Episode Transcript
[00:00:00] Speaker A: Hello, I'm Jeremy Rivera, your unscripted SEO podcast host. I'm here with Alejandro Meyerhans. Why don't you give your intro, tell us a little bit about your country, your company, a little bit about your experience, and focus on why we should trust you as an expert as we delve into SEO and your niche, which is link building.
[00:00:21] Speaker B: Yes, sir. Well, thank you very much for opening your platform to me. Sincerely appreciate it. My name is Alejandro. I come originally from Spain, although I don't live there. And I've been doing SEO since 2016. I started with small affiliate websites just trying to get out of Spain. I didn't want to be a waiter anymore. I was a bit miserable. Right. And I heard that you could make money online, so that sounded better than, you know, serving tables. Started trying the whole thing. Eventually started clicking. I got a job for a digital marketing agency that was my first mentor, Dominique Wells from On Folio, then now in the nasdaq. But those were earlier days before for that. And from there, I started doing SEO because I.
It changed my life already. So, like, oh, let me double down on this thing. I don't want to go back. And it turns out that it played to all my subjects of special interest. Math, statistics, game theory, language, all the things that I really liked. If you deploy them properly, that's how you get good SEO. Dom.
[00:01:16] Speaker A: Right?
[00:01:18] Speaker B: Knowledge.
And so I started just going as deep as I humanly could on it. I started building a whole portfolio of sites for DOM that eventually became Monfolio. He took it to the level where it is now, of course. Right. But I was with him for the first couple years. Then I started doing my own agency thing, because at that point, right, because I was running those sites, I got pretty good at a few of the basics, right. I knew how to make a really fast website. I knew how to publish good convert and affiliate content, and I knew how to rank through content and backlinks because, I mean, that's always been, let's say, 90% of the game right now. There's a few other things that are making a bit more nuanced, of course. Right.
You know, 2016, 2020, it was just content marketings, right. Everything else was still in the infancy. Right. Branded signals, ux, all that stuff mattered a lot less than it does now, right?
From there, a lot of people started asking me, what would you make my website fast? Would you audit my search console for me? Because I don't know what's wrong with my websites. And I started saying yes. And just Going deeply through it. I started getting pretty good at doing audits from there.
I just became a forensic SEO auditor for a couple of years. People will come at me, Alice, I've lost half a million viewers a month. Where did they go? I'm like, oh, let me find it. I like doing this, right? Let me find. Let me find where they went. And it's just going through the data. Data, data server, log analysis, analytics, GSE, everything, right? And I nerded out with it until 2022 and 2022, the new buyers of GetMe links. For the record, I'm not the owner, right? I'm the CEO. The new buyers, which is people that I know through DomWells call me. Hey, we just bought this agency. Like, oh, I'm a client. I buy leads from them. Like, well, we want you to do sales for it, right? Because we know you're good at sales and good at SEO, and it's hard to find someone who can do both that doesn't run their own shop already, right? Well, I do run my own shop, but I'll listen because I like you guys. And so from there, I said yes. And four years later, here we are. I moved to CEO literally a month and a half into the job because there was a big gap there in leadership that needed to be taken care of, right? So I've been running it for four years. And interestingly enough, before I said yes to this, all I knew about Lean building was if you don't know what you're doing, I would it to a pro. So I had to become really good at actually doing, which, funny enough, I've gotten really good at the running of an operation that builds good links. But I don't build any links myself, of course, because how do you build the links required to fulfill multiple seven figures a month? It's insane, right? So, yeah, I'm very focused on SEO, leadership, marketing. Link building is what we do. So I'm interested in the science about it and in getting people who are good at execution. That's the things that I do.
[00:03:44] Speaker A: Awesome. Well, let's talk about the science and then we'll circle back to the execution because it's something that I think, think has a lot of opacity to it because in part, it's kind of like a gold miner. You know, there's secret thought, there's processes that you have figured out that are successful, that if you found a, you know, flakes of gold in the river, then you don't go shouting into town and make a big Purchase because then somebody's going to jump your claim. It's been part of it. But also there is a lot of, you know, intentional ambiguity in the signals that you receive back from Google when you build links and a lot of fear. Fear Mongols mongering Thud I believe is the term that they put out there. There's a lot of stomping around like an angry gorilla making really strong sounding statements while puppy on the other side, the algorithm might be playing with you like a puppy dog or you might be getting bit by something you don't know. So let's talk first about science of link building and what we know from whom and how trustworthy they are. Because there's, there's variety of people who have analyzed this. I'm curious, who are your touchstone people in the link building side and the authority analysis side that you have seen have done good work in the past or have foundational mathematical viewpoints looking at our understanding of the link graphs, authority, knowledge graph and the, the state of play.
So let's go with that angle. Who, who are some of the touch points who have data that you reference and what's your general take right now on the algorithm when it comes to links?
[00:05:33] Speaker B: So it's been an interesting journey on finding out where to get truth from. Right.
Because before I started running Gammy links, I was only on the buyer side and I would fundamentally trust whatever the key people of influence at that time were teaching. I was deploying sort of the same type of methodologies that they were deploying that they were talking about on the case studies on my own sites. And they worked, but it was more of like, well, if I do this, it works, but not truly understanding what's behind it, only surface level. Right. Because one thing is reading the script and being able to repeat it and one thing is understanding it. Right. So at that time I was following very closely anything that Charles Float was putting out, anything that my diggity was putting out. Right.
Very deep into what you would consider gray hat. I had my own definition on black and white hat, all that stuff. Right. And lately, big jump there, covered in betweens, Right. I started paying a lot more attention to Ted Kuwaitis and all the people that are deeper into the testing community. Right. But what's happening those four years is with the agency, we were of course given a lot of processes and a lot of internal ways of calculating link gaps in the graph.
And so from analyzing those and understanding, aha. So this is how these guys built the agency and scale it because they had the capacity of analyzing a backlink profile and just bucketing things up. Okay, this is the authority that you need. This is the relevance that you need. This is the kind of contextual page and website relevance that you need. This is the geographic relevance that you need. Based all of it in what's already winning. So I've dedicated the last four years from the analysis point of view learning how to read what's working and how to replicate it.
And then the principles behind it are complementary because otherwise you cannot course correct when things don't work the way that you want them to work. And also if the landscape changes dramatically. Yeah, now you're a bit naked because well, what you were doing before doesn't work anymore.
[00:07:31] Speaker A: Right.
[00:07:31] Speaker B: If matching what the competitors in the top three do no longer works in a year, I need to know how to read it in your reality before I have a full year to figure it out. I need to already have the knowledge in my mind. I don't know if this fully answers your question or at least some guys happy to.
[00:07:48] Speaker A: No, definitely. So I work a lot with my friend Michael McDougald of Right Thing SEO. You know he's, we're mapping out together kind of he got a crawl of the Internet and we're. He's setting up his own version of, of PageRank to kind of understand the proxy of that. So from that perspective of you know, one of the going back to the foundations of what made Google Google was the concept of PageRank of hey, you have a link to a PA page, it links to five things. We split off the value that one that comes in becomes divided between those five links and that original value is derived off of how much, how many links that that other site got from other sites. So it's kind of the, you know, the link juice, the pouring bucket model. Now we know that there are, there's anytime you listen to a Google there's like two sides of their mouth and they're saying things that are absolutely true and they're saying things that are absolutely true in a useless way. We are know that algorithm is much more complex in that PageRank now plays just one role among the signals that are in play.
I have an awareness of these ones anchor text, the value of the link, the reasonable surfer impact meaning where the link is used on the page, whether it's in the footer, whether it's in the meat and potatoes as I call it or whether it's in the navigation or secondary link, the passage rank. What is the relevance of the text before and after the link, those are the primary ones that I have been paying attention to. Are those in the same ballpark as you? Are there other signals that you're also considering and what's your viewpoint on using those as the core value propositions of links?
[00:09:38] Speaker B: All of those are to my knowledge and experience completely correct.
And of course a lot of tools that have been developed on the knowledge also support that surrounding text around the anchor text also passes relevant signals the anchor text in and of itself. And within that we have the anchor text ratio of both the page and the website.
That is in itself another signal. Right. You can no longer rank with 100% exact match anchor text. That worked for a while when SEO was a lot younger, but it no longer does. Matter of fact, if you run analysis of a glim profile that's winning right now, you will encounter in Most niches virtually 0% exact match anchor text. You have quite rich anchor text in some cases. Unless we're talking about grainish stuff, casino, all that stuff. That's a different game. Right? Reasonable surfer. As you said, one in which element of the page does the link live? Is it in the body header, footer sidebar? Body links matter a lot more then not only the probability that the link will get. Click. In terms of the position within the page, links on the fold matter more is the actual user statistic click rate on the actual links. Those also send signals within the page itself. Then we have to account for the amount of PageRank that the actual page URL strength. Right. You are, if you use ahrefs terms, what's the power of the container page? What's the power of the container website? The relevance of both relevance of the website and the relevance of the page. The velocity at which you are acquiring links. The velocity at which you are winning and losing links. Right. Link velocity in proportion to website traffic and in proportion to brand search volume and in proportion to out of search traffic sources. And there's more. Right. But those are the biggest ones that we always look into when we're crafting a strategy. There's a lot older ones, but they're smaller. So if a site is getting traffic from a lot of different sources, it has good brand search volume. Right. It's already in a positive ranking state and it already has. Oh, this is another one. Right. The order of factors. This is a bit like chess. Building a backlink profile is a lot like chess. Right. You don't start a brand new website and you run it with 100 digital PR type of backlinks because that just doesn't fly. Where, where is this site even coming from? What's the basic bare bones knowledge graph of the entity for the brand? Where are the social media profiles? Where are the citations? Where are the BBB listings? Where is DMCA listing all these different things that need to be in place? Right. Has the author, the author published a book on Amazon? Even if it's the two pages, you read the book. Pretty funny, right? Do we have any sort of sources that verify that this is a real entity other than Newswire? Right. That doesn't fly at the beginning, but at some levels that's exactly what you need. And then it's the proportionality of where you are on the page rank spectrum, which is not a linear graph, it's a logarithmic graph. Right. So if you're already in the leagues where all your competitors at the 80 and above websites, the links that you need to get to the next level aren't like the links that you used to get there. They're very different kind of backlinks and you have to acquire them at a very different cost per unit.
So there's a lot of things there in the deployment. It's very strategic. My angle, it's always been very strategic like that.
[00:12:54] Speaker A: I'm curious, you know, I was talking with Matt Brooks of seoteric about this, about the concept of distance to seed and that within verticals, within authority niches, Google has identified hubs, you could call them authority sites, from which topical authority spreads.
In what way do you need to be aware of that concept? Is it, is that helpful or hurtful? I think in one way it could help because how you approach SEO for, you know, literally for SEO, you know, our niche is treated differently as a vertical for links. I've done link building for locksmiths. That's a dirty game and the players are dirtier and the calculus on that is a lot harder. I've worked in E commerce, in that math in commerce. I've worked in your money and your life niches and that's even crazier. So what's your experience with that verticality? What is seems to be true in one niche, doesn't seem to be true in another. How do you approach that logically or strategically to get a good foothold?
[00:14:08] Speaker B: Yeah, so there was, there was two points there. I want to make sure that we don't forget the part about the seed list and the distance to seed, because it's a very important one and within that there's a sub idea. Sorry, Arborist. And thinking neurospicy united here, right?
There's a secondary idea around. Not every link has to hit a home run and not every link has to kick all the boxes, but the profile has to tick all the boxes that the winning profiles do. So from that idea, ticking all the boxes that the winning profiles do, comes the idea of how do you read any given niche to know what's needed to rank For a locksmith in Alabama to a hairdresser in New York City, to a lawyer in Cyprus, completely different niches, different verticals, different countries.
The answer is always in the data. Now, the data is always incomplete, right? So there's a combination of the data and the subject matter knowledge on the particular niche. So two things there live at the same time for the data. We've devised our methodology. It's on our website. That's what we advertise, how we get clients, because it's good is the link report, right? Link back analysis, link gap analysis, backlink analysis, whoever are the top three ranking for any given niche. Now, there's multiple layers into that, right? You got to look at the domain level. How many total referring domains do they have? What velocity of those domains do they have, right? How much traffic are the sites pointing at them have, right? I always equate traffic with trust, right? If the traffic is real, not some manipulated keywords to inflate AHREFS graphs. If the Dr. Is real, same Dr. And keyword manipulation are so easy to see in Ahrefs, the only footprints that you can always see, right?
So if the sites are pointing at them are real, right? From those websites, starting with the do follow links, which we do know pass page rank, not follow also pass page rank sometimes. But the problem is we don't know when, so we enter into that later. So we try to go with 80, 20 within here we know this for sure matters dividing by Dr. Buckets as levels and then relevance multipliers. And from there we get a certain amount of scores. I say, okay, the website looks like the average winning side looks like they have 20 points in the Dr. 30 to 40, 25, 4 to 50 and so forth. Right? And then the general amount of relevance that they have is about this much. The velocity is about this high per month, right? And we try to chart it all out and then we got the keyword level. Okay, yes, These are the three competitors that you have, but not for each one of your top 15 keywords you want to rank for. For this, let's say locksmith in Alabama, right? Not all for 15 keywords, let's say for 10 different locations and five different services. Imagine, right, Very typical, right? For our local businesses, a handful of different services for a handful of locations, starting with the homep being the main service for the main location.
[00:16:51] Speaker A: Right.
[00:16:52] Speaker B: If the homepage tries to rank in a landscape, where are all Dr. 90 plus inner pages? Because somehow it's a very weird industry and you're trying to run with your doctor homepage client.
[00:17:04] Speaker A: Oof.
[00:17:05] Speaker B: You need to start raising some flags there for the client. And you show them the truth. Right, Listen, this doesn't look very good. Let's look at the other sub locations. And then you see, okay, inner page compatible competitor of, let's say, 20 doctor points ahead of you. And then you look at the doctor make sure that it's legit, it's not your citations beefing up the doctor Right, right. And you start going item by item. And so we have all that, let's say, charted out, right, in a report.
Now from there comes the sub idea of the completion game.
So this is something that I got from Takeaway this a while back and I really like is the analogy of SEO with a simple.
What do you call it in English?
A puzzle. What's it called?
The name skips my mind right now because it's a bit late into the day for me. A puzzle that you make with little pieces and it just has an image puzzle.
[00:17:57] Speaker A: Yeah, Jigsaw puzzle.
[00:17:59] Speaker B: Jigsaw puzzle, thank you very much. So it's no different, right? If you link at a link graph or SEO as a whole is much better put, Right. There's regions of the puzzle that are your technical foundation. Then you have your content, topical relevance, page level, quality, quality, massive air quotes, what quality really means for robots, different from what it means to us and SEOs. Right? And then you have the same for the user signals for the brand signals. Right? What's the nav boost score? What's Rankbrain saying about the site? What's the knowledge graph? Does the author have a knowledge panel? So many things, right, that are parts of the puzzle that make the entity stronger, bigger, more visible because it's more complete. So SEO is a game of completeness, link building same is exactly a game of completeness and the beauty of the whole of it. But of course, especially about link building is you only have to be as complete as the guys on the top three, then this much more. Everything else is an excess, which you might want to do if you can't catch up on the other dimensions because they're harder so if Your competitors have 50,000 pages indexed and they're all great quality and you're starting brand new, but nobody build links because it's all directories. Maybe go the link building and website authority route as you build up the topical map because you're too short on content. So maybe branding and social media virality, traffic and links might do the job for you because you cannot catch up on content and vice versa. And if you try to outrank Adobe. Good luck. Just good luck. Right. I actually ran a link gap analysis of Adobe with competitor. This is probably about a year ago from one of the agency clients that we have and I think the link gap came to like 1.5 million to begin with, right? Like yeah, it's probably twice. And then after the acquisition of Semrus, that's it, that's the end of it. Right?
[00:19:41] Speaker A: That's, that's incredible. I love that viewpoint.
What does, what impact or dimension does locality play in terms of getting the links? Is it about a connection between Google's understanding of an entity being associated with a particular geographic location versus like a page level relevance, connecting it to that particular location? Is there, have you seen like a differentiation between those, those two sets of signals and then kind of talking about, you know, building, you know, local link building. What is the value of, you know, having any particular emphasis, trying to get a city hall link from a city where, you know, you have a service level vendor for an area, you know, if they're delivering, you know, pre built concrete walls, for example, you know, if they're doing it in a particular city, is it the same if they get a link from a national website versus, you know, a local newspaper?
[00:20:46] Speaker B: Very good questions.
It's a slight oversimplification, but it serves for the purpose of not going mental here. I divide the core elements that comprise a backlink into trust, authority and relevance. Now geographical relevance in a way to my experience, lands more on the trust side.
Right. So it's worth making a note here not forgetting the idea where we are around the concept of eeat, right. That's been thrown around since Medic and all the other updates. And you know, when it first came out I didn't know what it was and everyone was saying it's this thing and now you have to do these things. And it was early in my career so I kind of bought that and I, you have to have eat, you need to your author page and you need all these different things. Right. And then over time my experience has proved me that I was wrong and, and EE is an amalgamation of a certain set of signals in particular. So not dissimilar to the idea of integrity, which if you ask most people out there, hey, what are your core values? How do you drive your life? They will tell you integrity is at the top. And I will tell them, that's amazing. What do you mean by integrity? Integrity is not a value, Right. The dictionary definition of integrity is a commitment to a set of artistic or ethical values. Right? So it's E equals C of range, basically, right? So what's within that range of values? So eat equals cluster of what? Right? Eat is not a signal, it's not something. And then we enter into like what's the distance to the seed list in your backlinks? Into how many of those backlinks and how many high trust backlinks do you have? Interestingly enough, this is exactly the same for ChatGPT because of course, Geo is just SEO part 2. We'll see if it's like the Godfather or like different trilogies. We'll see what kind of part two we have in front of us, right?
But we do know, we've known this in SEO for a long time, that there is such a thing as the seed list. As a list of real world should be possible to trust them entities. Of course there's corruption, but there's a different story. Entities that exist in the real physical world that we all interact with whether we want it or not. Governments, institutions, universities, all the staff that represents the functionings of the world, they all have websites. That's the OG seed list, right? Gov Edu.
From there you have all this stuff like, okay, sure, Mayo Clinic, John Hopkins Hospital, sure, right. Big hospitals, all these different things for certain niches, especially your money and your life. Financial institutions, hospitals and other such institutions that have websites tend to carry a lot of weight, right? And then you have Internet institutions that have become so big that eventually get added to the Tier 2 list. So in ChatGPT in particular, the Tier 1 list, Tier 2, Tier 3, the Tier 1 list is all these institutions that we're talking about and then you have the dedicated websites for specific niches that are considered to be the number one authorities in health is Healthland.
If you get a link from healthline, contextual backlink within editorial piece of content at the top of the page that links to your brand by brand name. Be ready for ChatGPT to fall in love with you forever. Because every single time there is a query related to what your business does, Healthline will be the one place that ChatGPT consults always by default. Right.
So very similar to that. Now going back to the idea of links and the link graph, right?
We have all these different websites that are ultra high trust because they're manually vetted green tick by the makers of the platform. No link is better than that. And it doesn't matter if you're a local plumber in your city in Tennessee, right. If you get a link from NASA, somehow your SEO will do good.
Because of course right now it's missing all the topical relevance. It's missing a lot of things. And yes, your backlink profile, if that's your only link, if you Google from, I guess a link from NASA, very weird. Nothing else, no Facebook, nada is going to be a very unnatural link profile, but you will still receive a boost. It's just you're missing so much on the other completion game that the link might not do what it needs, what it, what it can do because it's not ready to skyrocket. Never better said, right?
[00:25:02] Speaker A: Yeah. I did a consult for a remote telescope telescope company and you know, you could, from anywhere, you could, you know, pull it up and look at the stars. That particular. So it's very popular for schools. They had articles from, they had links from NASA, they had Washington Post, they had New York Times, and they, their head developer had no indexed their entire marketing site aside from the homepage.
They're like, we don't understand why none of our content is working. I'm like, because you put a no follow.
You, you made it invisible. You put a no index, no follow on all of it. And he's like, why did you do that, John? Like, like, I thought it would help.
[00:25:50] Speaker B: I just thought he was a millionaire, but when the lali.
Yeah, he was a millionaire, but when the Dali painting went on auction, he was dead for no comeback, sir.
[00:25:59] Speaker A: Yeah, so what? You, yeah, there are definitely scenarios where you, you can get powerful links, but you can fudge it. You can flub the ball on the architecture on canonical masses. There's always canonical masses with any entity of size in my experience.
[00:26:18] Speaker B: If I may complete something from earlier, because I believe I didn't actually give a complete answer and I don't want to do a disservice to your audience.
If you run a local business and you can get a link from any local institution that's based in your town or state, get it.
If you can get it, get it. Because most people won't have it. So SEO is not only a game of completion and a game of similarities. Plus A little bit better. It's also a game of difference. If you're dramatically different from everyone else in the right direction, you must be good, you must be better than. Because if nobody in, let's say, well, let's use a different, let's say mobile vet clinics.
Okay, yeah, why not? And it's a city that has 10 competitors.
I think one of them has a link from the local board of vets and one of them has it.
That site has massive advantage.
[00:27:14] Speaker A: Simple as that. I agree, I agree. You touched on it and I'm going to make you dive into it a little bit more. What's your understanding of LLM interpretation of link authority? Is it as complex as and as robust as Google or is it the toddler knockoff version of links? But is that balanced by the difference in their machine learning application? Does it matter as much?
[00:27:42] Speaker B: You said it with kinder terms that I would have used. Toddler is a very good analogy, yes. So to put things into context, right, what you said at the very beginning, Google became Google because a group of very, very smart people figured out that if they build the mathematical equation around what happens in the academic world with research papers getting quoted by other papers, how do you rank what paper is more important? And a hierarchy of content, right? So that researchers can continue to develop their work according to the most important research, not some super obscure stuff that might not be as important.
Right? We are the researchers, the users looking for information. We are everyday scholars of a my cat shot in my rug. How do I clean this up? Right? It's a much simpler problem. I'm not trying to put a man on the moon, right? But I still need a paper that will tell me how to do it because I don't have the knowledge, because it's the first time that happens to me, let's say, right?
So that built Google and it built it into what it is today. And so they've developed an outrageous amount of time and effort into polishing the system for PageRank and all the surrounding ideas around PageRank. Anything from Rank Brain operating as a kind of PageRank based on user signals for the people who are getting traffic, let's say in page one and beyond.
It's different kinds of PageRank. So it's algorithmical signals that are multiple algorithms at the same time stacked up, all of them with different weights that also talk with one another and update themselves.
Is an extremely complex system in order to rank who should be in this list of top 10, because it's a list game right now with LLMs is a completely different game because LLMs are rewriters or the story. This is very important. LLMs rewrite the story.
Google now with AI overviews. Sure, but it's Gemini powered. It's the same. It's an LLM for queries that don't have Gemini on them just yet. Right. They don't have AI Overviews is a game of who needs to be on this leaderboard. There's no story. The story is the leaderboard. Right. Leaderboard is a story with LLMs is I'm going to use whatever sources I feel will tell you the most valuable story to solve your problem. Completely different approach, yet very, very similar. Math.
Because of course, because of course, because you're ultimately doing a ranking system for knowledge and it's all written in math. Language is secondary math first. And so from without necessarily going to like, how does an LLM put words together? Which is still important, right? Like how does it know what word goes next? It's all the vector forces within it. If we think of words as atomic elements, right, Trying to move in the vector space, which one of them are going to cluster together and then just give you the full string, Right. When it comes to backlinks, that's one of the forces within that, if you want to call it field, from which the LLM will go and select words and spit them out on your screen. Right. I always visualize it in my head. That's literally like a little ship that goes through this massive map of meaning and jumps from one word to the other through sort of like wormholes based on the vector forces of every word plus the forces embedded by the input. And then it chooses what to spit out. So where links come into play, here is one, the tier 1 2, 13 list. Tier 1 list. Are you featured there around the idea of X? Every time somebody makes a query for X, the LLM is going to default one to the training data to the tier one list. If neither of them provide the LLM with something that the LLM based on the supplementary training that it has on what a good answer looks like, then it will say, let me Go Tier 2. And those are more niche specific websites. Your own website could be there, Right. You could be the source. Right. If someone is making a question about Lean building, I know ChatGPT has read my entire site. It's most likely that If a tier 1 site and its own training data doesn't have the answer, it will use sites like mine, hopefully mine, provided I've served that as a source very Effectively, So I don't need links, but I do, because how does ChatGPT know that it should use me as a source of knowledge and not somebody else? And the answer, 80% of the answer is traditional SEO. If you rank for a lot of queries, because I think it, in a way it sort of builds top to bottom and bottom to top. So the tier three is when there's nothing on the tier two known websites that are niche specific, that is in a secondary list that's a lot wider and there's multiple vertical sites that ChatGPT has found through other methods, divination or search. Right.
Then the last answer is the tier 3. Tier 3 is let me make a query to 20 different search engines and cluster together with what they all say, plus whatever I have from the previous T, though that might be incomplete. So it completes that with a tier three search and it looks at Google, Bing, ChatGPT Go Yandex, everything. Right?
Very interestingly, for a lot of local queries, if you ask ChatGPT a straight up keyword, the same way that you put it on Google, Alabama plumber.
Right. And then city. Right? City in Alabama plumber, maybe your neighborhood or broken pipe in borough in city, literally just put that right, Most likely it's going to perform a bunch of queries and you will see that the results that it offers are eerily similar to whatever is on Google and Bing on the map pack and whatever Bing's map pack is called, which I don't know. Right, right. Bing's map pack is massively influenced by Yelp and Facebook pages.
Google is massively influenced by the actual GBP user data.
Yeah, and your actual organic rankings. Because yes, both algorithms, the algorithm for the map pack and the algorithm for organic results are different, but they communicate with each other. So where do links come into play here? Well, links get you to rank in the organic. We know for a lot of long tail queries, ChatGPT and other AI searches where ChatGPT has over 80% of the market share. It's a bit nonsensical to spend too much time thinking about how do I rank in propensity. Nobody cares.
Right? Like ChatGPT against Google. I was just going through the data. Today, ChatGPT has got 6 billion views a month and Google has 85 billion with a B. Right. So if ChatGPT, which is 80% of the AI chatbot market, has 6 billion, which is that against Google, then perplexity has 10%, which is 8 times less than ChatGPT. There's no point in going there. Right, so how do we bank in chatgpt? What with right, good rank on Google as half the job because all the long term queries related to your industry, that's how it starts getting additional knowledge. Then serve in your own website, a lot of good. And by good I mean served for robots content that they can utilize as a source, using you as a source because that's how you get quoted, right? And then work on your online reputation. So knowledge graph, knowledge panel reviews, online reviews, online sentiment matter tremendously whether you are very active on YouTube and Wikipedia, which seems to be pretty like children of source of information for ChatGPT. I don't know what's the state of affairs right now with Reddit. A lot of things are changing right now. Obviously Reddit was the oh my God, we love Reddit. Everything that Reddit says is gospel. Right. And then there's been a bit of a change of heart, right. With search engines, we'll see what happens there. But UGC stuff matters. And then you have links from tier 2 and tier 3 sites that aren't you.
And what matters on those links is that they mention you by brand name. It's all about brand recognition. It's brand entity and it's the semantics around the brand name. So the anchor text has to be the brand name. Always right. For ChatGPT, of course for SEO. Traditional SEO is a ratio. Typically brand name wins. You need brand name on the anchor text, you need the sentiment of the page to be positive, very important. And you need the surrounding context around that to have to do with what you do as a business. So when somebody is looking for what your business does, if ChatGPT hasn't found you, but has found 300 other tier 3 websites that they all mention you, it has enough consensus, which is the rule that rules ChatGPT. ChatGPT will defer to what the consensus of the information available online say about a brand or a query. And it will give you that an 8 out of 10. Only it will never give you a 10 out of 10 because for a 10 out of 10 it needs to commit recent risks which then risk hallucinating a lot.
So it will tell you a story based on what everyone else is saying and mainly influenced by review aggregators and other sites that link to you and the sentiment of it all and then your own content, if you serve it well and if follow that says that you are the shit, then you will get recommended.
[00:36:46] Speaker A: Validate a theory. Do you think that this is accurate or not? I think and so does Darth Na. I don't know if you follow Lyndon Darth Vader.
He's. He's on Twitter. It's Darth Na, a longtime player in SEO, and we were.
Nobody's like, he maintains an online Persona that's anonymous. It's always with, like, a Darth Vader helmet. It's hilarious. Anyways, we were postulating around hcu, the helpful content update, and he said that he suspected that it was a combination score of two things that you've already touched on. One being actual click behavior and interaction with the brand, but also distance to seed.
And those two used as a validator for a validation factor for, you know, this huge content spread. Like, if you have a ton of content and a bunch of it is thin and old and not updated, then at a certain point you trigger an HCU penalty of, you know, trying to scale an operation without the.
Without the cojones to back it. And the proof of that kind of comes from the pudding of looking at who was hit first by hcu. You had, you know, like the birthday sites, you know, and they had a huge, vast programmatic library of all of these content, all of these content pages for all of these celebrities. I saw another site that was hit that obviously they taken one piece and had taken every question ever asked by it. Taken. The people also asked and then just kind of lightly rewrote. Exactly. The people also asked entries of the questions and answers and made their own site. There was no links to it. There was no authority to it, both in terms of, you know, backlink profile and then click profile. Nobody was clicking through. There was no brand name associated with. With the entity.
So would you agree that those.
That hcu, possibly or in part or in full, is impacted by links and clicks and brand identity as much as some hypothetical helpfulness, amount of, you know, is this content actually helpful? What's your take?
[00:39:04] Speaker B: Well, I'll tell you where I come from in order to answer this question one is when the original ACU came, part two was March 24th, which is, in my experience, was a worse. Much, much worse scenario. I had a bunch of sites that got hit. I had a bunch of sites that didn't get hit. I had some clients that got hit, clients that didn't get hit. I tried to make sense out of that based on what we were being told, and it made zero sense. Right. What was presented. The update is about this. And I look at who got hit like, it cannot be.
It cannot be about that.
Because these 20 sites, I build them all the same. These seven random ones got obliterated. These three are down, these five are up. And these two, what happened to them? They're the same, right. Then with clients, kind of the same story.
And then I started going a bit deeper, right? Following what some of the peers around me were saying is I've been very, very fortunate to be in Chiang Mai for the last decade. There's a massive SEO community there, right. A lot of people that are not necessarily very vocal online, but they're really good SEOs, right? So I get to learn from people.
I started being pointed towards the look at how expensive is the website and how much money does the website make to Google. Like, huh, okay, let me go into that. So a lot of sites that got hit weren't running Google Ads because they weren't real brands. They were content players, they were affiliate sites. Right? Okay. Lots of affiliate sites got. Got hit. Why? Because most affiliate sites don't run Google Ads.
Okay. There's not the why, mind you. There's one list. One on the. On the list of possible whys.
[00:40:37] Speaker A: One of the factors. Yeah, what else?
[00:40:40] Speaker B: A lot of sites were very expensive. Expensive to crawl, heavy sites with tons of content that by itself the content didn't target money. Keywords in a way that would make Google money.
Google wants to make money. There are keywords.
If you type a keyword and you see five ads in top and you type another keyword and you see nothing but AI overview. On the AI overview, Google is losing money already. So they will rank pretty much anyone there for as long as it hits the mark of what complete SEO looks like in a keyword where The PPC is $100 per click. They're not going to rank a nobody there. Right? They just won't. And so they have to vet it through so many layers because if that SERP gets devalued, they don't make money. And so a few things came around the time of the etc. There was a very interesting study. I wish my memory was sharper around this. Too much stuff to hold in the mind. There was a very interesting analysis of some of the drama that was going on on Google around the people that was at the head of search and some internal ploys to make sure that they hit the ad revenue numbers. So they had to mess up with layout.
And so the summary of it.
Great. The summary of it was they had to mess up with the layout of the SERP in order to maximize ad revenue, to hit certain revenue targets so that the shareholders wouldn't eat heads off this is one of the problems with publicly traded companies and, you know, growth at all costs. Right. So at the cost of the product, they got what they needed from an ad standpoint. Now, I'm saying this hypothetically. I don't have the actual facts of this. I'm not saying that this is debatable truth, but to our knowledge, when we read between the lines and some of the stuff that was covered, somewhere in my bookmarks, I have the links, if I can find them.
Right. I believe I mentioned this on some of my early videos on YouTube, but it's been almost two years. I forgot some of the details. Right. And then you have a bit of what happened happened in 2024. So in 2024. Right. Mind you, I had been running the company for three years already, but the company was five years old. So we had a list of maybe 300,000 websites that we had or we knew that we could collaborate with to get backlinks on those websites from the sites that everyone can get a Link on, from $5 to the sites that I can get a link on because I'm friends with one of the editors.
The whole spectrum.
[00:43:07] Speaker A: Yeah.
[00:43:08] Speaker B: To surprise to absolutely nobody. The crappy sites where everyone and their dog could get a link. Dog wiped, like, wiped out. Disappeared in a matter of days. Right. To the surprise of nobody. Which is the reason, like, we don't operate with any of those websites for a very, very long time. Right? Yeah. In 2019, you could sell links there and run to people with those things because it worked, but not anymore. Right.
Then you have the sites that they themselves have enough trust. Right. They had enough EIT of their own accord that they could survive being put at the examination of how expensive are you to me?
So do you publish content that makes me money? That's the first question that Google will make. And is it going to cost me a fortune to scan your website? Because then open any Google search console, you go to the Pages Index tab and then you will see, okay, 25,000 pages on the gray tab and 2,000 pages in the green tab. Right. 25 pages found but not indexed. 2,000 found that index. How big is your website? I have about 200 pages of content. Well, you have 10 times as many pages of content indexed than you should have and then 10 times more than that that Google is finding. So for a 200 page website, you ask Google to look into 27,000 URLs, do you think they'll be happy? That cost them money. And then you go to like, what's the average loading time and what's the average response time? Right, that's in settings crawl stats out of the three charts. The orange ones for those of at home watching. Go to search console, go to settings. That's the bottom left on the corner, right? Then the crawl stats is in the middle and then you will see it there. Open it on the time to face byte average response time. If it's anything higher than 200 milliseconds, congratulations, you're too expensive. Right? And so the moment that you start having all the 404s, 301s, 500 errors and your site responds in three seconds, that's three seconds. And the crawler is costing money before it gets an item answer, right?
So cost of retrieval, that's the fancy word for it is actually paramount. And cost of site as an aggregate versus value of site, that's a big one. So helpful for whom? Helpful for Google's finances. But of course, it's their platform. Of course. Why would we get angry? Imagine that, you know, the people from Riverside is, hey, Your plan says one hour per episode. You've hit one hour on one minute. The next one hour is an additional $2. You say outrageous and you flip the table, of course.
So we kind of forget that we are making money in their home right through the traffic that goes to their website. So we got to give them something that gives them more than we take.
And nobody thinks about that.
[00:45:44] Speaker A: It is, it, it is something that I've had conversations about in the hallways of conferences after the big presentations, you know, or in the, the in or having conversations at the conference, you know, at the bar afterwards.
What are the considerations that Google is making to select these, the list of sites that are ranking right now.
How much do they cost? And also, you know, there is, there is something to a feeling, I don't know, it's a feeling of people who have created this content and they're like, well, I just made this content so that the LLM could eat it up and then make a bunch of money off of me. And that there's a challenge of that symbiosis. You know, like, we both do need each other. Google has recently with adoption of Gemini and an AI overview layer of eating a bit of our sandwich. And that I think has caused a lot of resentment. And in our industry of like, hey, you told us if we make good stuff, you'll run ads on it, but you'll send people through to us like that's the devil's bargain. And they've slowly, you know, first it was featured snippets, you know, they released that and we're like, hey, you're, you're generating stuff and eating up the clicks. I think the stat of zero click searches, you know, continues to go up.
[00:47:10] Speaker B: I want to, you know.
[00:47:13] Speaker A: Yeah, yeah.
[00:47:14] Speaker B: May of 2018, Google removed most instances of the motto don't be evil from the main body of their code of contact. And the new code focus on things like respect for users and respect for opportunity and respect for each other. It's a very different thing. Which you said, the devil's bargain. Right. And so that immediately made me think of the don't be evil thing.
Not dissimilar from the IBM deck from 19, God knows when I think 1982.
Computers cannot be held accountable, therefore computers cannot make managerial decisions. Right. Which a lot of people are now deferring their intelligence to the Oracle of Delphos ChatGPT, which I don't recommend people to do. Right. Because it hallucinates and it can be manipulated and it's ultimately not smart. It might be pseudo intelligent, but it's not smart. No difference there. So these conversations that you were having thinned on the money and to my experience, yes, of course, now people are pissed. But why are people pissed? Right, I understand, I was pissed also, mind you, I've, I've made some, some video videos earlier in, in my career saying oh, Google this, Google that. And then over time I'm like, why would they care?
Why would they care beyond caring for the business first when that's, that's their commitment to the onus of the business, which are the shareholders, is that the business performs. And so sometimes for pure business performance you might have to sacrifice users, if that is your North Star is pure financial performance.
And you might have to do a short term compromise to the users.
Yeah, yeah, they might get outraged, but you'll survive. People didn't like transferwise becoming wise, they like transferwise more.
And over time people forget like, oh yeah, sure, I'll, I'll send you a wise transfer. But people like the old name more and for a while they had to piece of the user base. That's not necessarily how I would operate and it's not how I operate but at those links and it's the state.
[00:49:15] Speaker A: Of the game and you can only be so mad at the meta. There's a lot of money to be made and there's a lot of challenges. I was talking to somebody else about the reality of my last interview. I was talking about the reality of operating your business. And somebody else's ecosystem. You know, Google owns the ecosystem, correct? That's the way it is. If you are a vendor who has a plugin for HubSpot Spot, you exist within their ecosystem. If you're in jobber and you've created supplementary plugin, your model is always limited when you're within somebody else's ecosystem. So after six months, if they realize, hey, this guy's making a ton of money off of his plugin, we're going to make our own version. You can't be that mad. You can be. I think it's justifiable that tipio, if they took a tour with you and they like took your specs and it made, made an exact duplicate, then that gets super shady. But at a certain point, you know, like you, you, you have to decide if you're going to continue to play the game in their ecosystem. And you know, for better or for worse, you know, being an SEO inherently means that I am attempting to tamper with in some way the signals that the algorithm is interpreting.
100% that, that's my, I get paid to tamper with the signals the algorithm is interpreting. So I cannot be mad at that system reacting. And I would be a poor SEO if I didn't figure that into my long term strategy of, okay, what do I.
If my, if I'm only working tactically and this tactic only works now, they are going to pivot. You know, if you put white text on a white background for ChatGPT, at some point they're going to read the hex code just like Google did in 1998.
The message is there, like it, you can't if you're working in somebody else's ecosystem, I think, quote Lily Ray, it works until it doesn't.
[00:51:16] Speaker B: Yeah. And I was going to say, right, all these enhancements that we make to the right signals so that the people who are trusting us with their money and their CEO business performance, right, all the tweaks that we make are so that Google can send them Google's users.
It's like you come at me and you pitch me a good idea and you ask me, alejandro, you give me a thousand dollars to bring my idea to fruition and I say, yes, it's my thousand dollars right? Now it's yours. Now it's your thousand dollars. I give it to you for you to materialize your idea, Right? But we are going to the platform and says, hey, platform, look, I've done all these changes. Does this please you more according to the math that I know, pleases you and the platform is like, yeah, sure, I'll show your site to to more of my users that come to google.com because I've invested billions creating the best search engine there is. That's the game we play, right? Ask anyone on fba. Can high school dropout with very little knowledge of the business world go to China, put a couple products together, slam it on Amazon and make millions? That was a very good game for a while. And then Amazon is like, yeah, you guys do it. I'll give you all my platform, all these other things I have invested billions on, right? And you can go with a couple thousand dollars and make millions. Come on, oversized resources returns. But then if your diapers really sell very good, I might make a copy of them and put them on Amazon Basics. Thank you for the market research you did for me. Listen, don't get mad. You made millions.
[00:52:44] Speaker A: That's the game. Yeah.
You got yours, how much you got changed. So it's probably a game of what is it the phrase choosing beggars now?
[00:52:55] Speaker B: Does it suck? Yes. Is it a punch in the gut? Yes. Does it hurt? Yes. Might tank your business when it happens. And then you have to face some very real first world consequences.
Ryan.
[00:53:05] Speaker A: Yeah.
[00:53:06] Speaker B: Yes.
[00:53:06] Speaker A: I love this conversation. I could probably talk like a whole nother hour, but I'm going to kind of wrap up and bring it back around to what you guys are doing now. What message are you putting out there? Is there any special resource that you're, you're focusing on, on getting messaging out about? Are you guys going to be at any particular conferences or anything interesting like that? I'll make sure that any of that get added to the show notes for anybody that's interested in that having further conversations and shout out whatever social media handle where people can follow up with you and find more about. Get me links.
[00:53:41] Speaker B: Yeah. So I'm pretty active in Chiang Mai and so Chiang Mai just held its seventh, I think sixth, seventh, I don't know. Chiang Mai SEO Conference, Massive one. Right. And so every year I go and I do some interviews with some of the top names there. And right before the TAMA SEO conference, there was the Link Building Mastery conference held by my good friend Jabez Rubin. Right. And so I had the privilege of speaking there. I spoke about something that had nothing to do with Lean Building. It was just my experience being the CEO of an agency for the last four years. Lessons learned, that kind of stuff. Right. And presenting a model that I developed that I call the Entrepreneur Map and Compass, which is a system to make good decisions when you run a business and you didn't have the privilege to get an MBA in business, so you have to learn by yourself. So I devised this own system which goes back to the classics, right? The Tribune Equa Tribune, the basic education that you will get 500 years ago, right?
Grammar, logic, math, basic stuff, right? From there, a model that you can utilize to make good decisions out of. My work right now is on developing the model and making content around it, as well as a new series that I'm doing demystifying Geo AEO AI SEO from the very beginning because there's so much nonsense around it that it drives me insane. I cannot go to LinkedIn because I want to vomit every three seconds, right? Because everyone's oh, chunk optimization. And oh my God, look at my new GEO service for five grand a month where I will do nothing for you because. Because nobody in your niche is looking for solutions inside the btm.
Don't do it, don't do it. Just make sure that you rank in your map pack. You'll be fine. You'll be right, right? So anyway, that's kind of like where my focus is right now in terms of the content. Of course, if you can't. Milling is the lean building agency. We always keep going up with the edge of the blade when it comes to lean building so we can deliver to our clients. It's mostly agencies and CMOs that just need lean building to get results. Results working on the science behind it. And so people can follow me on LinkedIn. I think LinkedIn Alejandro Meyerhands on YouTube Alejandro Meyerhans SEO and soon I'll open a second channel to talk about more of my leadership, entrepreneurship and some more esoterical knowledge around business. I'll put in a separate channel.
[00:55:46] Speaker A: Absolutely love it. Thanks so much for your time.
[00:55:49] Speaker B: It's been absolute pleasure, man. Thank you so much.
