Entity SEO & The Future of The SERPs – Jason Barnard Interview
In this interview I sat down with Jason Barnard, the literal writer of the book on Entity SEO – We had a long discussion about entities, topical authority, AI and much more.
Follow Jason on X – / jasonmbarnard
Timestamps:
00:00 – Episode Intro
01:27 – Introduction
01:54 – Beginning of SEO Career
03:27 – Google’s Understanding of Entities
05:32 – Google’s Documentation and Marketing
08:33 – Google’s Recognition of Expertise
09:09 – Sponsored Message
10:19 – Evolution of Google’s Knowledge Base
12:54 – The Killer Whale Update
16:39 – The Future of ML and AI
20:51 – The Problem of Duplicate Entities
21:49 – The Importance of Being Recognized by ChatGPT
23:35 – Discussion on AI and Hardware Evolution
25:16 – Manipulating Google and ChatGPT
29:49 – The Future of AI and Quantum Computing
33:57 – The Role of Links and Entities
40:04 – KaliCube Pro System and Entity Equivalents
43:46 – The Impact of Being an Entrepreneur
44:12 – Conclusion
🔗 Stay Connected:
Follow Me On X – / charles_seo
Follow Me On LinkedIn – / charlesfloate
My Free & Premium SEO Training – https://charlesfloatetraining.com/
My FREE AI & SEO Newsletter – https://reenewsletter.com/subscribe
Jason Barnard [00:00:00]: Have done every single trick you can imagine in the book during the first years when it was really, really easy. After 2010, started to get a little more difficult, but we could still play all these tricks. And now since 2023, it’s going to get significantly more difficult to trick the machine because the machine is getting super smart. So if we say Jason Barnard is the CEO of Kalicube, it understands Jason Barnard, it understands Kalicube and it understands CEO. And therefore, it understands these two things, Jason Barnard and Kalicube, and the relationship between them, which is, is the CEO of. Somebody said, I made a million dollars in the three months that I happened to rank number one. Remember, two months later, Gary Illyes said the helpful content update was all about classification. So there’s entity understanding, confidence in that understanding, and the credibility of the entity that has been understood. And you need all three or you’ve got no hope. So the day that we get into trouble as a human race is the day quantum computing becomes a practical reality in production for Google and Bing.
Charles Floate [00:01:00]: If I can no longer manipulate Google, what happens if ChatGPT is the future? And I said, well, can you not manipulate ChatGPT?
Jason Barnard [00:01:07]: You’re a writer who specializes in SEO, which is much more powerful in terms of content creation than an SEO who happens to be a writer sometime.
Charles Floate [00:01:27]: Hey, guys, welcome back to the channel and welcome back to another episode where I’ll be interviewing SEO experts and entrepreneurs from around the world. In today’s episode, I’ve got somebody who’s been in the industry for a long term who has literally written the book on entity SEO. Please welcome to the show Jason Barnard, ladies and gentlemen.
Jason Barnard [00:01:47]: Thank you, Charles. Delighted to be here. The book. Woohoo.
Charles Floate [00:01:51]: Yeah, the book on entity SEO, in my opinion. So how did you kind of get started in your SEO career?
Jason Barnard [00:01:57]: Right. You said a long time, it’s 25 years now, coming on for 26. I mean, I started in the year Google started. So me and Google have grown up together. I’ve seen every iteration of Google since the very beginning. And that I think, or if I know, is information knowledge you cannot buy. There are very few people who started with Google and are still with Google and have done every single trick you can imagine in the book during the first years when it was really really easy and now, after 2010 started to get a little more difficult, but we could still play all these tricks. And now since 2023, it’s going to get significantly more difficult to trick the machine because the machine is getting super smart.
Charles Floate [00:02:43]: Yeah, it is. It’s getting to a point now where the kind of Google’s understanding of entities and everything can understanding of the web and crawling and stuff is at such a fast pace and then also with all these new technologies and stuff, it’s just unbelievable. And that’s exactly what I kind of want to talk to you about today. So for the beginners and the non-SEO folks in the back of the room, what is the meaning of entities within SEO’s kind of context?
Jason Barnard [00:03:08]: Right. Well, lovely question, and it actually beautifully follows on for the last one, is that Google announced from strings to things ten years ago, from strings to things mean we’re not going to count strings of characters anymore. We’re going to understand what the thing is that those characters represent. So if I take my name, for example, Jason Barnard, the string of characters, J-A-S-O-N space B-A-R-N-A-R-D is a person and that’s me. But it’s also multiple other people called Jason Barnard. So we take two things here. Number one is the string of characters, Jason Barnard.
Jason Barnard [00:03:44]: Google would, in the past, just count the number of times that appeared. Then it would wait it according to the number of times it should appear on the page. The algorithms were written by people, so the people would give it a number so another person could then guess what the number is or figure out what the number is and then beat the machine because it was written by humans. Now we’re getting into machine learning, where the machine learning, the AI, if you prefer, is understanding what the string of characters means. So if we say Jason Barnard is the CEO of Kalicube, it understands Jason Barnard, it understands Kalicube and it understands CEO. And therefore, it understands these two things, Jason Barnard and Kalicube, and the relationship between them, which is, is the CEO of. So from that perspective, Google understands like a human being. But it announced that from strings to things ten years ago. So I built Kalicube Pro in 2015, thinking I’ve built the machine that can figure out the things part of this, which means we can now beat Google hands down from today.
Jason Barnard [00:04:53]: Turns out it took Google eight years to catch up with me. In 2023, Google finally nailed it. It did a huge update. To its knowledge last year, we called the Killer Whale update. And I’ve been here for 25 years. All the time Google’s been there doing its thing. I have never seen an update this big. It was so huge.
Jason Barnard [00:05:15]: That was in July. And all of a sudden now, the machine I built eight years ago is the best way, the only way to manage entities for Google. If you want to beat the strings to things deal.
Charles Floate [00:05:33]: Yeah. Well, it’s impressive that you kind of preempted Google and that Google only took that, well, it only took eight years, but it took a long eight years to kind of catch up. It’s always kind of been the case that Google’s documentation, I think, has been what it wants to be, rather than necessarily what it actually is, which is why I’ve always kind of told people to take with a pinch of salt and kind of look at what the SERPs saying and not necessarily what the documentation is literally saying.
Jason Barnard [00:06:00]: Which is a great point. I mean, a lot of what Google says publicly is marketing and they’re trying to market their products to us as SEO so that we will do what they want us to do so that they can get to where they want to go. And what was lovely about this is I took them at their word. I was still quite naive, less so today, but it meant that I just leapt so far into the future, as it were, that I now have this amazing product and service that people can now buy and use to great benefit to themselves. And from that perspective, I was lucky. But now I am taking it much more with a pinch of salt. Google say this, Google say that, Google say the other. But look at what the machine is doing. And John Mueller from Google said, outside of Google, nobody knows more about entities and Knowledge Panels than Jason Barnard.
Jason Barnard [00:06:54]: For them, it’s just algorithmic. They’ve written the algorithms, they give it over to the algorithms. But who else in the world can actually make the algorithm jump to my tune? And that’s the entity machine, it’s not the SERP machine, it’s how does Google understand? And the idea of Google understands the world like a human being does is something that people pull me up on sometimes and say, well, it’s not quite that, it’s vectorial representation in the mathematical space. Bloody bloody. But I can’t remember what the silly phrase is that they use. And I was talking to a friend of my daughter who’s doing a PhD at Oxford in robotics, and she does the sight part of robotics, and she said, oh, we understand the world, but we’re struggling to understand it over time and over space. So I asked her to clarify and she said, well, if the robot is standing still and doesn’t move its eyes, to all intents and purposes, it understands the world. And in AI for robotics, we just say, we’ve understood the world as it is.
Jason Barnard [00:07:51]: As soon as I get it to move its eyes all stand up and walk around, then we have a problem. So we can say Google can understand a static world today. It can’t understand very well the passage of time and the movement in space. And if you look at Google Maps, it does understand movement in space. But that’s Google Maps. And that’s a part, that’s something different. That’s a different Knowledge Graph. So Google is working towards this real human understanding right now.
Jason Barnard [00:08:21]: It’s a static, childish understanding. And I’m writing a book now. We’re going to release it in a couple of months. Google is a Child. Learn to make it love you. And if you treat Google like a child, you win the game.
Charles Floate [00:08:34]: Well, that’s a pretty cool quote from John Mueller as well. I’m hoping that he wrote it on Twitter or something so that it feeds into AI for your own entity and that you have Google recognizing that you know more than anybody about it with your own entity. Right. Entity inception, if you will. So how has Google’s knowledge base. You said last year that there was a massive Killer Whale update that changed the knowledge base and kind of changed panels and entity understanding and entity recognition. How has Google’s knowledge base and kind of definable entity recognition changed over the last year or so, or even the last few years? And now for a quick message from our sponsor, search Intelligence. If you’re looking for digital PR campaigns, what you’re about to see should impress you.
Narrator [00:09:19]: What a crazy campaign. How to sleep on your back this campaign got us links in Huffington Post, Glamour magazine, Mirror, and lots of other great news publications. Let me show you how we’ve done it. It was so simple. Our sleep client provided us with expert commentary about how to train yourself to fall asleep on your back. They also gave advice on why it’s best to sleep on your back. Once we’ve had this information, we went to macrack and searched for journalists that consistently write about sleep and well being. We’ve sent these journalists the advice provided by the client and within one day the links started flowing in.
Narrator [00:09:54]: Glamour magazine, a DR 81 website, picked it up. Huffington Post, DR 88, Mirror UK, DR 90. A massive avalanche of links blasted through our client’s website with this simple yet effective campaign about how to sleep on your back. I hope this inspires and I hope you’ll use this technique to land massive links to your or your client’s website.
Jason Barnard [00:10:20]: Well, over the last few years, it’s evolved in kind of these steps. I don’t know who out there remembers the Google dance, but the Google dance was the idea that you update your content. Google crawls it three months later, it then goes through its index again and updates the entire SERPs index. So there’s what we call data lakes. It stores the data in a lake and then it trawls through it once in a while. And so when you change something today, you waited three months, you saw the result of it, it didn’t work, you had to change it again, then wait another three months. But if it did work, you made a fortune for three months. And people were saying, somebody said, I made a million dollars in the three months that I happened to rank number one.
Jason Barnard [00:11:03]: Brilliant. Wonderful. And then it went straight back down again. But I had my million dollars, I didn’t care. And today in the SERP index… Sorry, in the SERPs, we expect things to update in a day, a week maybe. That wasn’t always the case. So people who haven’t gone through the whole history of Google don’t really realize what the Google dance was.
Jason Barnard [00:11:22]: But now we can look at it and say, well, we have three Google dances going on. The index Google dance, which is between a day, a week, maybe two weeks if your site isn’t trusted, and that’s important. It’s a day if Google trusts your site. It’s two weeks if it doesn’t. So if your website is taking two weeks to update in the SERPs, you know Google doesn’t trust you, you’ve got a problem. The second Google dance is knowledge, and that’s three month iterations. So I update today, Google digests three months later. The changes that I’ve made will be reflected in a Knowledge Panel or in the Google Knowledge Graph. And the third one is going to be large language models. Right now, that’s a one year iteration.
Jason Barnard [00:12:01]: If I add information to my site today, it has no chance of being in the LLM until this time next year.
Charles Floate [00:12:01]: Because of the training data and things and stuff, right? It’s like a training delay as well.
Jason Barnard [00:12:12]: Sorry. That’s exactly it. If you just think about lakes and rivers, a data river is just flying by and the machine picks out the nuggets. So if it doesn’t trust you, it’ll just let everything flow by into the lake to deal with later. If it does trust you, it picks out the nuggets on the way past and sticks it straight in the surf. And that’s a really nice analogy, because it does show you, does Google trust you? And if it doesn’t, you need to do something about it. And then the data lake idea is what we’re looking at with Knowledge Graphs and the Knowledge Panel and LLMs is let it all flow by, and I’ll let another algorithm deal with that. And it will deal with it when it’s good and ready, when the engineers are ready to feed it more data.
Jason Barnard [00:12:55]: And that brings me to the Killer Whale update, which is that the number of person entities in the Knowledge Graph. The number of people Google understands, tripled in four days. Now, if you think about the scale here, if it understood 500 million people, which maybe it does, it understands 50 billion entities, by the way. If it understood 500 million people, it went from 500 million to 1,500 million. That’s huge. And that was in four days. So what the engineers did is they had this huge lake of data, they cleansed it as much as they could, and they gave it to the machine and said to the machine, now figure out people.
Jason Barnard [00:13:34]: It didn’t increase the number of companies or songs or events, only people, and it tripled them. So they gave this machine huge amounts of data, unimaginably huge amounts of data, and said to it very specifically, two things. Find the people and reclassify all of them. Reclassify all the people you understood. Then what happened in September, 2 months later, Gary Illyes said the helpful content update was all about classification. We have improved our classification. And you look back two months, what did they do in the Knowledge Graph? They reclassified everybody. So this is the first time, and this is why it’s so huge.
Jason Barnard [00:14:20]: Not only did it triple in size, but they reclassified everything with a much higher bar to be classified as anything in particular, and then rolled it into the helpful content update two months later. And that for me, is a direct relationship. And what’s classification? When you think about people, it’s what job do I do? What’s my role in life? And they reclassified by taking away the classification of many, many entities, and the only ones that increased were writers and authors. So Google reclassified, looking explicitly for who creates the content. What does that say to you, Charles?
Charles Floate [00:15:05]: Yeah, I fully understand. I think that it’s also potentially feeding into Google’s idea of how to classify content for its own future AI, if you understand. So it can specifically go, okay, these group of people are authors on this subject. Go find every piece of content they went and when it created. And now this is our data source for this piece of expertise, right? I see it kind of, these systems are effectively stacking on top of each other and creating a new kind of form of technology or a new output, new media. And I think that’s also why Google is having these unintended consequences that you see like sites getting hit out the blue that you wouldn’t expect to get hit, and also sites also going up massively that you really wouldn’t expect to go up massively at the same time, right? And Google itself is even kind of confused and to an extent, Danny Sullivan, I think last December, asked on Twitter for the SEO community to send in examples of the helpful content update with these wrong classifications.
Charles Floate [00:16:08]: So I do think that it’s like a stacking of data with also Google’s ability to just try and define that data into the correct clusters, as refined as possible, right? And I think that will be kind of the future. And that’s my next question. How do you see ML and AI, which is the future of this technology? Kind of the future of every technology, it seems, at this stage, anyway. How do you see that moving, especially in the next year, the next two years, three years, and changing entities and Google itself?
Jason Barnard [00:16:40]: Right. Firstly, I’d like to say, I didn’t mean to sound condescending just before I was actually passing the baton to you because I knew you would have something to add to that and expand on it, and you’ve nailed it. There are a couple of extra things I might have said, but I think we can safely say Charles just nailed what does all this mean to SEO? The future is really interesting. For me, I’m super excited and I’m making the mistake yet again of trying to figure out what’s going to happen in eight years time. And I need to focus on making money today. So I’m going to try and balance those two out now. But if you look at Google, the last twelve years, I’ve been educating Google like a child. I started in 2012.
Jason Barnard [00:17:23]: So yeah, twelve years, and realized that Google is an unknowledgeable child that learns by repetition. So it doesn’t know anything about you, me, Kalicube, anybody, the audience’s person, sorry, or company, and it’s trying to figure it out. The world is incredibly messy. The web is an incredible mess and we don’t realize it. Our little ecosystem, every mention of us is incredibly messy. All you need to do is clean up your digital ecosystem, create one central hub, which is called the entity home, and link out to all the things that corroborate what you’re saying on the entity home. Link back from them to the entity home. And Google goes round in circles like this, seeing exactly the same information.
Jason Barnard [00:18:10]: And after a while, like a small child, it will say, well, I’ve seen that so often it must be true. Bingo. Bob’s your uncle. You’ve educated Google, the child. That’s really simple. Anybody can do it. The only problem, of course is, it takes you 14 hours.
Jason Barnard [00:18:25]: We’ve tried it to compile your own digital ecosystem. We built Kalicube Pro, which can do it in ten minutes. And Kalicube Pro will find 40% more than a human being, and Kalicube Pro will prioritize it according to how important it is to Google. So each source is then prioritized. You go from top to bottom of the list and you sort it out. And it takes an afternoon rather than taking it, should take two days, but people generally spend two months doing it because you get tired, you get bored, and it’s so easy just to give up. But that prioritization is really important, because if you’re spending your life trying to convince an author who’s written an article about you to correct something, but Google doesn’t care about it, you’re wasting your time. And some of the surprising…
Jason Barnard [00:19:12]: Sorry. Some of the ones that aren’t important are very surprising. And some of the ones that are surprising… Sorry. So if you take Aleyda Solis wrote an article about me, Google doesn’t care. Matt Artz wrote a page about me. Olesia Korobka wrote a page about me, Google does care. So I’m focusing on those two people and not Aleyda Solis.
Jason Barnard [00:19:34]: But logically, within the SEO community, you think Aleyda Solis is going to be more important, not the case. And so Kalicube will give you that prioritization. Now, what I’ve just described is a simplistic child. I’m desperately trying to not say stupid. Simplistic learns by route, learns by repetition. What it’s doing now is learning to learn. And that’s what happened last year as well, which is, once again, why it’s such a huge update is the engineers gave the child immense leeway and the child just went, okay, and it created three times more people than it had before. Now, a lot of them are duplicates, a lot of them are wrong, but that doesn’t matter right now because Google need to push this forward as fast as they possibly can.
Jason Barnard [00:20:22]: And they’re not worried about the duplicates. I am. I think it’s a huge problem. It’s a huge problem that’s waiting to bite us all in the, as you know… Thank you very much. It’s going to bite Google in the backside in a few years time. It’s going to bite all of us in the backside in the next few years as well. It’s a huge problem just waiting to explode in our faces.
Jason Barnard [00:20:45]: I would suggest we need to sort it out now because you don’t want to get caught out in a few years time because that would be disaster because think about what it does. It’s if it thinks there’s two different Jason Barnards but they’re actually the same one, it splits any credibility equity I’ve got between two. So the primary Jason Barnard suddenly lost some of its equity to this other pretend Jason Barnard, imaginary Jason Barnard. That’s a huge problem. I spent all of this time building links, building up EEAT credibility and I’ve just given half of it to an imaginary person and lost it. And I don’t ever know which one of them is getting the equity every time I build it. So I need to get rid of that duplicate. But then when I do and I merge them together, I reunite all of my equity and poof.
Jason Barnard [00:21:30]: It’s like I’ve just built 10,000 links in a day because I’ve managed to unify my equity. That doesn’t tell you where ML and AI are going, but now I will. Google is going to grow into adulthood. It will learn to learn and the engineers will lose control. The machine will become smarter than we are, at which point we won’t be able to teach it explicitly in the simplistic way that I’ve described. We will need to learn to manage how the machine understands and how it represents us. And that’s going to be the key for the next eight years, which I’m now figuring out. And I’ve got a good idea how we’re going to do it.
Jason Barnard [00:22:08]: But right now, if you don’t sort it out today, when that does happen, you have no hope of managing what that machine is going to be doing with your data and how it’s going to understand you and importantly how it’s going to represent you to your audience when they Google your name or when they come across you in the search results. And both happen. And I think people forget that. In order for people to see you on Google, they don’t necessarily need to search your name, although they do. They’ll see you across all of the different results where you appear and increasingly in search generative experience. As that evolves, how do you get Google search generative experience to push you in front of the audience? And that’s through understanding. If it doesn’t understand you, it can’t recommend you. A search generative experience is Google’s analysis and its opinion about the results that it’s delivering below and if you look at it that way, it’s much easier to deal with. Google’s got knowledge on the right hand side in the Knowledge Panel, SERP results, recommendations, takes the knowledge and the recommendations, brings them together, pushes them to a large language model, gives its opinion.
Charles Floate [00:23:20]: Makes sense. I think that what you just said is kind of scary to some extent, but also quite intriguing around how you kind of said the next eight years Google is going to have this technology, right? A lot of people are saying very wildly different statements, right? So as an example, right, I saw on Twitter recently, there was quite a big discussion amongst kind of Marie Haynes and Aleyda and a few other people and stuff, and they were having some kind of debate around when Google would implement it into their systems because obviously there’s a difference or a delay between when the technology is available and when it’s available for commercial implementation, right? And so they’re saying, hey, this technology is probably going to be here within a few years, right? But we’re lucky because it’s not going to be implemented commercially on an extent to be within Google’s index maybe for a decade or longer, right?
Charles Floate [00:24:23]: I don’t necessarily agree with that because Google, I don’t think that they take into consideration that the AI is also making hardware improve at such a rapid scale as well. So because the hardware is improving, two new Nvidia chips every year that are twice as powerful as the last year, which is just unbelievable. That kind of rapid technology growth, I think, will allow that commercial implementation to be sped up faster. I think it will allow Google to implement some of these chips into their data center. And Google have already kind of done it, right? They’ve ordered the most compute in the world. They’re getting all these H100 chips from Nvidia and they’re getting the most compute in the world. My question is, how many years do you think it’s kind of implemented? Is it eight years? Is it four years? Is it five years? How many years do people have until Google makers all their overlords?
Jason Barnard [00:25:17]: Yeah. Right. I think there are several things we need to look at. Number one is a lot of this has already happened and people just haven’t realized. Entities in the SERP have been around for a long time. You’ve got entity boxes now. You’ve got, if you search for the best of, instead of giving a list from a website that they’ve found online, they’re now actually listing out the entities. And we’ve got several clients who’ve come on, we need to be in best in HR software, for example. And they’re not because they’re not understood as an entity and or they’re not understood as a credible entity.
Jason Barnard [00:25:54]: So there’s entity understanding, confidence in that understanding and the credibility of the entity that has been understood. And you need all three or you’ve got no hope. So all of that is already here and people are saying, oh, we don’t need to worry yet. And they’re all being happy in the little corner, and that’s lovely, but you’re missing a huge opportunity. It isn’t because not everybody’s doing it that you shouldn’t be doing it, and it isn’t because it isn’t right now the most obvious way to appear in a Google SERP, that you shouldn’t be doing it. You should be doing it because it’s already important and it’s about to become, as you said, significantly more important, especially with SGE. SGE actually now gets data from the Knowledge Graph. For the moment, it’s only dates and places of birth, or in my case, because I’ve been doing some experiments.
Jason Barnard [00:26:41]: One of my songs that I wrote 30 years ago, it gets the date of release of that song on a specific album from the Knowledge Graph and not from the results underneath. So Google’s already experimenting with integrating the Knowledge Graph in real time into SGE. And at Kalicube we know how it does it. And it’s not the Knowledge Graph directly, it’s an intermediary system of lookup table that Bill Slawski mentioned to me several years ago that allows us at Kalicube to then, and I’m going to say manipulate. I’ve been trying not to say manipulate my whole career, manipulate Google. That’s what we do. We do it in a nice way because it’s all truthful and honest and I’m such a lovely guy, I never tell a lie. But we’re still manipulating Google’s focus, Google’s understanding, Google’s attention.
Jason Barnard [00:27:33]: And one thing that I learned, somebody said to me the other day, is the currency today is attention. If you can get attention, you can sell. If you can’t get attention, you can’t sell. Wonderful. How can you get the best attention or the most attention? It’s being presented by Google, either in the SERP or now in SGE, and SGE being all about Google’s opinion and its recommendations. You really have to be understood confidently and be seen to be the most credible by Google. So I would argue that saying, oh, we’re okay for the next ten years is a hugely foolish mistake. Over the coming years. I agree with you.
Jason Barnard [00:28:13]: The technology is advancing, and we can’t necessarily understand how fast it’s going to evolve, but it is evolution and Fabrice Canel. Who’s Mr. Bingbot? He builds Bingbot, and he’s built it since the very beginning. He’s been doing it as long as I have, but obviously, slightly more important than me.
Charles Floate [00:28:35]: No need to worry about that.
Jason Barnard [00:28:36]: Pardon me?
Charles Floate [00:28:37]: Nobody uses Bing anyway, so you don’t need to worry.
Jason Barnard [00:28:40]: But the really cool thing about Bing is I interviewed six of their team leads, the people who build the algorithms, because I realized that they’ve got nothing to lose and nothing to hide, so they just shared all their information. So I’ve learned how search engines work from the people at Bing who actually build the thing. People said to me, oh, yeah, it’s just Bing doesn’t really matter. It’s not Google, but Bing and Google function the same. It’s the same audience, it’s the same technology, it’s the same data set and the same aim. They work the same. It’s obvious. And then Gary Illyes said it a couple of years ago, and I’m very happy and pleased with myself for that one.
Jason Barnard [00:29:17]: But understanding how they work, what’s important, what isn’t important, how the flow goes through that entire system. Reading patents is great, but when you talk to the people who build it, you know what they’re actually doing and not what they could potentially be doing with the patents they’ve got, they’re not necessarily doing, but we’ve just digressed from the topic. Fabrice said to me that 2023, we would see most of the growth, the exponential, revolutionary growth in AI. And from now on, it’s evolution. It’s going to be very fast evolution, but it’s still evolution. The next big moment in tech is going to be quantum computing. So basically, we’re on a rising plateau. We just went up this huge hill, this mountain, this vitigenous mountain.
Jason Barnard [00:30:03]: We’re now on a rising plateau. And the next vitigenous mountain is quantum computing. So the day that we get into trouble as a human race is the day quantum computing becomes a practical reality in production for Google and Bing.
Charles Floate [00:30:19]: Because you can run those algorithms, even the current algorithm, you could run an efficiency of zero point whatever percentage what you’re currently running it at with using quantum computing. So it would be mind boggling how efficient that would be. And I also think that there’s some level of unlocking quantum computing utilizing the current AI technology. I think it will fasten us to the point we can unlock it even on the stage, that the AI can start simulating variations of the technology to see the output and performance and things, and then we can get it to do it. But anyway, I think, like you said, we’ve been digressing against a lot of topics here, and we could probably do that all day because there’s just so much to talk about when it comes to this technology and entities and Google and everything, right?
Jason Barnard [00:31:03]: No, 100%. I mean, I’m just really excited because for me, it’s not even once in a lifetime. It’s once in the entire human race, this is going to happen. And we’re sitting in exactly the right place at exactly the right time. And obviously, it’s scary and hugely, hugely challenging, but it’s also a unique opportunity with the most interesting thing in humanity ever, in my opinion. And sitting right in the middle of it is so much fun. So for me, this is a game, and it’s a game. I’m going to figure this out and I’m definitely going to figure it out. And that’s what I’m going to do for the next years.
Jason Barnard [00:31:42]: And in the meantime, I need to make money. Which is why are entities important today? Why should we be working on them today for our SEO of today, tomorrow and the next day? And if we come back to why entities are important, I mentioned this to Lily Ray four years ago. Without entity understanding, EEAT means nothing. And she said, oh yeah, maybe. And now she’s saying it everywhere. Because it’s obvious. If Google doesn’t understand your entity, who the person is, who wrote the content. If you think to the Killer Whale update, why did it triple the number of people in the Knowledge Graph? Because it wants to understand who the authors are.
Jason Barnard [00:32:21]: Then it can understand what is and isn’t AI generated, because it can also understand author vectors. So it can look at your work and say, this is written by Jason Barnard, or he had it written by AI and is just putting his name at the bottom. And the other thing, of course, is EEAT. What is it? It’s credibility. We call it NEEATT. We’ve added notability at the beginning and transparency at the end, thanks to Jarno van Driel. So if you look at it, it’s expertise, experience, authoritativeness and trustworthiness, plus notability in your niche. Not Wikipedia notability.
Jason Barnard [00:32:52]: We don’t care about Wikipedia notability. Google is incredibly niche. It’s more niche than you can possibly imagine, to the point at which my website is a reference for the Knowledge Graph for my music group, for my cartoon characters, because it knows that I created them. It knows who I am, it knows I created them, therefore I’m an authority. That’s niche. That’s lovely. Notability for Boowa and Kwala, who are the cartoon characters. Notability for The Barking Dogs who are my music group. Sorry.
Jason Barnard [00:33:24]: They are. I’m sorry. I’ve got authority for them. I’m getting myself mixed up there. And they are notable within their sphere, although they weren’t famous. So french alternative rock, The Barking Dogs, that’s a niche. Blue dogs and yellow koalas and cartoons who were vaguely famous online for two or three years, that’s another niche. But they’re notable enough for Google to take notice and I’m authoritative enough to be able to educate Google about them directly.
Jason Barnard [00:33:53]: Where it doesn’t even double check what I’m saying. I can get it to spit out anything because I’m so authoritative about them. Google spent so long, 25 years, I can’t believe it. Spent 25 years relying on links. Links are popularity. You can possibly get…
Charles Floate [00:34:09]: High popularity?
Jason Barnard [00:34:11]: Well, then you can look at it and say, well, it is authority. I mean, I use the example of The Sun newspaper, it’s popular, but you wouldn’t believe what it says. So the popularity and the authority are two very different things. So EEAT then starts to measure all of the different aspects that aren’t just popularity. I mean, I’m simplifying links and I don’t want to dislinks, they are very important. But what you then have is Google had or has links and people are saying they’re still important. They are still important, but then you’ve got all the rest of it, which Google can only apply if it’s understood the entity. And in the quality raters guidelines, they talk about the website owner, not the website.
Jason Barnard [00:34:51]: They want to understand who owns the website. They can understand that. Then they can start applying all of the credibility signals, such as awards, time in market, references to them across the Internet without links, employee feedback. And if they can then identify the person who wrote the content, which is what the author update from Killer Whale from last year was, then they can look at qualifications, other articles, what I’ve been writing about for the last 25 years, peer approval, social buzz. All of these things that then add on top of the links. So it’s not links aren’t important. They are. But look at all that stuff you’re missing if your entity isn’t understood.
Charles Floate [00:35:40]: Yeah, it’s sort of like links are the base layer and you’re building your entity on top of them, basically. And the idea is that you’re trying to get as authoritative or as notable. I’m trying to think of the correct word as a high-up-there positionally on those links as possible. So, for example, as we already said earlier, some links have higher priority within Google than others and you’re just wanting to try and get that entity information on that priority list as high up as possible, right? And currently, the way to do that, seemingly normally, especially with a new site anyway, is probably with links. But going forward, it’s going to be a combination of a lot of different things, right? And I think that will be the future of everything.
Charles Floate [00:36:25]: It’s just when is that future going to become reality? And what will happen? To your point earlier, as always, you talked about manipulation. A lot of people have said, what happens if I can no longer manipulate Google? What happens if ChatGPT is the future? And I said, well, can you not manipulate ChatGPT? Are you not doing the exact same thing where it’s reverse engineering something, positioning something in the correct place, and then hoping that output is in your favor? And it’s the exact same thing that we’ve been doing with this system of SEO and Google, but we’re just going to have to adapt to these new systems.
Jason Barnard [00:37:04]: And you’re going to have to get used to the Google dance. So people who never went through the Google dance are going to have to learn how to deal with the Google dance. And it’s hugely frustrating, I can tell you. An interesting thing about ChatGPT is I was at a conference the other week and I was saying, Google search results, look at how well Google understands me, look me up on Google and you’ll see I’ve got this amazing Knowledge Panel, amazing results. And the person said, I don’t care about Google, I use ChatGPT. I said, we’ll try it on ChatGPT and ChatGPT says Jason Barnard is a world-renowned expert in Knowledge Panels and Brand SERPs. He’s been in SEO for 25 years and he created Kalicube in 2015, I think. But basically, he says he’s a world-renowned expert.
Jason Barnard [00:37:48]: Then we tried somebody else’s name and ChatGPT had to look it up on Bing and came back with the answer, which was simply a summary of the search results. And that doesn’t sound like such a problem because it got a good answer. But what was happening there is you ask about me and the child, ChatGPT, can recite from its own brain without looking it up in a dictionary or an encyclopedia, exactly who I am and give an opinion. World renowned. Whereas with the other person, it had to go and look in the encyclopedia, which is Bing, look at the current results and then summarize. No opinion, so no recommendation. There are two things.
Jason Barnard [00:38:32]: Number one, I didn’t get… The other person, sorry, didn’t get the recommendation. Number two is those search results can change from day to day. So the result that it’s giving and the answer it’s giving is going to be different tomorrow than it was today. So the LLM behind ChatGPT was trained in April 2023 was its last data set. So I was already in there in April 2023, but to get there, I had to prepare in April 2022, let’s say. So doing this over time is going to become increasingly important that Google dance is going to be a huge problem for people who don’t understand how it works.
Charles Floate [00:39:07]: Yeah, I’ve managed to get myself in ChatGPT as well about a year and a half ago when I was using it, and I was really frustrated that whenever I put who is Charles Floate, it came up with like, I don’t know who that person is. I was like, I’m going to make you know who that person is. And lo and behold, the April 2023 training date to it did. So, yeah, the way I kind of reverse engineered it was I looked up other people that it had found within our niche and that it did recognize, and then I just reverse engineered where those sources came from. There’s likely source sites went and bought articles on those websites about interviewing me or about me, and then suddenly, lo and behold, it’s within ChatGPT’s training data. So it works out pretty well. And like I said, it’s the same kind of thing with, what have they been doing with SEO, with ChatGPT optimization, it’s just reverse engineering and then trying to manipulate the output. And I think that is inherently what SEO is at its core.
Jason Barnard [00:40:02]: Yeah, that’s a good quote. And 100% of what you just said is something that we do at Kalicube, which is look at the competition. What we do is take, we call them entity equivalents. So it’s same entity type, same geo region, same industry and we put them into the Kalicube Pro system. So we try to get 50 to 70 entity equivalents, and then what we can do is template. And we can then template, not just which are the sources of your hated rivals, as it were, but what does Google look for as a template for this cohort, for this group. And that means you can immediately say, well, this is going to be better than that.
Jason Barnard [00:40:44]: Even though that person is in this particular source, the industry or the cohort or the market in general is going for this. That’s where Google’s looking. And that’s the key. It’s not just where they are, it’s where is Google paying attention. And that’s what technology does for you. It does at scale what you would do by hand, much faster, much more accurately and with prioritization. And that is hugely powerful. But we also do what we call competitor envy analysis.
Jason Barnard [00:41:12]: So we take the hated rival, we analyze them, and that’s where you poke the client is, say, look at what they’ve got that you don’t. And that’s the kind of competitor envy thing that we can get them going with. But the actual real power is the cohort analysis, the market analysis of 50 to 70 equivalents. But we don’t talk about that too much to the clients because the competitor envy works much better to get them moving.
Charles Floate [00:41:36]: Yeah, it turns out ego is a pretty good selling element when you start talking to them. I really appreciate you coming on, Jason. Is there anywhere anyone can find you online?
Jason Barnard [00:41:45]: Yeah, Google my name, or look me up on ChatGPT. Ask ChatGPT, who is Jason Barnard? And it will tell you he’s a world-renowned expert in Brand SERPs and Knowledge Panels and has been in SEO for 25 years. He created Kalicube in 2015, and next year maybe I can get it to add, he’s a genius. That’s a joke, by the way.
Charles Floate [00:42:07]: No, get it to say the full iron man thing, you know, playboy, billionaire, philanthropist, the whole lot, right? Yeah. I need to get some extra names on my entity, I think going forward with ChatGPT as well.
Jason Barnard [00:42:20]: I’ll tell you one last thing about classification, is that Google, for the last six months, has been explicitly looking for writers and authors. So if you’re an SEO who has now been classified as an author or writer, go with it. Go with the flow, because as you said earlier on, Charles, you’re a writer who specializes in SEO, which is much more powerful in terms of content creation than an SEO who happens to be a writer sometimes. So that’s point number one. Point number two is, once that happened, I was a writer in Google’s brain, and I have been for a while. We decided, as always at Kalicube, to experiment, and we decided to change me from a writer to an entrepreneur, which is very much going against Google’s current flow and current focus. And we did it. It took us two months.
Jason Barnard [00:43:07]: So now, Jason Barnard, entrepreneur in the Knowledge Panel, why did we do that? Because I don’t really care about Google search results, ironically, even though we spend a lot of time on Google. We use Google to learn what we need to do for our clients because what it doesn’t understand, we can teach it. We can teach it by improving the digital ecosystem. By improving the digital ecosystem. We’ve improved the funnel and all the touch points for the potential client. Google then reflects it in the search results for their name. Sorry, we’re learning from it and then we’re teaching it. And we’re teaching it by what we can call it implicit semantics. Being about entrepreneur is why is that important? Because every single project I get involved in now benefits from my NEEATT.
Jason Barnard [00:43:57]: So next company I create, Google, will say, well, it’s created by Jason Barnard, who’s a super duper entrepreneur and has been for 30 years. That’s going to get some extra credibility just from being associated with Jason Barnard. That’s my next project.
Charles Floate [00:44:12]: Well, I look forward to seeing it come to fruition and I appreciate you coming on once more. That’s been awesome. I think some people are going to be completely mind blown by this interview and they’re going to walk away with just so much information to take away, so much entity research to do. I really appreciate you having you come on as well and I’ll see you guys the next one. Peace.
