted you on as well. I'm, I'm hyped that you're here.

I appreciate you having me as well. Thank you. I actually have had a fair few like friends from high school and things go, "Hey, you know, I saw you on TikTok or like on Facebook or something the other day," and I was like, "Was, was it about SEO?" Like, "Why were you reading about SEO?" And half the time, apparently, it wasn't about SEO, which I thought was even more interesting.

Yeah. Well, you put up a lot of just interesting stuff, just all around interesting.

Um, all right. So, I wanna, I want to start actually with the end in mind. I think this is, this is a fun question. What should every SEO do when they finish listening to this, to this podcast? Aside from following you on X and and on LinkedIn and checking out your products and all that stuff. Aside from that, what should every SEO do when they finish listening to this?

Buy links. No, I'm joking. Um, I gen, I, I would always take everything into account that is you and my is my focus coming out of here. I think it's all good.

Yeah, there it's good. That's good. Now,

I, I would always take into account what you can learn from what you just took, took away, right? So, to a lot of people that I meet and stuff, and they're like, "Oh, I heard this idea. I heard this idea. I heard this idea." Right? And they put a lot of stuff into their own world, but they never really build upon things, right? So, right now, what I've been doing, especially from podcasts and things like that, is just downloading YouTube transcripts or exporting YouTube transcripts, taking all of the bits, or taking that transcript and all of my notes from the podcast, or from anything like that, or even just literally a summary, a paragraph of what I took away from that podcast, and then getting them to go and exfiltrate all of the actual key points that I want to put there. Right? Any podcast that really switches it on for me. And I think the last one I did was Neil Patel did a podcast with, I think Eric Si or something like that, and they had some really, really interesting points about AI and SEO and things like that. Um, but it was like an hour and 45 minutes, and I couldn't go and go and again, it's where AI makes it a massive growth hack. Before, you would have had to physically go through and jump through and try and find the bits that you were trying to remember and memorize, right? Whereas now, I literally just get it to transcript, put it into the thing, exfiltrate, put my comments, and then it goes through and just notes all of that down. And a lot of what I've been doing now is adding those notes into my knowledge base. And then that knowledge base is synced into a, and again, this is where it gets a bit technical, synced into my Superbase vector database, and then that's into my live model. And then I can communicate with that model. And it's just a continual database of my conversations, history, uh, information, transcripts of my favorite podcasts, and, um, talks and presentations. Even I've actually uploaded a couple, like slide decks that I've taken from some of the best slides and things that I've seen at conferences and things like that. Um, and I've just created this massive, massive knowledge base. And anytime I want to reference information, or anytime I want to try and pull things back, or even check potentially my own advice that I'm giving to other people, I then go and talk to this model, and it will give me bits and pull and pull bits and piece back. And I constantly get that deja vu moment where you're like, "Oh, yeah, I do. I remember that vaguely in the back of my head of trying to pull that bit of information." But the fact that I'm able to communicate with somebody who has that all-encompassing knowledge just allows me to very, very easily go and double-check that what I'm thinking and what I'm going to do is the correct source of or correct action that I would have done if I had got all of that information in my brain at the same time consistently. Right.

So you have, all right, so you have this like this massive database of everything that you're learning.

All the interesting takeaways that you've gotten.

It's all vectors.

And it's massive. Yes.

Yeah. And it's okay. Yeah. So, it's vectorized. So, I guess, I guess like the, uh, context window isn't really a problem.

So, context windows have, to some extent, gotten a lot better because you have like 10 million tokens and things like that, right? But it's degradation that's more of an issue, right? It's when you start like asking it nine different things to pull from. That's when it starts having some issues, some hallucinations, some stuff like that. Um, it's just when you're kind of, it's, to be fair, I would, if you put it into like AI as a person, right? And you give the person nine tasks in the same message. True, right? I, I do kind of give the AI a little bit of leeway when it comes to things like that. Um, but no, you are right in that it doesn't get everything 100% of the time, but it will get a hell of a lot more information that I could store actively. Right. This is the thing. I can probably store this information in like the deeps of the backs of my brain, but it would take me 10 minutes just to pull it and try and memorize it perfectly to be able to for it to be actionable, right? Whereas this is just a constant memory store. And there's so much information in this world now, and there's so much content, and there's actually there's so much good content as well, to be fair, um, that it's really difficult to constantly be pulling all of this stuff, whereas this is.

How do you prioritize with that amount of information?

So, I everything's split generally into PDFs anyway. So, it'll be like a 900-page PDF document or something like that, right? Per one, because generally the file size, like a thousand-page PDF or something, that would be the file limit size. Um, we tend to try and prioritize the PDFs into what they're about, right? And then the data at the top is the data that we want to be sourced for that the most. And the, and the further it goes down is the data that we want least sourced. And that often means that if there's a, if there's a really, really good podcast that I really enjoy, I will literally rank the podcasts in the PDF and then I'll just go, "Insert this new one, number two," right? And because this is better than the podcast that's number three, and I'll try and use that as a kind of prioritizing as a memory source.

Yeah. But but since, but since like you're forgetting that you're forget, you're listening to these great podcasts and then you're forgetting them. Isn't it hard to like go in there and rearrange things where something is like five months old?

Not really, because I annotate a lot of stuff as well, but okay. And I, I generally don't forget a lot of the like, a lot of the memory files that we're building at the moment is from the last two or three years of data. And even realistically, it's probably a lot, a lot of the stuff that I've been collecting, I've only been doing this for like the last six, seven months. So a lot of the data, well, because the models weren't good enough for it to actually work beforehand, and, and the context size.

Do you actually think you've been getting better results across your businesses from from doing this?

100%. Like, not only, and I'll be super honest here, right? Um, I'm not that great of a CEO. I'm not that great of a manager, right? I'm a great CMO. That is what I'm good at. I'm good at marketing, right? I don't necessarily enjoy managing people, and I don't necessarily enjoy sitting on Slack telling 10 different people what to do all day, right? Um, a lot of the time, if they do stuff wrong, I will ask the AI to reword into a cleaner way, right? But the problem is, ChatGPT has zero context about what I want it to reword around, right? So, if even if it's got the memory updates, even it knows about me, this kind of stuff, right? It still doesn't know enough for it to fully know this client's individual revenue numbers, whatever, right? Whereas my memory base does. So I can go through and have specific interactions around specific people, clients, hires, all sorts of things. Whereas I, I couldn't do that before with ChatGPT. So it really has allowed me to have a conversation with like a higher level of myself, almost, if you know what I mean? Somebody who's less emotional, more focused, higher levels of memory, more access, more focused on what the good information is versus the bad information. And this is another thing that I'm trying to bake in more as well, is negatives. So you said, "How do you prioritize stuff as well?" Well, I also am trying to put in negative information. So I'm trying to say, "Actually, do you know what? This podcast, I hated it. I really didn't like this one. And I, and I don't want you to be talking or tone or style like this, right?" Um, and I'm trying to do that more because I've actually found that that is a better way to kind of fine-tune and get better engagement out of your models than it is to actually give it positive information a lot of the time, right? Because it's kind of the, it's telling it what not to do versus what to do. Um, whereas it can, because it, it has so many pathways for it what it can do, whereas if you start telling it what not to do, it starts limiting some of those pathways to the, the more fine-tuned lane. Though sometimes it does overcorrect, which is a huge pain.

Yeah. So, we, we've been testing different models, right? And OpenAI models and Gemini, they're just too safe, to be very frank, right? A lot of them now, check like GPT-5, Gemini 2.5 Pro, whatever, all these things. If I start, unless I'm doing prompt injection, if I start talking about like Black Hat, I'm like, "Okay, I need a PHP, uh, or I need like a Java JS redirect script to like hidden redirect users, but the bot sees the normal page, right?" Yes, it's Black Hat. Yes, blah, blah. It's not illegal. It's not all these things. It should still generate it, but it doesn't, right? Right. Gemini 1.5 doesn't. Chat GPT doesn't, right?

I'm sorry. I can't help you with that.

Right. But exactly. Um, but other open-source models, they're more than happy to do that kind of stuff, right? And

To some extent, you can have ones where you have no limitations whatsoever. Like I, I, I, I don't know if you know Plenny, the the Liberator, Plenny the Elder, Plinus.

Yeah. Yeah. Yeah.

Yeah. The very famous like, uh, uh, prompt injection hacker, right? Prompt hacker. Um, he did a really interesting one where he got Grok to sing "WAP" from, um, what's her name? Nicki Minaj, and then teach you how to make meth. So that was a fantastic example of how you can really remove all of the limitations, right, by get it to think it's doing one thing and then actually it's doing another.

Um, so how are you actually? Wait, shoot. This is a great, this is a great segue. Okay. So, how are you, how are you using, uh, AI that doesn't have limitations to get SEO results?

This method of marketing is so effective, I had to make sure it wasn't against Google's rules before I kept doing it. It's a form of SEO I call compact keywords. Whereas most SEO focuses on putting up articles to answer questions, how, what, when, compact keywords focuses on putting up dozens of pages that sell to searchers who are actually looking to buy. These pages rank on Google and convert so much better than normal that when I discovered this years ago, I couldn't believe this was allowed. It's less work, too. The average Compact Keywords page is only 415 words. Compact Keywords is a 13-hour deep course on getting sales with SEO. A customer recently said, "Each lesson is dense with information. You're giving years worth of experience boiled down into 15 to 30 minute lessons with no filler or fluff. I feel like I'm gaining a new superpower." Compact Keywords is about setting up an SEO funnel that brings you sales for years and years and years. It works with AI. It's less work than traditional SEO and it makes way more money. You can get it now at compactkeywords.com. Back to the podcast.

Yeah, all the Black Hat stuff all day long, right? So like scripts, hit it, like cloaking, everything like that. Right now, I'm not coding it myself anymore. I'm just getting AI to to code all of it. Um, even domaining and setting up sites and like configuring WordPress scripts and installations and things like that. That's all AI these days. I was just AI agents to deploy everything, really. Um, so with Black Hat, it's changed in the fact that it's made me a lazy Black Hat, to be very frank, right? Like, five years ago, I like, right, I need to be a developer. I need to be a designer. I need to know how Google works. I need to know how all these algorithms work. I need to know all this stuff. Nowadays, half of it is done for me because I can just tell the AI what to do, right? Um, but again, you, you can't use the major models to do a lot of this stuff, which really sucks.

Yeah. Yeah. Yeah. I feel that, dude. I use, I use AI to code all the time. I, I have, I have made like so many custom plugins just on my own WordPress sites, and it's Claude that has done that, and, uh, yeah, it's like, it's fantastic. Um, I want to actually, so I want to ask a few questions from, uh, from people who watch the podcast. I have like, actually a ton of questions that people have, people want me to ask. All right. So, the first one is, uh, this is, this is from someone named Glitcher, who asked, "Tips for PBNs and making them not filtered out."

So, so the, so the filter, so PBN links are not getting filtered out. They're still worthwhile. So like adding authority links, or using social signals, um, or parasite link building techniques for ranking parasite pages.

So, with PBNs, the very first thing is just your domain name, right? Um, too many people try and buy expired domains, they haven't done due diligence, and then those domains end up having zero equity, zero history, zero anything. Um, right now, we're just mostly concentrated on age domain names, or because blogs have kind of fallen off a cliff because Google killed a lot of blogging over the last few years. Helpful content being the kind of stepping stone to to that. Um, we've found that actually there's tons of blogs out there for sale, or there's tons of blogs that are just dead, and that if you DM them, they'll say, "$500, I'll I'll sell you the site." Wow. Yeah. And a lot of the time, it's actually a lot lower than even the domain itself is worth if you're, if you're actually contacting them via contact forms and things, or email, or LinkedIn sometimes, even actually. Um, and a lot of the time, that domain could have been worth $5,000. You just bought it for $500, but it's actually a real blog, and you don't have to change anything. You just keep what's already there, and then you build off of it. A lot of the time, people are buying, uh, expired domains, repurposing it from a gardening blog to a casino site, and then wondering why it's not passing any juice, right? That's not number one, Google's got anti-spam systems to stop that in the first place. Um, number two, Google's a domain registrar, right? So anything that you can see in the back end of the domain, Google can see too. Um, maybe WHOIS history, you know, that kind of stuff as well, is is is discounted. Um, so when it comes to aside from the domain, actually building things out, it's just differentiating things. So, it's making sure that your, that your site is actually still active, and that it still has posts coming out. And if that means that you're using AI to schedule a post every week, do that. Um, trying to maintain link velocity. So, a lot of sites that when you actually build the, the PBN, you put it live, the links just slowly, slowly trickle away. The DR goes from 90 to 70 over the course of three, four months. And as an effect, you get link rot, you get link decay, that juice is significantly devalued, all of those things. So, we try as much as we can to maintain the links that are already there, but we'll actually try and build some new ones. What building new links looks like is very dependent on the domain and its worth. So, if we've gone and bought a $50 age domain, we're probably not going to go and build a $100 worth of links at it, right? We're probably going to go and find some cheap alternative to build maybe like $3, $4 niche edits and just put like 10 of them at the, at the, at the page. If the domain is $5,000, $6,000, that's when we're going to probably do maybe like a Google News press release blast just to get some freshness, get in Google News, get some high authority domains, plus probably do a few niche edits, maybe do, um, some pillow linking, blog comments, and stuff as well, just anything to kind of get your velocity back on track and also get the domain, uh, to be reactivated and kind of juice flowing. Anything aside from that is just due diligence when it comes to like WHOIS and who used to own it, if it used to redirect, if it had a Chinese farmer site spam on there, if it used to be an Indonesian casino, all of these things, right? Um, and you just want to make sure that the domain, like I said, still maintains history, still maintains equity, and still maintains the old link profile. If it doesn't do those three things, then it's not going to pass juice. And then that PBN is dead. So the only reason you build PBNs is for them to pass juice, and if that domain doesn't have the ability to pass juice, then it's not going to be a very good PBN.

Crazy answer. Uh, what are, what are huge mistakes? Some other huge mistakes that people make with using expired domains?

So, in general, most people are either using tools to to just go and find them, um, or they're using some sort of, uh, list. Like, there's, there's like sites like expired domains.net or something like that, which have just like these lists and things, right? Um, millions of other people have access to the same thing, right? They've all got the same information. The best ones you can do from what we've found, the best domains I've ever got from expired wasn't really expired. It's drop catching, right? We've, all the best by far domains that we've got from expired is just drop catching, um, for PBNs. The problem with expired.

Can you talk about that a bit more?

Yep. Let me, I, I'll start first. Um, so the problem with expireds is that you don't know if they will hold that juice equity or anything until you get the domain and test it, right? Um, it could have lost the juice. Google could have penalized it. It could have an unnatural site penalty. You can have all sorts of things going on, but you don't know until you get hold of the domain, set it up, and test it, right? Um, and from what we can tell, expired domains have like an 8 to 10% uh success rate. So, for PBNs passing juice, right? So, every 100 PBNs that you build, only 8 to 10 of them on on for expired domains are going to actually be able to pass juice, which again, is a really bad PBN. Um, when it comes to drop catching, right, it's, uh, niche dependent, or sorry, it's geo-dependent because it's all about your local registrar. So, .com, .net, .org, that kind of stuff, that's very easy. When it comes to like .co.uk or .com.au, or something like that, right? That's when it becomes a little bit more difficult because of local registrant, local registration requirements. Like, for .au, for you to even bid or own a .com.au, you have to have an Australian business license. So, that in itself makes it quite difficult to go and, uh, buy them. When it comes to everything else, it's pretty easy. You're just looking for domains that are about to expire, and then you're essentially going through and there's,

There's lots of lists, or what we normally do is we scrape sites, Wikipedia, BBC, whatever it is, right, for sites and try and find their expiration date being tomorrow, next week, whatever, right? And we'll drop catch those. But again, it's mostly because we're trying to target links on specific sites. Um, and with Wikipedia, it's even better because you can, you can scrape niche categories or niche specializations. So, I can be like, "Okay, I only want to scrape pages that have this word on it in Wikipedia for links, right?" And that means that, okay, casino, crypto, credit card, whatever, right? Domains that are actually going to have a high value. Um, and that's what a lot of vendors do. A lot of vendors will go for those specific areas, specific sites, specific pages, specific topics, because they know that those domains, even if they expire and they pick them up, and then they don't even have the PBN value, just those niches mean the domain is is valuable. And because it's an expired domain, you're picking it up for $12, $11, $9, right? Um, and then you can sell it for 100 times up.

All right. So, uh, okay. You, you actually wanted to ask you about this. This is my last question, but I think I can, I can bring it now. All right. So, we're talking about about, uh, tactics like this.

And you, you said recently on X, you said, "Most entrepreneurs sabotage themselves by chasing passive income. News flash, there's nothing passive about building an empire." So, how does like what you're talking about right now relate to building, I and maybe you're not into this, but I, I think you are, building long-term assets.

Mhm. So, I go about long-term assets a little bit differently as well. Um, I like, I knew about crypto, and I was actually blogging about Bitcoin and things in 2013, and there's a lot of people in SEO who still come to me and they're like, "You know, you were the first person who introduced me to Bitcoin." Um, and the only reason I knew that is because of the hacking world being involved in Bitcoin and them using it as the currency for for hackers, uh, originally, like 2010, 2011, 2012. Um, when I try and build long-term assets, I have different classifications, right? So, I've got my active assets, which again, is what I'm currently doing. I'm trying to build this business into a $10 million, $50 million, $100 million business. I have my passive assets, which again, I say passive income because it, I never said it didn't exist, right? It definitely does exist. Um, I get paid dividends all the time from Microsoft and things like that, right? I own a lot of Microsoft stock. Um, and again, I use that to buy more stock and stuff, right? That's the dividend cycle. Um, and then the third classification is, let's not say anything too bad here. Um, would just be short-term gain capital driving assets, right? To parasite projects, parasite pages, um, domain acquisitions to sell, anything where I'm doing like website flips or stuff like that. Um, and my 90% of the time, I'm going to be focusing on the big picture. So, I'm just going to be focusing on the business. Right now, it's Preswiz, right? We're trying to grow it to be like a $50 to $100 million business, potentially even go public. We're already closing in on a $10 million valuation, which is amazing since the business opened last September. So, it's pretty good. Um,

Can you share what it does?

Yeah, so it's an AI link building marketplace. So, we literally, uh, about a year ago, I had somebody approach me, Dan, who's the CEO. He is actually a former Counter-Strike professional player. So, he used to be a Counter-Strike pro. He was building this, building this linking marketplace. I'd already worked with a couple of the ones, and I wasn't really happy with the state of link building marketplaces. I thought that, A, the biggest ones out there, I'm not going to name names obviously, because competitors and all that, they'd gotten too lazy. They, they were price gouging. They were charging, you know, 300% of the cost of a link, right? So, you're paying a blogger $50, they're charging you $200, you know, right? It's insane markup. Um, and there was no innovation. So, there was no, there's no new tools, there's no new, you know, tests, there's no new methodologies, there's no link types coming out. Like, there's nothing, right? Um, so I was like, this business is worth billions of dollars a year. Like, the buying links and buying digital PR and buying all this stuff is a multi-billion dollar a year business, and nothing has changed in the last several years. So, I got approached by Dan. He was saying, "Look, I want to build this marketplace." I said, "I'll come, I'll come on board as long as I have control over the brand, the tools, the vision, the trajectory. You run ops, right? You do the stuff that I don't want to do, right? Which is managing people and managing clients, and I'll do the everything else." Um, and as a result of that, we put a lot of time and effort into making our entire outreach system, and when, and this is not just outreach, but also link delivery system automated. So, there's no human in the line anywhere unless you ask for one, right? So, from when initially, obviously, we have, we have a human set up the outreach campaign, set up the average AI, but from there, the, the AI is just going out and emailing. I think it's, we're close to like 10,000 people a day right now at the moment, which, yeah, it's a huge number of people. Um, and it's just trying to get them to sign up to be a publisher for Preswiz, right? Um, secondarily from that, when the, uh, client comes and orders a link through the marketplace, the, our AI agent sends an email to the publisher. It says, "Hey, look, this publisher's email, this, uh, person's ordered." It'll attach their guest post if they've, uh, put content, they've uploaded their own article, if they've ordered an article from us. Um, and it will send it to them. The publisher will then upload that guest post or change, upload their niche edit, whatever it is, into their site, send the URL back to that email. All they have to do is send a URL, and they'll have that very, very clear instructions. Just send the URL back to the, back to the email. The URL then goes back to the AI agent. The agent validates that it's the correct anchor text, the correct URL, the correct placement, the correct page. Once it's all correct in there, it will then go to the order dashboard and it will say that your order is now live. So, we've had as, so the only human in the middle is the publisher, right? That's our only bottleneck is the publisher. So, we've had links go live in two minutes before from when somebody's ordered to when it's actually live in their dashboard. Crazy.

Because the publishers on their email, they've seen it, they put it straight into the thing, set the validated by the AI.

Exactly. Yeah. And so, we've had customers who are like, "We cannot get this speed anywhere else." Right. Um, and in some industries, as an example, in gambling, some of the biggest parts of gambling is fixtures. So, it's like, you know, UFC betting is coming up, the football game is next week, and they want to rank specifically for that fixture, or they want to rank for that UFC betting term and things, right? Well, they need links tomorrow, right? And we're the only ones that can deliver that for them. So, we've ended up kind of absorbing tons and tons of the, especially I gaming gambling clients, we've got a lot of them now, um, a lot of the biggest markets there. So, we've been able to grow. We're close to $600,000 a month in revenue now. So, it's, it's growing massively. Yeah.

Great. And and how many employees do you have?

12, I think. Yeah.

So, 600K MR.

Yep. Mhm.

With 12 employees.

Yeah.

Dope. Y. That's cool.

And half the team's in Serbia as well. So, it's.

Really in Serbia.

Oh, man. I used to live in Belgrade.

Oh, yep.

Uh, how long ago did you start that?

So, Dan started in September, and I started working with him in December.

Last year.

Okay. Okay. Yeah. Yeah. Yeah.

Awesome.

Uh, okay. I want to, I want to change, uh, I want to go back to AI and ask your thoughts on actually ranking, like ranking on AI, getting visibility within AI tools.

Yep. For sure. So, I have got an all, all, you know, um, cards disclosed, I have got an AI ranking service. Um, so we're doing AI visibility. I, I, you know, there's this massive debate at the moment, go AI, AI SEO, whatever.

Yeah. What's your take on that? If there's a huge difference between, uh, what to do to rank on Google and what to do to get more visibility within AI tools.

So, I would say no, just because right now, so I, I would have said last year, yes, because GPT-5 changed things, right? So, I don't know if you saw this, but GPT-5 was the first model that had lower training data or lowering training data access or something along those lines. I don't want to get the, there's a correct technical word. I'm not sure if I have it. Um, but essentially, it's going to use less of its training data in its actual, um, commencement of the query than the last model, the previous model. Whereas every model before that, it would use more of the training data. Right? The reason for that is because that it wanted to have the model be more intelligent and be less reliant on data for its intelligence, and be more reliant on up-to-date and real current information. And so, how do you do that? You use search to go and find today's information. So, GPT-5 has started to essentially be a grounded intelligence, which just means it's an intelligent search assistant, right? So, all it's doing is going and 90% of the time, it's searching the web or searching its knowledge base or searching whatever and finding that information. And again, most of the time, right now, from what we can tell, it's searching the web, especially for queries like "best X" or "who is whatever," you know, things like that. Um, it's searching a lot of the time right now. So, I think if you go into Chat right now and put in "best SEO expert 2026," we should be number one, me, Kazra, duly, and people and stuff, the God of SEO.

There you go. We just did this stupid little manipulation experiment, um, a couple weeks ago, and with our own sites and some pages and stuff like that, and it ended up coming, and it's literally like verbatim putting our our list in there. Um, that's again, just more proof that it's grounded in search. So, if, if its information is being pulled from search, then it's a hybrid SEO model, basically, because it's all, all the data is still coming from SEO. It's still coming from organic search. So, if you can rank your pages in those organic search queries, then you're still feeding the visibility inside Chat and inside AI. The different models and the different companies have got different preferences, right? So, as an example, Perplexity loves YouTube, right? It loves YouTube like big, big time loves YouTube. Um, and it will prioritize YouTube for some reason over most other things. Um, I recently did like "best Black Hat SEO." Three of the seven topics were like YouTube videos in Perplexity. Chat GPT hates, hates press releases. It hates, uh, to some extent, it doesn't even like YouTube videos that much either, to be fair. It prioritizes Reddit and other blogs and like news, right? So,

It's really, dude, it's really interesting. So, for, I've, I've seen people manipulate. I don't know if you saw. So, there's this guy, Harper, he's been on the podcast two times already. He, yeah, so he, he used, uh, $6 press releases from AB Newswire to, to one, just get, uh, AI Overviews to say that his newsletter is like the best AI SEO newsletter. And then these $6 press releases worked for Chat GPT as well, and Chat GPT was literally citing, uh, one of the publications where the press release was syndicated to. But the interesting thing about Reddit, actually, is so, do you know Jackie Chu?

Yeah. Yeah. Indexy, of course. Yep.

Mhm.

Yeah. Yeah. So, he, so he was on the podcast last week, and I was talking about Reddit and Chat. He's like, "No, Chat GPT dropped Reddit." I'm like, "What? Really? No. No way." And I was looking, and I'm like, "Oh my god, Chat GPT actually dropped Reddit." What is going on here?

Yep.

And do you know why? I'm actually curious. I, I think it's something to do with that Google deal or something like that. And apparently Google is.

I thought maybe it was like the removal of the num 100 parameter.

No, I, I think it's something to do with Google's actual deal with Reddit where they don't want OpenAI having the information that they're paid for, because Google paid for that, the, they're paying like $6 million a year plus stock and all sorts of things and stuff. So, they probably don't want OpenAI having that information. Um, but no, yeah, the, the only difference really is the niches. So, if you, if you know, if you're dealing with something very, very competitive, it's going to be a lot more difficult to create enough sources and rank those sources to be able to take over the summary versus if you're like what I just talked about, "best SEO expert 2026," there was hardly any other pages out there already about that topic. So, it was very, very easy for us to go and rank those pages. And then because we are ranking number one, number three, number five, number seven, Chat is going to summarize or pull at least, you know, half of those, and it's the same entities over and over and over again. It's just reinforcing each other. So, it means that when it's pulling the same page eight times out of the 16 sources that it's got, it's pulling the same entities, uh, eight times in the exact same position. So, it's reinforcing it, and it ends up creating that same, uh, the same kind of list format in Chat. We, we've seen it literally just copy and paste from what we can tell. Like.

Yeah. Um, it, it was funny too, because, so, okay, I had on, right before Jackie Chu, I had on, uh, David Quaid. Yeah, David. David said that he wanted to actually rank for "God of SEO," and then he's like, "Oh, Charles Float ranks for God of SEO," so I'm going to take "King of SEO." So, he shows up for "King of SEO." And I think if you ask Chat GPT, who's like, "Who is the King of SEO?" It's, I'm not sure if it's Chat GPT or Gemini or both of them, but it's saying like, David Quaid. And it's probably, if you ask "God of SEO," it's going to say Charles.

Yeah, there's some Indian dude who's trying to steal it from me at the moment, but I don't know.

Really? Yeah.

It's a fight for it.

Well, I, no, I, I stopped using it obviously several years ago. So, um, yeah, he's put himself as Harsh Harsh Car or something like that. And then it comes out me and Bruce Clay.

Okay.

But no, yeah, it's, uh.

That's somebody else that I need on the podcast, I guess.

Yeah. And then I see, I see "King of SEO," primaryposition.com. That's David. Um, I actually, so, I, cuz we were talking about that, cuz I also am showing up for "King of SEO," but it's not for me, it's for, uh, my friend Yas, who was my mentor in SEO. And Yas, um, I went and worked at Dentsu, and Yas was my manager. And then I left, and Yas became head of SEO at Omnicom, which for a while was the most valuable publicly traded marketing agency in the world. And Yas was head of SEO there. Very, very, very close friend of mine. And so, I had him on the podcast, and, uh, and I called him "King of SEO." And then that was showing up, um, I don't remember what position, but it was showing up for, because it's YouTube, and YouTube is ranking extremely well, and it was showing up for "King of SEO," and it's like a term that not many people are targeting either. Not, I stopped targeting, like I said, the only reason I ever used it in the first place, um, was because Rand Fishkin on, um, on Twitter, like, this is probably 14, maybe longer than that, like 15, 16 years ago. Um, he, I was the first person he blocked on Twitter, basically. And it was, and it was, uh, it was because I, I basically, um, I'd ranked for his name. So, I did this whole case study thing where I ranked someone for his name, um, using like some, and it was like, I called it "Ranking for Rand" as like a case study, and just used Black Hat stuff. And I, I took the piss where the links that I was building were about like window cleaning and stuff like that, and I was just doing it to prove that you, there's nothing about relevancy, and it's all [ __ ]. Um, and Rand basically turned around on Twitter and replied to my case study saying, "Oh, you really think you're the God of SEO, don't you?" And I was like, "I'm registering that right through now."

And so I registered the domain name, replied to him saying, "I'm, I'm writing the case study for the new blog right now." And then he blocked me.

Oh, no.

So, yeah.

Oh, Rand, Rand has been on the show as well. Oh, shoot.

Yeah.

Oh, man.

So, like, he's the only person, I think, in SEO that from all the way back then still doesn't like me.

But really?

Like, genuinely. So, I, cuz I've spoke to like, you know, Cyrus Shepard and people like that who were.

Yeah, Cyrus has been on.

Yeah, exactly. So, like, I'm good friends with them all, like all of those people who are like former Moz staffers and stuff, right? And like, they all say that Navan still dislikes.

Wow. That's a, that's a funny story for God of SEO.

Yeah, exactly. I, I can also, people have always come up to me and goes, "Why did you call yourself God of SEO?" And I was like, "It wasn't." This is my new goal with the podcast. So, let me actually. So, so Jackie Chu came on, and, uh, and we were talking about Parasite SEO, and I brought up Lars Laughrin, and Jackie is like, "Oh, yeah, Lars has me blocked on, uh, on X." And I'm like, "I'm going to make an episode happen between you and Lars." And so, now actually me, uh, Lars Lofgrin, and Jackie Chu were supposed to do an episode together in January. Okay. And, uh, yeah, so I'm going to try to, I want to try to make it happen between you and and Rand on this show.

That's my goal.

I will be very, very impressed if you make that happen. I'm not going to lie.

I'll be impressed. I will be impressed with myself. Well, I mean, I know I'm, I am a huge fan. I just, I'll say here, I'm a huge fan of of R. I'm a fan of yours. I'm a huge fan of Rand Fishkin. Um, and I told him that too when he came on the show, and I'm always commenting on his LinkedIn post saying, like, "This is great." He just came out with a new tool last week, called Alert Mouse, and like, people, like, 10 or 15 people a day try to make, try to get me to make videos about, try to pay me to make videos about their marketing tools, and I just ignore all of them. But for Rand, I just like, he even asked me, and I just made a video for like my 700,000 plus marketing followers, and, um, because it was a cool tool. It was, it's like 10 times, it's like so much better than Google Alerts, and it's free. It's, it's called Alert Mouse, and, and I just made a video about it because I'm like, "Man, it's Rand Fishkin." So, I don't know. Maybe.

We needed a replacement. We need a replacement for Google Alerts. It was, uh, it was amazing to be fair that it came back, cuz I, so, I, I big up Rand the whole time, like, like I said before, I think he just took it the wrong way initially because I, I wasn't ranking for his name to be like a hater or anything, you know? I was just, I was doing it as a case because he was the most publicly known person in SEO, but I just think that he didn't, uh, appreciate my Black Hat endeavors to do so.

Crazy story. Well, Rand, if you're listening, I hope, uh, we can make this episode happen.

Yes. I, I, so, I was supposed to do a debate with Gary Ellis as well, actually. You know that.

Really? Yeah. Yeah. Yeah.

So, he's Google's head of sunshine and happiness. And then the conference that we were going to do it at closed because of COVID, so we never got to go back to the same conference. I saw him in China last month, or a month and a half ago. Um,

And he saw up for it. So, I maybe you could be the, the moderator for the, the Google versus Black Hat presentation, because that would be one hell of a debate when you get the actual Google engineer versus a Black Hat, you know.

Yeah, that would that would be crazy. Uh, all right. I want to move on to questions from Gagen Gotra, who I'm a fan of. He's a friend of the podcast. He's been on like three times. He, he literally gave me a list of questions to ask, and so these are going to be fun. All right. Um, site actually, and it's completely on topic of what we were just talking about with, uh, Black Hat versus Google. So,

Are you editing this quite a lot by the way?

Uh, I'm going to edit the stuff that we talked about.

Okay. Yeah. No, I, I wanted to ask about Lars as well quickly. Yeah.

We, Oh, uh.

We'll do that after I'm make. Yeah, I'm going to make a note. Remove Lars. We'll do that after I stop.

He's coming to my event next month. So, yeah.

Oh, he is. Okay, cool. Um, I mean, you could just ask me right now.

Yeah. No, I, I was going to ask why did him and Jackie fall? Like, cuz he likes me and.

Dude, he doesn't even know. He like actually asked him on on LinkedIn, and he's, he's like, "Oh, shoot. I, I like must have done that by accident or something, cuz I don't even know why I blocked him."

Yeah, cuz he's, he's like, he's not anti-SEO at all. Like, he, he's.

I know.

Yeah, he, he's anti.

He's a Black Hat.

Yeah, exactly. So, yeah. Yeah. So, I was very confused why they didn't like each other. I, because Jackie was going to come to my mastermind as well. Um, Oh.

Yeah, but it's, I, I host.

Like a private annual mastermind in London every year, and it's, um, in Knightsbridge. It's really, it's really cool. But, um, I just invite like a bunch of millionaires from me to come every year, and this year Lars is coming. He's in James Julie's group, so it's like,

"Oh, awesome."

Yeah. I, I think, I mean, Lars said he did it by accident.

"Okay, that makes sense then."

Yeah. Yeah. 'Cause I was going to say it didn't make sense why he would block him at all. So, I was he was going to,

"I mean, doing it by accident. I don't know. I, I mean, yeah, it is possible he didn't realize it was Jackie."

Jackie. Yeah, that's it's probably something like that.

"Um, but yeah, so we're supposed to do an episode in January."

"I'll look forward to that, too."

Yeah, that'll be good. All right, so from Gog and Go, we're talking about, uh, we're talking about Google's site reputation abuse policy. Um, so first question on this is, why did Google introduce this policy, and what is your take on it? I think they introduced the policy to have some level of PR to stand on, right? The issue that they were having is that it was blowing up to the point where, I don't know if you saw, um, the tweet from Jake Ward that went viral last year, and it had the CEO of like Twitch, who was the interim CEO of OpenAI, EMTT Sphere, and stuff. He was replying to it. Got 22 million views or something. Um, and that was essentially about using AI content to steal a competitor's rankings by putting it on parasite pages on your own websites, right? And just stealing their pages, but it was using their own content because he was rewriting their own content with AI.

"I think I saw this."

Yeah. So, that was a massive PR disaster for Google.

"And then secondarily, Outlook India and Forbes Advisor were massive PR disasters."

Forbes Advisor was one that did the rounds with investors and did the rounds with people who probably had, or people who were definitely shareholders of Alphabet, and may have been considerable shareholders in Alphabet. As soon as you get those voices, the ones that actually get the CEO sweating, that's when things happen, right? Um, I now, do I necessarily think it was an amazing update that had all these crazy things? No, because there was no algorithmic component, right? It's purely manual interventions. And anytime Google has to do manual interventions in its search is proof that it either a cannot control, or b doesn't have control of its algorithm, right? Because it can't fine-tune that out of its algorithm. The algorithm is promoting it. So, it's having to manually devalue it. That just means that all the other sites that weren't manually devalued have been given an opportunity on a silver and golden platter, right? So, it's, it's kind of twofold. One, yes, you're going to kill off a lot of the negativity, a lot of the PR, all that kind of stuff because you did this big bold move, and all the white hat influencers can reference that as this is not dead. It's obviously not, right? We see it all the time. Um, and they only did it in English. So, they only affected websites that were doing parasite SEO, that were doing certification views in English. So, every other language, Spanish, German, French, you know, whatever language, Thai, Mandarin, they were all completely unaffected. And if anything, a lot of those countries, especially countries like Brazil, um, are now ruled by authority sites and parasites, right? Because the algorithm promotes it. I think that again, it's proof that Google has lost, or doesn't have control of its algorithm. And it's a good thing for black hats. It's a terrible thing for white hats, right? Because it just reinforces that what we're doing is working inside the algorithm, and they have no way to algorithmically stop it. They can only go in and manually do it. And if they're having to PR announce that they're doing it in English, I'm just going to go do it in French. I'm just going to go do it in Thai, in German, in Spanish, or whatever, right? Um, and arguably the link systems and the NLP systems and everything as well, they're weaker in those languages as well. So, it means that I have to do less for content, less for links, less all this stuff. So, it's cheaper for me to do it there anyway. Um, so I'm a very big believer that Google just does this stuff for propaganda and PR reasons, and that long-term they aren't going to try to put anything algorithmically in it. They're not going to continue doing these banan waves of sites that are doing these things. They're just going to do this one-stop shop. Maybe a couple updates in the future, kind of thing. Um, but it's just a point that they can, they, it's a milestone they can point to to say that they cleared it up, they canceled it, they stopped it. They didn't. It's just pure propaganda. It's just pure PR at this point. And a lot of updates are, this is the thing, right? People need to stop reading in so much to the names of these updates, right? Helpful content update, spam reputation abuse update, site reputation abuse update, all these kind of things, right? They have very little meaning from the title versus what they actually do. And helpful content update was probably the best example because we got, because of the DOJ filings, we got internal Google communications showing that Danny Sullivan actually didn't want to call it the helpful content update because he knew that it had nothing to do with content. So, he was trying them to trying to get them to change it to something else. One of the engineers wanted to call it the SEO first update. We know that from DOJ filings. Thank God we didn't get blamed, you know, for helpful content, right? But this is the point. A lot of people need to stop looking at the surface and start looking at the subs and start looking at what's changing and what's moving, and you'll very, very quickly see what Google is rewarding, what it's not rewarding.

"That's good advice. Start looking at the SERs, looking at the behavior, looking at what's changing."

"And Forbes and similar sites, so they had manual actions, but they're now back."

Yeah. So Forbes Advisor has a specific interesting one because they got hit with a manual action twice, right? And then now they're fully back anyway because they changed their procedures. So they changed their editing guidelines and their writing team. They fired half the third-party writers and things like that. But the funny thing is the authors of those articles from what we can tell are pretty much all the exact same. It's still the exact same content. So if it was site reputation abuse, that they're abusing their site reputation to rank this content, why is it now no longer site reputation abuse just because they changed the editor? Like,

"So what do you think? What do you, what do you think the future of this policy is?"

I like, I think Lars did a fantastic blog post. Lars Offen did a fantastic, and it doesn't actually make sense that you can't have third-party writers for your content. Have like, what happened to columnists for newspapers, right? Like all of these people that were third-party writers as a part of the actual content is now, that's now not allowed just because Google doesn't want you to put affiliate links in from a third-party writer. Like, it doesn't make sense on the tin. So I don't think that they're going to continue pursuing it, or they don't think they're going to continue act, I don't think they're going to, there's not going to be like a site reputation 2.0 update, you know what I mean? I don't see that coming because the PR wasn't that great from the first one. The results weren't that great from the first one. And the algorithm, if anything, is just continuing to reward authority and continuing to reward links and continue to reward spammers, if anything, to be very frank.

"And so, do you think this is going to, this is going to work with AI? This is going to continue to work with AI. You want to get, you want to get more visibility within AI. You use tactics like this?"

Probably. It's all, it's all about where the AI is sourcing and grounding the search rate. So, if it's in Bing, if it's in Google, the results are not that much different anymore. Um, and a lot of the pages ranking is still the same pages ranking. It's just, and again, it's because of authority, right? Like, it's just coming back down to if you can get pages on authority and trusted sources that the AI trusts and references, you are going, becoming more reliant on current information than it's old train. It, like, I, I, I think it doesn't again make that much sense to be reliant on information from a year ago, right? For like training data versus grounded search information. I just feel like grounded search information is a lot better because you are, because it can be current. It can have edits from yesterday, whereas the training data can never have that. It can only, it's, it can only ever be up to its cut-off point, right?

"Yeah. Yeah. You said something pretty interesting that I've heard you say before, which is that, uh, non-English SEO is so much easier, which I also agree with, like, a thousand percent. It's so much easier. But can you, uh, can you share more about about why? I think you even said in another interview that most of your sites are not in English?"

Yeah. So, most of my personal sites, yeah, because the algorithm changes less, and it's a lot less, uh, it's a lot, got a lot less anti-spam and a lot of NLP signals and things like that, right? So, as an example, um, just in English, the algorithm has been allowed to develop to a point now where it's got testing data, it's got reinforcement, it's got all of these things. They've been able to show that, okay, if we change this signal, it affects English language results to this degree. They know that, right? But they don't have nearly as much insight for other languages, right? Um, and to some extent, they don't want to because those other markets aren't worth the same as what English is. If you look at the revenue from the US for Google Ads, it's like 50% of its entire revenue is just the US, right, for Google Ads. So why would it try and focus on another market that is 0.2% of its market share? That doesn't make any sense economically or ROI based. So, it's 100% just going to focus on the markets that are the, that a, it's going to cause the most political and legal headaches for them. Um, which, to be fair, some of the e-markets, it has been making moves in German and in French because they do regulate there. Um, but it, like, do I see those same anti-spam systems in Thailand or in Vietnam? Hell no. You know, like, I see hackers and malware ranking every day in these searches, literally. Um, and Google, at the end of the day, Google doesn't care. And if it has to, especially for parasite, if it has to be reliant on manual interventions, I think the, um, US has something like 300 search engineers, right? Working on the algorithm, working on search, things like that. In Japan, Google has one, right? Like,

"Wow, really?"

One. Yeah. One. How are you going to,

"Wow."

get anything done, right? So, I really don't see,

"them going into those other markets, so I just, for the next few years."

AI overviews, AI mode, that kind of stuff. It does affect it. It does change things, obviously, because you're not going to be able to get the same affiliate clicks. You're not going to be able to get the same CPA sign-up in your form, that kind of stuff. Your conversions are going to be dropped. But black hats, theoretically, you should be able to control AI overviews even more so than you can control the number one position because the number one position has volatility, whereas the AI overviews are just a consensus of sources. So, if you can put more sources than everybody else in the SER, you can build consensus in the AO view. Even if number one says something different.

"But, okay, so how, but how do you stop? Uh, I guess, I guess it's just the amount of links. I was going to ask, how do you stop affiliate links from being stripped?"

So, so you can't. That's what I mean. So, you just have to deal with lower clicks, lower conversions, lower everything. And just like, more, more visibility, lower clicks. Basically, that's what. And unfortunately, this isn't just SEO. This is social marketing, video. Like, you're going to have to get more visibility for lower clicks across every platform, right? Like, it's just what every platform is moving towards. Google wants to try and have lower clicks, Facebook wants to have lower clicks, Twitter wants to have lower clicks, YouTube wants to have lower clicks. Every platform wants to try and keep you on that platform. And so you're fighting.

"Credit to Twitter though, because I think they just stopped, um, suppressing posts that have links in them."

Yes.

"I don't know if you saw."

Yeah. So, because even Elon Musk's advice was to put your link in the reply. Did you see that?

"Yeah, I did see that."

And I was like, look, if the owner of the company is telling you to do it, you should. Then they know it's a problem, but you should probably be doing it, right? Um, but no, yeah, that is, it's good that they have done that because otherwise, I think Twitter is also very much so like the summary information source site. Like, I don't go on there and apart from like the occasional thread or the occasional long-form post, I'm not sat there reading a thousand words, you know, I'm sat there reading a hundred or 50 or something like that to summarize the information, summarize breaking news, things like that. Um, even SEO, I'm not reading like huge chunks of information. So, having the ability to click off to a website and actually go through something that's formatted in a correct space and things for long-form content makes sense.

"Yeah. Also, like, uh, from a product perspective, it used to, before, before that was in it, like it used to be, you would go on on Twitter and discover cool stuff."

Yep.

"And and now it's kind of, I think it's like it could kind of get back there. So I think that was, uh, actually really welcome that. I think that was cool."

They need to fix the whole new timeline, but that's about it. There. I, as far as I know, they're working really hard on it.

"Good. Cuz it, dude, at some point it was bad. Like it was bad."

Oh yeah. Oh yeah. Oh yeah. Yeah. Yeah. LinkedIn. LinkedIn though, uh, posts with links are still suppressed, and you have to drop your link in the comments.

"Yeah. Yeah."

Um, uh, actually, speaking of X, who are your favorite follows on X? Who, who, uh, yeah, who drops like crazy value? Who should everybody follow?

I, I like, um, Djon AI SEO. DJ SEO.

"Yeah, he's he's a legend."

Yeah. So, he, a lot of people don't know him, and he dropped just really interesting AI. Dan Petrovic.

"Yep, exactly. Dan, Dan Petrovic."

Um, he's really, really good.

"Is that my guy? Say his last name?"

Oh, shoot.

"And then, uh, trying to think of other people."

Cyrus is always good on there, but he's trolling half the time, to be fair.

"Um, Gagan is actually a really good follow."

Gagotra. Um, and there's a, there's a couple other ones, but Twitter's fallen off unfortunately. My X has fallen off, but Harriet's pretty.

"Harriet. I, I mean, I don't know if you like her. I love Lily Ray. I'm always sharing Lily Ray stuff."

"Yeah, she blocked me, so I can't share her stuff, so."

"Oh, no."

Another, another, uh, threeperson podcast. I need to meet Catherine.

"Oh, and Barry Schwartz."

"Yeah. So, yeah, Barry Schwartz." Well, he's just new. The problem if you follow him is he takes up your entire feed. That's the, 'cause he posts every article, so it just comes up all the time with various stuff. So, I, half the time I mute him and then I unmute him 'cause I see something really good, and then I mute him again 'cause it's just taking up my whole timeline again. So, I'm like, damn it.

"He's Barry, Barry's actually, um, we're going to do an episode. It's going to be me, Barry, and David in December where we're going to predict the future of SEO in 2026."

God damn. Okay.

"And I think, yeah, I think that's going to be a really fun one."

"Um, all right. Yo. Okay. So you shared, you shared a post recently about Black Friday, Cyber Monday."

And because it's coming up, and and you said Google is desperate for fresh, time-sensitive content on trending queries. So, what should SEOs do right now for, I mean, it's kind of late, but anyway, they should have done it probably like a few weeks ago or longer for for Black Friday, Cyber Monday. So, any affiliate CPA offer that you have already, and especially if you have exclusive or you have higher commission rates than other affiliates, immediately just go and research and see if there's anything, any volume at all. And bear in mind that it's seasonal volume. So, when you're in HFS, if it says 300, look at the monthly chart, and if the top goes bam in November and it goes to 50,000, that's what you want to be jumping on, right? Um, ideally, we're just trying to look for things that we already have, especially because we've only got a few weeks. We're trying to look for things that we already have. We can already tick off and we can already just immediately create. If you're, if you don't have anything and you still want to try and take some advantage of it, um, I would go in, especially in other languages, again, trying to not be English here because English is quite taken up by newspapers and things like that and stuff these days. Um, go into other markets, create your main post that you're wanting to try and rank, and then just rewrite it into AI and distribute it over parasite free parasite pages. So, LinkedIn Pulse, Medium, all that kind of stuff. Aim for big but not too competitive queries. Um, and try and make sure that each AI rewrite has some longtail variant in that the last one wasn't necessarily trying to target. Ideally, what happens then is you build 10, 20, 30, maybe even hundreds of pages if you have the time to go and create it. And bear in mind, people who have time, but not as much capital, have an advantage here because people who have capital but don't have time currently, we're in a time-limited environment where you only have a few weeks. So, actually, you have an advantage if you have time. Um, go and create as many as you can because the more visibility that you get, the more likelihood you are to get clicks on those things. And bear in mind that it's only worth it if you're getting clicks that are worth it. So, if I'm like, let's say I'm trying to promote like adult coloring books for Black Friday, that's probably going to get me a 30 cent commission. It's probably not going to be on Amazon, right? That's probably not going to be worth my time to go and create all of those pages for. But ride-along lawnmowers, right? Even on Amazon, you might be talking a $400, $500 commission, right? So, something like that might be worth my time to go and do. And there might not be very many competitors. So, just putting myself,

"So, what, what keywords would you target?"

See, that's where I, off the top of my head, I can't give anything that isn't, I don't already target. Um, maybe like seasonal holiday things, anything that comes and goes. So, um, over the last few years, there there's been like, you know, hoverboards and fidget spinners and things like that, because Black Friday is always in the lead-up to Christmas. A lot of people are buying stuff in Black Friday for Christmas. They're buying it for Christmas presents and for things like that. So, you can do a lot of stuff like that where people don't even think. Like, I'm, I'm sure that when fidget spinners were happening, fidget spinner at Black Friday, or Black Friday fidget spinner has probably got hundreds of thousands of searches in the course of a few weeks. Um, that kind of opportunity is rare. And if you can kind of double up where you get the seasonality plus you get the limited time trend, it kind of compiles and goes to the moon. I'm sure this year it's going to be Laboo, Black Friday. Is probably going to be on there with some stupidly, uh, stupid search volume following me going up there as well.

"Dude, I just bought, I actually, I'm using a fidget spinner right now. And, uh, yeah. Yeah. And I, and I, so I just bought two from Amazon, but I was trying to find them in person because I didn't want to have to get them online and I couldn't find, like, they're so, they're so like, not trending now that I couldn't find them anywhere in any stores. And so I had to buy them on Amazon. They just make doing like long pod, because I do so many long podcasts each week, and I'm just like, I need, I need fidget spinners. I'm like, drinking coffee and I'm like, yeah, I couldn't find them in person. So, uh, that's funny."

"Um, uh, okay, here's a question. What are two to three trust signals that you look for when you're building links?"

So, entity is number one, right? So, a lot of people don't actually go and look at the, the, uh, brand SER for this site you're building links on. Um, I'll go and look through and see in the SER, three little dots, more about this page, and see if Google has a description for the company. See if there's more information about this company in there. If there, if there isn't, a lot of the time, and you're buying links from marketplaces, those are domains or sites that have been bought and then rebuilt to sell links off of. So, if they have no, if it says there's no description available, or Google couldn't find information about this company, a lot of the time that's because they've rebuilt an old company into, um, a new one, and the previous information says it's closed, or the citations were deleted, things like that. When it comes to the actual building links, I due diligence on metrics and traffic and, uh, where the guest post or the content is placed. So, metrics, super easy to manipulate. You can boost DR, you can boost DA, you can boost that kind of stuff for like $15. It's ridiculously easy. You can even get an, an, a redirect domain. Buy an expired domain that's DR25, redirect it to your site, now you're DR18, stuff like that for $12, right? Um, traffic, you can manipulate with fake keywords. So, as an example, if I created a page on my DR30 blog and I put it as like, Disney Plus login page on my blog, right? So, I created this page, which is Disney Plus login page, and then that page ranks number six, or no, number page six for Disney Plus login. Well, Disney Plus login has 18 million searches a month. So, the fact that I'm page six means my site has 2,000 visits a month, right? My page has 2,000 visits a month for my Disney Plus login page. That's faking traffic. That's faking traffic for to sell links. Um, you can fake traffic for other keywords like software error codes, um, things like, uh, Windows variations, all sorts of things and stuff that you can, um, basically have high search volume and low page count in terms of actually what's ranking there. Um, those are really easy. And then, uh, the final one is special categories. So, a lot of sites or a lot of people who are selling links will only put guest posts in this one category. They put, they'll put like slash, they put like breaking news, or, um, special news, or some [ __ ] like that, right? And they'll only put guest posts in that category. I don't want that either because that's creating footprints for Google.

"All right. So, we've been talk, you've been, we've been talking about some more advanced things. So, I want to ask, what are some parasite SEO platforms that even like an idiot is not going to mess up on?"

Yeah. Uh, LinkedIn Pulse, easy. Medium.com, easy, right? Um, Reddit, by far, number one, easiest. Uh, and a lot of the time as well, if you're new to a platform and you're trying to use it for parasite SEO, go into the AI, especially like Grok, because Grok lets you get away with more blackout stuff. Um, go into AI and just ask it to give you information and recommendations on how to use this platform and how to not get banned, right? Um, so many people that I've seen, I tried Reddit, but I got banned from every subreddit that I was trying to comment on and things like that, right? Um, go and ask the AI for the rules first. Go and figure out how to use the platform, then actually try and implement things, right? Because otherwise, you're just wasting your time. There's a lot of people who waste their time.

"And, uh, yeah, they don't, they don't read the rules. I think something, uh, a lot of people just underestimate, um, you don't do this, you know what's up, but a lot of people just underestimate, uh, spam algorithms. It's that and how easy it is to detect irregular behavior."

Yep.

"It's not that hard actually to have a profile look legit."

Yep.

"But people don't put in that like 20% of extra work to make a profile look legit."

It's like they create a Reddit account and then right away they go into their favorite subreddit and start spamming their their company and,

"Exactly. Yeah. And they're trying links and stuff and all sorts. It's like, what?"

"Yeah. They're spamming links and all that stuff."

Yep. Zero karma account created yesterday.

"Yeah. Zero karma. Yeah. It's not, it's like, uh, and it's like that's not natural. That's not normal. Of course. Like, of course."

"Um, what do you, this is a kind of similar question. What do you think the biggest mistake people make, uh, when they're trying to do black hat search engine optimization?"

The number one mistake that you've seen.

"Uh, picking the wrong niches, right?"

So, so many people are like, "Oh, I'm going to go do weight loss. I'm going to go do pay the loans. I'm going to go do cryptocurrency. I'm going to go do casino." And I'm like, "Yeah, but the best black hats in the world, the ones that are ranking for those terms, right?" And I, if you're just starting out, you're not going to be anywhere near their level. And you won't even be able to reverse engineer half the stuff they're doing, literally. Um,

"So, I would always say go and test things in easier niches and then scale up what's working for you, right?"

Don't go and start throwing, and especially don't go start throwing money at problems. Too many people are like, "Okay, I'm gonna build this weight loss site. It needs $20,000 to do it. $25,000. Put this in." Then when it goes nowhere, well, you've spent your budget. You don't have that much time. You don't have that much experience or history, and you don't know where to go from here. And any actions that you do go from here could actually make it look more unnatural than what you're doing. Anyway, so,

"Let me ask, let me ask you this. Do you think every SEO or every webmaster should even be doing black hat?"

No. Not at all. No, no. Because there's some, like, as an example, like, we, I was speaking to, um, this earlier, right? Law firms are probably a terrible one to do black hat for because they will sue you. But, as, there's there's lots of other examples as well where sometimes it's just easier to not do it. But, w, with with that being said, actually, um, as we move more to AI overviews, more to AI mode, I don't see a world in which we can't, if that is the default, right? So, if, if, if AI overview/ AI mode,

"Do you think it will become the default?"

If Google figures out how to monetize it in the same way they can monetize the current ad system, yes. Um, if they don't, then no, right? But at the moment, it's looking like an inevitability. And that's what they're saying in earnings calls and things like that. That's what they're trying to move towards. Um, you, with one website, if it's a consensus of 25 pages, you're, what you can do all the SEO you want on your one website. It doesn't matter because you need a consensus of sources. So, you need those 25 websites to say something at least similar to what you're saying for the AI overview to say what you want it to say, right? Um, and so the only way that you can do that is by going and buying and building pages of the same content, the same consensus on other websites.

"Across domains, right?"

And that's black hat. Like, there's, especially if you're trying to create for the same piece of content over and over and over and over again, right? That's spamming. That's,

"Yeah, that's that's manipulation for sure."

I'm not, I'm not so positive that, um, but I mean, lots of us, like, to your point, you know, there, uh, LLMs are crawling search, or are going through search engines, and a lot of the times that's what's ranking on page one will show up and get visibility within LLM.

"Exactly. Yeah. 100%."

I just, the, again, the only problem is the LLMs, still, it's still a consensus of sources, right? So, it's never normally just that one page that's being, that's been taken from from there.

"I mean, if you're only doing, if you're only doing, if you're like a, if you're a startup, and you are only doing SEO, and then again, you could still, like, for some, uh, for some niche queries, you could still show up."

"Yeah. Yeah. For, for niche queries, 100%."

Right. But I'm talking,

"If we're talking about the most competitive stuff."

Not even most competitive. I'm talking like, even if you're just like some random city in Idaho or something, right? Like, best lawyer in some rank there. There's probably going to be like multiple law firms, multiple press releases, blog posts, newspaper coverage, like all this kind of stuff, right? So, you're going to have to create multiple sources to outrank that. And I don't see any way that that's not black hat, right? Like, because it's, you're a, you're doing multiple pages of the same content. B, you're doing it across pages that you inevitably have to pay someone for. Like, one of those websites is going to want payment for you to post on the website. Um, which is black hat. And third, you're again creating the same content. So, it's spam because you're doing the same over and over, which is spam. That's the definition. Do the same thing over and over again, right?

"Right. That's spam."

"Yeah."

Know that I don't see in any way of the three ways it's not black hat because they're all black hat, right? So, I literally, I see the future of AI SEO, AI modes, whatever. Again, I don't necessarily think, like I said, I don't, I don't see them rolling it out now or soon because they haven't got the monetization there. If, if they can get the monetization as part, and it's the same monetization level that they have got on ads right now, then they would do it tomorrow.

"You think so?"

100% because there's already, um, research that actually the AI overviews are cheaper to run than the search algorithm. So, it makes, it's more economical sense for them to run the, the AI overviews on top than it is. It's cheaper for them to run the AI overviews than it is for them to run the search algorithm. So, 100% I think they, they would do that. Yeah.

"Interesting. That's a question that I should ask some more guests. Is, uh, do you think, do you think, um, 10 blue links will be entirely replaced with AI mode?"

I, I don't. Entirely is also like, I, I think they'll probably still be there. This will be available, right? Um,

"But that's what, that's what you're saying that like my mom will go to Google and then it will just be a chatbot."

It won't just be. It will be the default. So, what will come up above? But I still think there'll be still stuff underneath, right? Like, yeah. So, but the default, the first thing that she will see when she's searching Google box will be the AI mode. Like, that's. And right now, for 20% of queries, that's already happening.

"So,"

"Yeah. I mean, you can, you can, you can follow up with AI overviews."

"Yep. 100%. Yep."

"Yeah. So, it's not."

"I don't think that they should have even rolled out to 20% now because there's no fact-checking, there's no verifiability, there's nothing. Like, I've seen multiple times where people have Googled like cancer treatments, and it's told the number one treatment is to go jump off a bridge."

Seriously? Like, stuff like that.

"Oh, yeah. I mean, this is, this is very widely, widely shared by SEOs is just how bad the results are. But also, it's malware and hacking and stuff. Like, I'm seeing, um, like an uptick from what I can tell anyway, in scammers using this. So, you know, scammers who are phoning people's grandmas and stuff, they're actively trying, they're absolutely manipulating AI overviews to tell their customers something different. So, when they, so, when they're, when they're on their scam phone, they say, 'Oh, go Google, you know, Forex Trader LTD.' It comes up, these are the number one most trusted thing in the world, you know, that kind of stuff, right? And that's because it's so easily manipulated. So, this is where it's going beyond just the internet. Like, it's,"

"When they are, when they're manipulating, when they're manipulating it, and they're doing that by getting covered in lots of different websites."

"What exactly are they doing to get these, these exact mentions syndicated across websites?"

The exact same thing we talked about before, press releases, blog posts, that kind of stuff. A lot of the time, it's more so press releases and things like that because they don't want to trail. They don't want to be able to say this was by published by XY and Z LTD or LLC or something like that, right? They just want to be able to have it, this is the cut off. Like, you go, you can contact the owner of the website. They can send you a PayPal email, which is for some dummy email from a random country or something, right? Um, that's where it's becoming, you know, and, and bear in mind that inherently, because it's Google, it's inside Google's site, there's some level of trust that the user has with Google automatically, tell, because it's Google telling them the information. It's not some random magazine or some random blog or something. And that's where I think it's even more dangerous for that side of the coin. Um, and I do think that there, there's, again, this is like getting really deep, but there's, uh, a lot of research into LLM induced psychosis. I don't know if you've read into this.

"No, actually I haven't."

So, people can go insane talking to LLMs. I,

"Oh, dude. I mean, I feel like I'm going insane, but they can no longer differentiate between reality and like real conversations."

I feel like I feel like that with AI overviews is going to go to the moon because it's, you can imagine already that AI overviews, if you have some very niche conspiracy theory, most of the information is going to be on that conspiracy theory. It's going to be available there. The AI overview is summarizing what's already there and it's reinforcing that. It could be completely incorrect. It could be complete nonsense, right? But the AI overview is telling them this is real. This is, this isn't in your head. Like, and again, that's so dangerous. Like, it really, really is.

"Especially for older people."

100%. Yeah. Oh, my, my grandparents are terrible already. Like, they believe half the AI videos they're seeing on Twitter and stuff already.

"Oh, yeah, dude. And I mean, even, even the videos where people use those like remove Sora watermarks."

And, and, and they, those tools do such a terrible job.

"Old people still fall for it."

It's insane. It really is. Like, yeah. But again,

"And then you think about, man, and then you think about doing, doing something like that and combining it with parasite SEO on YouTube."

That's bad stuff.

"This is what I'm saying. Like, this, this is why these scams are getting so prevalent. It's because these, like, these platforms are allowing them to be reinforced. They're allowing, like, before, if somebody phoned you and they were saying all this stuff, you could actually go and verify for yourself, right? But now, when you do go and verify for yourself, it's reinforcing it for them. Like, it's reinforcing the scammers."

Thoughts to yourself. So, I really do think that there needs to be some safeguarding going on over the next few years. And I, I, I hope they don't roll out default because then,

"The information wars begin."

I think I really like that. And also, it's this is more power and control in Google's hands. Like, we do realize this, right? Like, a lot of people don't realize that we are allowing Google to be the arbiter of information and truth and like all these things by doing this, right? Like, at the end of the day, they could be like, "Okay, let's tweak the algorithm for this specific area of information, this specific politics, and only surface X, Y, and Z in the search results, which then means the AI overviews are only going to consensus this information." Sure. Right. I really do, like, I'm putting my conspiracy hat on here, but I really think we're giving Google way too much control and power over information and c, and it's, it's centralizing information basically.

"Well, that's why that's why I don't think, uh, that's why I don't think it's going to become the default anytime soon because the results are, look, I mean, you're saying the traditional search isn't great. A lot of people are saying that AI results are even worse."

Yep.

"And, uh, and, and I do believe that Google wants to give the best experience for users, um, and the best information to users. I mean, sure, they mess up. Every company messes up, but I mean, like, they're literally dealing with the world's information. So, like, it's not like an easy job or or anything. So, I, I, the reason I am not, um, the reason I'm bearish on like AI mode just becoming the default is because I don't think the results are are good enough, especially like you said, for for, uh, medical queries where people are like, 'Yeah, jump off a bridge.' Or where AI is like, 'Jump off a bridge.' And, um, but, uh, I mean, even if it does, I mean, yeah, if it does, like, what are, what, what do you do for SEO? I mean, still a lot of a lot of things like, if we're talking, if we're talking white hat, a lot of things that would probably work for normal SEO, you would still want to do. You would want to do social listening. You would want to do, you would want to target the right queries and not waste your time on queries that aren't going to get you customers. Um, you want to, you want to target those queries. Exactly. You want to have like, there's a lot of things."

"But my main concern would be for people's, you know, for the industry as a whole, right? So, people's currently the structure of most of the industry is people are, you know, you're an analyst, you're a research, you're an outreach manager, you're a technical SEO, or something like that, right? Tomorrow, 90% of those jobs, if AI default does come into place, just don't exist, or or they are significantly devalued. That's the thing, right? Like, your job is no, a technical SEO job. If AI default, if AI defaults tomorrow, and you're doing technical SEO for one website, suddenly that's a $100,000 career to a $20,000 career, you know, literally because you're devalued overnight so much that it's, you're only worth one of 12. Previously, you were one of one, right? So, I, I, like, I'm bearish too, in the fact that it's, I'm just down to like, I, I don't believe that Google cares that much about its quality of results, to be fair. Like, I, because I've seen firsthand that they choose ad revenue over quality every day of the week. Um, and they want to be, they, they are trying to actively grow their profits and revenues from ads to fund their AI growth as well, basically. Um, so I'm pretty bearish just on the monetization aspect that I don't think Google can make as much money out of AI mode, AI views as they can out of ads. And until they can, they're not going to do that. And so far, we haven't seen any information that they can do that or that they are close to be able to do that. Um, and we've also seen that actually searches using AI mode are less valuable than searches, traditional searches."

"Are those ads then less valuable than traditional search?"

"Why, why are they less valuable?"

There's a massive study released the other day. I think it was like two or three days ago. Cyrus Shepard shared it on on Twitter. Um, and it, they did a whole study just showing user journeys basically and going through the AI overviews, Google AI versus traditional searches was 19 to 22% less valuable in terms of end spend than it was going through.

"But, but like everyone is saying that LLMs have higher conversion rates."

Again, it doesn't matter. Like, from, from that's also based on anecdotes from what we can tell, right? Whereas this is based on actual study of tens of millions of user journeys. Um, and from what we can tell, it changes the user journey quite significantly. Um, and it kind of, it stops as much comparison. Um, so they're actually not looking individually as much as they were previously, as in, you're not opening tabs, you know, to go through the products. Um, and B, it's again, centralizing information more. So, before, you might have had 10, 20, 30 different brands coming up. Now, it's two, three in the AI overview. So, it's creating a lot less, um, opportunity, or not less opportunity, a lot less options than previously, which is probably as a result, creating less spend.

"You know, this, it, it's interesting with these studies because there's so many conflicting studies and I'm sure, I mean, you've been in SEO for such a long time. I'm sure you've seen this like a trillion."

This, this is the biggest study though, from what I can tell so far. It was,

"I, it's the one that I believe, so, so far, the most out of all of them."

Because it also does make sense in the sense that if the user has less information, but it's more consolidated, then they actually do, then they have less information and less options and less opportunities to go and spend more money, or to upgrade things, or to buy add-ons, or to buy upsells, or to do all these things. Whereas previously, they had all that additional information. And also, um, the study took into account that you no longer get retargeting clicks. So, this is one thing, right? Previously, if somebody opened 20 tabs, all of those 20 tabs would be able to get, uh, pixel data and retarget them. Now, you don't get that. So, you can't retarget them with ads anymore. So, you are net losing that ability to retarget and gain customers again. Um, so there was multiple reasons on why it was, and but it did look to me that makes sense. Yeah.

"That makes sense."

"That makes a lot of sense. All right, this is the question that I'm going to end on. Uh, I, I like this tweet that you put out. You said, 'I've built, consulted, and audited well over a thousand different e-commerce and SAS sites.'"

And you listed, uh, you listed the worst mistakes that you see. Um, but I, I'd just like you, off the top of your mind, what are the worst, like, people who have websites, SEOs, what should you stop doing, or like, put some thought into right now?

Two things that come straight off the top of my head are meta titles and internal linking. Um, for some reason.

People don't optimize their metatitles properly at all. And as a prime example, um, the other day I added "service" to one of our metatitles on Press, and we ranked number one the next day for that query, just by adding "service" to the metatitle.

Right. Um, and >> so did the query had "service" in? >> Yeah. So the main query, so it was it was niche edits page, basically, and uh, then it was "niche edits best niche edit service." We were ranking top four, like "best niche edits," top three or whatever for "niche edits," things like that. But we weren't page one for this "niche edit service."

>> So you had "best niche edits" and then you just added "service" to it and it was it was like exact, you exactly match the keyword. >> Yeah. So I >> it was in the page title. >> And again, I try to naturally optimize metatitles. So I'll be like, uh, as an example, with the niche edits one, I think I put "niche edits - the best curated link insertion service." Right? Something like that. Right? But again, that's very clean, >> white hat, almost. Um, but it includes four or five different variants and modifiers inside the metatitle, which most people don't take that additional time and research to do.

>> That's true. >> Um, with internal linking, a lot of people are either so internal linking that it's just a mess, to be very frank. There's no system and there's no uh pre-thought and predetermination into where links are going.

So >> yeah, >> generally speaking, when I go and audit a lot of sites, especially e-commerce sites, they have two or three pages that are doing a lot better than every other page on the website. There'll be like a blog post or some guide or some news release or something, even they like acquired a business or something like that, right? There'll be something that they've that they that just results in this page having a lot higher number of backlinks than every other page or every other page like it on the site.

Backlinks or internal links? >> Backlinks, right? RD. Yeah. Um >> Uhhuh. >> And most people don't go and take those into account, right? So the amount of times that I've gone and be like, "Oh, you have these two amazing pages, 29 RD, 54 RD, you got Wikipedia, the BBC linking to these pages, amazing links, right?" You go on those pages, zero internal links from those pages to other pages on the website. But yeah, those are your best >> pages on your website by link. So the fact that you're not putting in links in there means you're not moving equity from that page to other pages.

>> You're not passing. Yes. >> So I'm all all I try and get people to do at the bare minimum with internal linking is go and look at your best pages and see if you could add or change internal links on those pages. If not that, then the best thing you could really do is try and do a system where the most competitive anchor text is on the most high high authority relevant page, right? So and it goes down. So you have the exact match, the highest RD, highest uh UR, whatever page here, this link, the lowest competition, you know, "buy niche edit service UK 100 guarantee links," whatever that link that's on the least authoritative page, you know, a blog post with 400 words, no links, no internals. That's where we're going to put the longtail, super longtail variant that has 10 monthly search volume, right? And we're just going to go down from the highest authority to the most competitive and kind of go down that list. And a lot of the time we're doing it um money page by money page so that we make sure that our team has got the most amount of time per internal link to actually get the equity moving proper properly.

Um, and with internal linking, a lot of people use VAs that don't have full English language training, um, or don't speak fluent English. And that's also a terrible idea when you're using internal linking things because they'll go in and edit paragraphs or edit descriptions and and it will just look like a mess and it won't be readable and it will devalue links and stuff like that as well. So, >> or or they'll use AI, but they won't even understand the AI. >> Exactly. Yeah. And it and it's I've actually had a VA use before and it's [ __ ] mentioned a competitor in the edit and I'm like, "Are you insane? Like, you're putting a competitor's brand name on our website."

>> I think but they just don't know. They don't know. >> Exactly. That's the problem. Like Yeah. But again, but that looks so bad to a client if you're doing that for client work. That just looks horrific, you know? Like Yeah.

>> Yeah. I'm I'm very I've been scaling my team and uh I'm like really really picky when it comes to hiring people and I yeah, I mean I've been seeing some just to me obviously dumb stuff. I'm like, "What are you how did you make this mistake?"

>> How did you make this mistake? >> Yep. >> I don't >> it doesn't make sense.

>> Charles Float, Charles Float, thank you again for coming on the you are such a legend, dude. I am like uh I I was I was telling a friend of mine this yesterday. Basically, when I decided I wanted to start getting guests on this podcast, it was uh it was like a year ago, actually. And I up till that point, this is so this is episode 854 of the show. I've done the show for 854 days in a row now.

>> Without missing a single day, >> and and when I when I decided a year ago that I wanted to start getting guests on the show, you were one of the first people who came to mind, but I'm like, "Oh, Charles Float is so big. I can't ask Charles Float yet." Charles Charles is Charles is huge. And uh it was actually it was you and Lily Ray who were like people that I just read on X all the time, which is very funny because I guess you guys kind of have like contrasting SEO viewpoints.

>> Um >> it's probably the best thing to read though, really, is you know, directly the opposite of one another >> and then go with something somewhere in the middle, probably. >> And you guys both have very strong takes and very strong language, which I like and which also creates good viral content.

>> Yes. >> And uh Yeah. And I'm like, "Yeah, I want to get Charles Float in the show." And here you are in the show. And it was I hope you'll come back.

>> Um if you ever like like we were talking about uh before, if you ever want to do like just a 10 or 15 minute episode, something that's on the top of your mind. U maybe you built something because you saw an opportunity in SEO that wasn't being addressed. I built this. You could even come on just share share stuff. I don't care. >> Sure. Sounds good.

>> Um >> yeah. So, uh, thank you again and, um, yeah, I guess, uh, stick around when I hit stop because your video will have to upload.

>> Yep. >> But, uh, for everybody watching and listening, thank you so much for tuning in episode 854 and I will talk to you again tomorrow. Bye now.