In the latest episode of the Standard Deviation Podcast, Simo and I sit down with Michael King, founder of iPullRank, to talk about some big topics in SEO and marketing. Michael shares insights from his 10 years of running an agency, explaining how he's evolved from hiring freelancers to building a structured team with solid processes. He emphasizes the importance of balancing creativity with scalability and how being adaptable has helped him stay on top. A significant highlight is Michael's deep dive into the Google algorithm leak. He breaks down what we've learned, mainly how Google uses click data, user behavior, and UX's role in search rankings. One key takeaway: SEO and UX teams must work more closely, as user satisfaction directly impacts rankings. We also discuss branding in the age of AI, touching on how companies can influence results in generative AI platforms like ChatGPT by focusing on authoritative content, proper markup, and structured data, which plays a key role in AI-driven search results. If you're into the technical side of SEO or just want tips for staying ahead in search and marketing, this episode is packed with valuable insights.
Latest content from Juliana & Simo
Also mentioned in the Episode
Connect with Michael King
Standard Deviation Podcast with Juliana Jackson, Simo Ahava and Michael King,
===
[00:00:00] Now we're back. Oh my god. It's been so long. So long. Months and months, at least weeks and weeks, if not days and days. Days and days, months and months, weeks and weeks. But true to Finnish cultural tradition, I was, like, we vacationed. Like big time. Like we, we did all the holiday trips and we did all the music festivals and that's awesome.
Took in a lot of sun outside, went swimming all the time. And that sounds good. That sounds good. So your summer was very fun. It was because it's so my son started school. in August. And, um, so this was the last kind of time when we can just travel as much as we like without having to worry about like absences from school.
So we, we, we made the most of it. And, um, we did a long 10 day trip in London, uh, which was a lot of fun, uh, around my birthday, but, um, our birthday, you mean? Yeah. [00:01:00] But, um, 10 days is quite a bit of time in a hotel room for a, for a family. Yeah, I know. But it was a lot of fun. Like London is a nice city and, and I, you know, I visited there so many times and I've, I've been there as a kid as well, um, because we lived quite nearby.
But, um, it was interesting to be there for the first time as a parent with kids because it, it is a, it's, it's, it's an incredibly big city, like such an incredibly big city. So trying to figure out what to do with kids. Yeah. Uh, was, was. It was a fun challenge. I bet. I bet. Oh man. But you have been, you know, like it's good that you have been somewhere this summer.
I haven't been anywhere. I've been at home again, celebration. It's been 45 degrees for a month. It was unbearable. And, um, at that time I was dealing with some changes because I'm leaving. Okay. So I'm leaving Romania. Hey, we'll talk about that later. [00:02:00] But because, you know, I. Stayed in this rent for so long. I want, obviously wanted to leave this house in a nice place for my landlord.
Cause you have to give it as you got it, or at least close to how you got it. So I had to take a few nights at the hotel because you know, I did a lot of stuff in the house and it was smelling, you know, like all the substances. It was interesting to stay with the kids at the hotel in Bucharest. Um, and actually my birthday I was, uh, at the Marriott and, um, they made me this nice little birthday cheesecake.
Oh, that's so cool. I cried because that was the only thing I was able to do my birthday besides running after these kids. They uh, it was nice It was cute. Um, so yeah, and uh, yeah, like simo i'm moving to uh, moving to malmo. I'm actually going there. Uh, Uh next week and uh, I think if you're listening to this podcast By the time you're listening that I am probably in malmo.
Uh if I manage to release it Um, if not, it doesn't matter. But yeah, I'm [00:03:00] definitely, uh, I'm going to be there. I'm moving to Sweden. That's it. I'm disrooting my Romanian roots, but, uh, yeah, just excited. Just, it's because of you guys, I told you, like I have zero social life in Romania for several reasons. And people ask me all this all the time.
. It's just that I always felt like I didn't fit in here for some reason. So the first time when I actually went out of the country was when I was 26. And the first country I ever visited was Finland, was Helsinki.
And let me tell you guys, it's a huge difference between Bucharest and Helsinki. Because Bucharest is like, very crowded, because you've been here, Simo, so you know. It's so crowded, noisy, there's so much stuff going on. And then you go to Helsinki, where everybody's civilized. And you have one person benches, which is the best thing.
Ever the best invention, in my opinion. Why do you want some random person sitting on your own? I know, I know, right? It's so fucking weird. And if you go to the bus, [00:04:00] it's like all the, all the two seaters people just take, everyone fills in only the one seat of those two. Yeah. And then they put their bags on the other seat.
Exactly. Yeah. Like that's, and I felt quite good because like I do, and I know I seem like I'm very, very extroverted, but whoever knows me knows that I'm the total opposite. And I don't really like proximity that much. I mean, I like it sometimes, but with limits. So that's why I, when I went to Finland, I was like, okay, like, and for some reason Finnish people like me, I can't, I can't make it up yet because I'm like so anti Finnish.
Like I'm everything, you know, like everything that I represent is anti Finnish, you know, like I'm loud. I'm obnoxious. I'm everything. For some reason I get along. You see, you, you tolerated me for two years. Yeah. I did. No, no. Yeah. You're, you're, you're like an honorary Finnish citizen now. Thank you. And that's what the Measure Chem Helsinki people are saying, but Carlos tolerates me, Lina tolerates me.
And uh, it was kind of similar with Denmark and Sweden as well. [00:05:00] So yeah, I decided to move there. It's been a process for my family, as you can imagine, but, um, you can come with Mari and the kids and visit me cause I'm trying to get like this beach apartment. I've been telling you, it's a bit expensive for sure, but like, come on, it's You know, like, come on, you want to, you want to do it in a nice way.
So yeah, that's it. Uh, excited about it. And that's kind of been my summer kind of like putting this together and uh, figuring shit out. You, you, you haven't vacationed all summer. I know you've been busy cause every week you release something. You were talking about something. I mean, we had the fun with the cookie deprecation.
That wasn't no deprecation. And I feel like your meme that you always share on Twitter every year, Oh yeah, yeah, the three third party cookie deprecation. Yeah, well it's old news by now, but yeah, so as it turns out, Google isn't deprecating them and it's like, um, yeah, it's yeah. So obviously like when I vacation, I do follow the news and I, um, try to stay on top of things, [00:06:00] but yeah, I did, I did vacation.
Like it's, it's, you know, one of my life goals has always been that I can just drop the pen whenever I want. Yeah. Or get into a, get into a position at work where I can just drop things, uh, when I want and I, well, I'm not when I want, because obviously there are some deadlines to meet, but um, I'm definitely getting closer to that place and it's, I found it all the more important now that we have kids to be able to do that.
So, but you've been looking at this for a decade, like people think, Oh, you know, it just came out. No, like it's. Well, yeah, yeah, yeah. It is. Yeah. That, that is what, that it is what it is. But, um, and it obviously it's required a lot of work to. To get there. But, um, and also because the summary summary in Finland is really quite short, like you get two months tops of, of good weather.
And, and now we have this problem. So finish lakes and, and the Baltic sea in general, well, the Baltic sea is really in a really bad condition. Like it's, it's like de oxygenized there's, [00:07:00] there's very little oxygen left. And, and it's, it's like just impacts everything, the, the, the flora and the fauna like, but now also like Over the last two, 20 years or something, the Finnish lakes, many of them, especially the ones that are popular swimming spots have this thing called cyanobacteria, which is, which is classified as, or it's the, like the common term for it is, is, um, like seaweed or it's lake weed or whatever you want to call it, but it's not actually, it looks like it.
So it's this kind of floating green stuff on the lake. Um, but it's actually a bacteria. Algae, right? How do you pronounce it? It's, the algae, the word I was looking for. Yeah, but it's, it's not algae, it's bacteria, but it looks like algae. So that's why it has, that's why it's called algae and Finnish. It's like um, blue algae.
Um, but it's a bacteria. So long story short, long story short, there are different. Types of it and some of them can be really, uh, like hazardous to you. You get, you get [00:08:00] nausea and you might get like little kids might be sent to the hospital if they ingest that water. And so it means that many of the swimming spots are like, they're only open for like the first couple of weeks of warm weather and then they have to close down because of cyanobacteria.
So that's the, so that's a bummer because we just moved, um, a couple of kilometers away from our old house to closer to a couple of really nice lakes. So we're, we're just a walking distance away from those now. Um, and they have really nice, um, swimming, swimming spots, but the other lake, especially the one with the nicest spots gets a lot of that bacteria.
So we only managed to swim there for the last first two, three weeks. It really sucks. Not just because we can't swim like that's a luxury. We can definitely give up. It's not like the end of the world, but the fact that we have this kind of pollution, um, and this deoxygenation stuff. And it's, yeah, it's just depressing because we have some of the clearest lakes in the world and the clearest freshwater in the world.
And then there's this [00:09:00] thing that isn't, isn't really pollution, but it's kind of an, a by product thereof and just, yeah, it really sucks. It's kind of like, it's dying, you know, like the sea is dying. Yeah, it's kind of similar to that, but there are some, there are some really interesting research initiatives where people are building technologies to harvest cyanobacteria for like industrial purposes.
Because if there's, if there's not a, if there's not like a monetization involved, nobody's going to do anything about it, but now they've figured out a way to maybe like harvest that stuff. And start turning it into, I don't know what is, is it, is it going to go into like, um, feed or, or is it going to be turning into like, um, you know, plastics or something?
I don't know, but there are initiatives for that. So that's, that's something to follow at least. I had no idea about that. That's crazy. Cause we have the same here. Cause I'm at the black sea, but it's just that black sea. It's so, uh. Cover constantly in transportation between Europe and Asia, because it's the only sea that connects Europe with Asia [00:10:00] and it's constantly like the Romanian ports there.
It's so much shit. Cause everything that comes from Asia, like there's so many sewage and stuff that happens in the sea. So I, I, I, I grew up next to the beach and this year I didn't go because it was like swarming in Algae. And, um, we used to have like this virgin beaches here in Romania until people figured them out.
And now everybody's there throwing random shit, unfortunately. And I'm ashamed of this. Like we're not the most. Yeah. Kept people when we go outside of our homes, it's funny because we do this when we travel. Like, I've seen this, I've seen Romanians outside of Romania and they're respecting rules, robbing citizens, you never see them throw, but the moment you hit the Otopeni airport.
They're like fuck this shit, you know, like typical this is like so and i'm, sorry And if you're romanian and you're listening this and you feel offended
So what she said [00:11:00] exactly so please but anyway, um now i'm glad that you had a nice summer I know you're also being busy. I just want to before we uh before we get our guest in just want to touch a bit on the Uh, new release that you did with PV pro for the server side template. Like I wanted just to see, you know, like how, how did that work?
What does it do? Kind of like the elevator pitch. So, so ironically, I, yeah, I did vacation, but I also did write a couple of templates, but that's coding and it doesn't count. It's like a side quest. Yeah, that doesn't, that doesn't count at all. Like it's, it's like a vacation and still write code. Like I know, wait, writing code is vacation for me.
So I, yeah, I worked with the pivot protein. There are, there are a lovely bunch of people and, and I got the chance to help them out writing some templates for server side GTM. So what we basically built is a pipeline, um, with two options. One is to use the PIVX Pro JavaScript tracker, the one that you would deploy on the site through PIVX, uh, PIVX Pros, uh, Tag Manager [00:12:00] or, or with the script snippet itself.
Um, and then send the information to ServiceLite GTM instead of directly to your PIVX Pro instance. And then in ServiceLite GTM, you know, you can do all the basic stuff like enrich, validate, lock, cut stuff out, fire other vendors tags with it. So you have this pipeline of taking control of that, of that, um, vendor data.
And the other thing we built was, uh, part of the tag template that we built for server side is support for taking another vendor stream like GA4 or amplitude or mixed panel or snow plow. And then use that information, map it to pivot pros format and collect it to pivot pro. So the two use cases here being the one first one I already described is kind of taking control.
So you get to choose what pivot pro actually receives. You don't have to trust that their JavaScript library has your best interests at heart. Um, and the other option is if you want to try PIVX Pro and you've already extensively tagged your site with [00:13:00] GA4 and you don't want to go through the motions of tagging it again, you can just use the GA4 stream, uh, with some small modifications.
Uh, you can then kind of map it and transform it into a PIVX Pro data stream and collect that to PIVX Pro. So it's been a lot of fun. Um, PIVX Pro has a number of things going for it and one of them is that it's, uh, so similar to the paradigm that Google analytics. represents like this discrete model of events and sessions and users, uh, more akin to universal analytics than GA4 though.
Um, so it's, it's kind of an easy transformation to make. You don't have to worry about like backend processing stuff. Much of what GA4 does happens in the browser. Much of what PIVX Pro does happens in the browser. So you don't have to kind of worry about, okay, I managed to do this mapping, but am I ignoring all these events that happen in the server?
Well, GA4 doesn't really have them apart from session start and first visit, which have, which have special logic, but yeah. So it's, it's, it's, it was a really, really fun project. And I think that the community has. [00:14:00] Really embraced it. And I'm really looking forward to just everybody was excited that day.
Like I couldn't open LinkedIn and nobody was talking about anything else. It was really, we're definitely going to talk about this more this year. So I have a, um, you have a webinar, right? We have a webinar with Pyrrhic in September, uh, late September at some point. Um, that will be definitely. Maybe next week already or the week out before the end of August will be will be promoting it quite a bit and then, um, I have a couple of talks coming up and I have no idea what I'm going to talk about.
So I'm thinking this might be one of them. So measure summit, for example, I thought that maybe I could talk about pay with pro. That would be a nice change. I would like to see that. That would be so cool. We're not talking about PWA Pro because I'm not an expert. I'm talking about the pipeline stuff.
Maybe in general about using server side tagging as a kind of a consolidation. I had somebody yesterday after they saw the video that I posted that we did sometimes. [00:15:00] In a random time in a random time of our life, she said basically like, and I'm reading, um, I understand the concept, but I'm not sure how would that benefit marketing the marketing in the end, since we're using different advertising and measurement tools to deliver our ads.
And get the data wherever the ad has been served. Won't that be a limitation for us and therefore advertisers? And I mean, I did tell her that instead of just giving everything to vendors, you know, client side, you choose what data gets processed, but you're not kind of like, um, entirely blocking, um, the data flow to the advertiser, but just kind of like you're creating a control layer.
Right. So am I, am I wrong? So, so the question was about server side tagging in general. Yeah, for sure. Like advertising as, uh, as much as it works without third party cookies, it works just as before, like when third party cookies come into the mix, it gets a bit more complicated, but Google has a solution for that.
It's not a, it's not a fun solution or a nice solution in any way, [00:16:00] but it does. But, but for like first party advertising, like, uh, you know, correlating clicks with ad impressions, there's of course, it works just, just as before you, you just get to choose the appropriate way of dispatching that data. And you can do piecemeal setups as well.
You can do ads through the browser like before, and then just focus on analytics through the server. You don't have to move everything 100%. So it's just, you're more in control. Yeah, exactly. More in control. Like in my, my, my opinion, um, taking control of that ad stream is way more important than taking control of the analytics stream because ads is historically more, uh, inclined to collecting far more than they should.
Um, so taking control of that would be, would definitely be a bonus, but it's also a more difficult pitch in the organization because you, you control. Unfortunately, control taking control with ads means that you actually reduce the amount of data that is collected. And many organizations are not happy with not getting, um, getting [00:17:00] that granularity that they think they need, which I could also argue that they don't actually need.
But they don't need it. Exactly. But it's, it's not my, it's not my, like, I, I've, I'm done with that fight, uh, companies can do whatever the hell they want. And I just, I'm just going to keep writing solutions. And every now that I'm going to stand on that soapbox and, and, um, tell them how the world should look like, but I'm luckily not in a place where I actually get paid to try to insert my worldview into other people's minds.
So I'm very happy that I'm not there anymore. I'm trying to myself, but sometimes it's hard. Yeah. So on another note, GTM tools, don't don't. Oh, yes. Let's not forget about that. We do a very quick Yeah, very quick pitch about that. Very quick update. Yeah. So GTM tools is a, is a tool set that I originally released, uh, a bit over 10 years ago and, uh, it's, it's a way to manage Google Tech Manager containers and it stayed largely untouched for the past 10 years, apart from a couple of features that I added every now and then.
Um, it's got like, you know, thousands upon thousands of lines [00:18:00] of backend Python code and, and at least the same amount of frontend JavaScript. It's not a beautiful code base in any shape or size. But you have those pretty nodes and stuff in her. Yeah, for sure. I love it. The, yeah, the visualization thing was a lot, a lot of fun.
That was that maybe the most pretty, but the, yeah, so I got, I got to a point where, uh, it needed a lot of work to get up to date and I didn't want to put in the work. So I decided to, um, shut down the project. And I, I kind of said that if somebody wants to buy this domain, then contact me because GTN tools.
com apparently was quite a valuable domain. And I got a couple of contacts and nothing really. Inspired me, but then the lovely guys from stape, uh, who are, you know, the, one of the number, just number one companies for pretty much anything these days, I think they put out so much content and just so such cool solutions, especially for server side stuff.
They offered to take over the domain and to re write the code base. And to re release the [00:19:00] tools in large, their original form, like look and feel, but of course, with a roadmap for updates and off, like I was so happy. Not because, not because I got like any, any, not because I got rich because I did it. Yeah.
But it took over this project that had really been a, uh, an important one in my own development work. So gtntools. com is where you can go. It's completely free. And hopefully we'll stay free. Um, you need to log in with your Google credentials that has access to a Google Tag Manager setup. And then you can do all sorts of things like copy tags, uh, update workspaces, do bulk stuff, save items.
And if you have more than 50 tags, you should wonder. Why? Yeah. Yeah. You get to, yeah, you get to remove, uh, unlinked tags easily and unlinked variables. So it's just a lovely way of, of maintaining. Yeah. So that was GTM tools. Shouts to the guys at STAPE and Dennis, if you're hearing this, I still would have loved one of those turquoise hoodies.
I [00:20:00] don't have the hoodies. Also, we got to have, we got to have Dennis and somebody let's have the marketing team over. Yeah. Yes. Let's have Dennis. And maybe Dennis will send us one of those turquoise hoodies that I had to fight with somebody for it. Major camp Copenhagen and didn't get it. Yeah. Yeah. We, we have them.
Of course. I'm so jealous. Yeah. I don't have them. I know we have all the hoodies now. Dennis. Cool. Well, uh, this podcast, before we, uh, move on and invite, we have Michael King guys. We have one of my absolute best friends, one of my absolute best friends from the industry. Michael King is an amazing person. I can't wait to chat with him about all the research that he's been doing.
Yeah, and he has a book coming up too, so really excited, but he's a hip hop guru. I'm going to try not to talk about hip hop, but let's see, you're going to fail. I can, I can look into the face, I can look into the future and I know that you're going to fail at that. I'm going to make a prediction that you're going to, you're going to start talking about hip hop.
And I'm going to be there like just nodding my head and thinking about when are we going to talk about [00:21:00] death metal? We'll see what happens. I'm going to bet you 1000 euros. Yeah. Okay, let's see what happens, but let's not forget this podcast is sponsored by team seymour. com. Wait, I'm going to point it.
That is, there we go. That's team seymour. com and we have Simo from Seymour here. Hello everybody. Hi, I'm Simo from Seymour and Uh, we do online courses in technical marketing. So go to teamseaman. com and check them out. Use the coupon code deviate. That's D E V I A T E to get 10 percent off any individual course purchase does not apply to bundles or other campaign products and go check out our technical marketing handbook, which is a free resource that you should absolutely get your team, your colleagues, your friends, and your family to read.
Especially your friends or your family so that they finally understand what you do for a living. So go to handbook. teamsimr. com and once you complete the handbook, so read through all the content, you'll get a nice certificate. And if you post that on LinkedIn and tag Simr, [00:22:00] you'll quickly find that Marie and I who wrote the handbook will be jumping there and congratulating you in the comments.
I've seen it. I've seen it. Yeah, we do it all the time. It's a lot of fun. Awesome. Well, thank you, Seymour from Seymour. And now. Thank you. Lovely podcast you have. I try. Cool. We'll have Michael in. Yes.
Yeah. Okay. Oh, yeah. Here we are with my king. Good morning. How's it? How's things going with you? My friend. First of all, thank you so much for being, let me be here. Um, yeah, things are crazy. Things are really crazy. It's been a very wild last three months for me, uh, since the algorithm leak. And you know, I was in back to back meetings every day for nine weeks after that.
Cause so many people were reaching out, wanting to work with us. So many people wanting me to do things like this. And uh, then also client work too. [00:23:00] So this is like, I think the second week where I've had like time to just chill a little bit. Yeah. So you kind of introduced a number of things I want to talk, we want to talk with you with you today.
But one thing we want to start with is that So it's, it's kind of an important couple of days being for you because your, your, your, your brainchild, your agency, I pull rank just turned 10 years. I got the feeling like you, I think you dropped the ball before the weekend saying that, you know, we're going to, something's coming up and you get the feeling you've got a lot of celebrations planned for this.
So can you say something in advance? What's going to come up? I can't really speak on it yet, but we're, we're going to do some great things. I mean, you know, we, we historically have. Celebrated very big and I pull rank like boat parties, like, you know, parties and houses and Hampton's like we've done things like that historically, or like rooftop parties at clubs and in the city.
But [00:24:00] since the pandemic, we haven't really done much and, you know, the team is mostly remote. And so we want to get everybody together and do something really big this time, especially because it's 10 years. You know, 10 years is like a milestone that so many businesses don't reach. So I'm excited to be here and I'm also excited to see like what comes next.
Cause you know, I, I think we did a lot of things by accident that worked out. I think now we can start being more intentional about what we want to achieve. Well, I think you're selling yourself short. Like if you get here 10 years and you've grown into this huge, like, well, you've grown into a big business and you have amazing big clients and you still managed to do that, like technological stuff that you do all the research and all the articles.
Um, so. Looking back on those 10 years, like how have you changed as a founder? Like when you started, was it just you? I think it was just you, right? Yeah, it was, it was just me at first. Um, and I was just using my friends as like contractors. And then [00:25:00] that next year, I was like, okay, I need some help. It was like, I can't, I can't like, when you just use contractors, they do exactly what you tell them to do.
They're running a business. People have to like pick up the nuances on things and help you out proactively. And so I needed to shift more towards that. Um, but really I just didn't know what I was doing at first, right? Like I'm not a inherently process oriented person. I'm someone who's like, cool, what's the problem?
Let's figure it out. You know? And like, I might use some parts of what I did before, but I'm usually looking for a new way to do it better. And that doesn't scale. Like you've got to come up with. You know, process documentation, training and so on. So the first few years I was just like, cool, let's just hire smart people and be like, we have these problems.
Let's figure out how to fix it. And I have no idea how we were getting anything done. Like, I think back to our project management solution, we, we went from like, you know, Trello to Asana to teamwork for a number of years. And when I think back, they were [00:26:00] just tasked with like names and no instructions on what we were doing.
So like, how did anyone know what to do? And. What ultimately ended up happening was like people would get really frustrated because there was just no process. And yeah, like I had to learn the hard way that like everyone isn't like me. You know, I assume that if I can do it, anyone can do it. That's just not true.
Um, and so you've got to have frameworks in place that work for other people because they're going to be people that you hire that are just like, I just want to know exactly how to do my job. And I'm going to come in at nine, I'm going to leave at five. And that's that. And frankly, that's the most people in the world.
Right. And they're going to have the, the amazing people who are like, yeah, I want to figure out how to make this mind and how to do this better. And all the things that I was just describing before. And those are the people that are going to like shift your business in the right ways. But if the lower end of the staff, Can't do their part.
That's just going to frustrate the higher end of the staff. [00:27:00] So you really got to put those frameworks in place. And I just wasn't something that I was taught before. Like, I don't think I ever worked anywhere that we had really specific processes. And if they did, they just never gave them to me. Cause I was So, um, I just never had that on my mind and that was like the thing I had to learn the hard way to be able to get to where we are.
So, by the way, do you know what, what other 10 year anniversary it is this year? Your blog? No, no. It's 10 years since we first met at the marketing festival in Brno. Oh, wow. Yeah, it was, yeah, it was, uh, fun because I, my talk was just before Mike's and, um, I like, of course Mike had already published a lot at the time and I've, I've read a bunch of his stuff.
And I remember when, when it was Mike's turn to come on the stage and, uh, you know, I was with my wife, uh, mother was there as well. And we were just on this, on the sidelines. I was, we were about to go grab a cup, grab a cup of coffee. And, and then Mike's your talk was, I think it was like, [00:28:00] What was it like 30 API's in an hour or something like that?
So you just drop different SEO or not just SEO, but different API's and how to use them. Now it's just like floor. Like, can you do a talk like that? Is that allowed? Like, can you do such a technical talk? Because I was the under impression that we're not like a thing. So people are expecting me to be like super, um, telling a story with numbers and all that bullshit.
And if I had known that we could do like a super technical What do you mean bullshit? Oh come on, storytelling is like the oldest trick in the book. It's, it's a, it's deception. So anyway, but yeah, but that was a, that was a fun trip. It was, it was a cool, cool event and kind of make, make, make a lot of the following events after that look bad because it was so well organized.
Like everything was so, so well catered to. Yeah. Yeah. He, he went to all the events, all the big ones and took notes and was like, here's how I'm going to make this one better. That's how I met him. It might've been like Maz Khan or one of those events. [00:29:00] He's like, I'm here. Cause I want to see how they do this.
Cause I want to do this thing called marketing festival in the Czech Republic. And he and I connected because, you know, In my rap career, I spent a lot of time in the Czech Republic. Like I spent like, I lived there for a summer. And so, you know, I was like, Oh, you're doing it in this place. And so it is so forth.
So like we connected and then I ended up doing it. Um, but yeah, marketing festival is, is by far like one of the best shows I've ever been involved in. And. You know, I really love how they do like a real party at the end and then they also have to perform at that too. So I had a lot of great time doing that one.
I would love to do that. And they always have a different, like every year has a different vibe, which is kind of different from these other events that are this because they have the same speakers every year and all the kinds of the same themes. I think marketing festival really, really manages to push the, so this actually brings me to one more thing because before I'm going to let some questions in as well, because I'm, I just, I just.
I miss you, man. It's been a while. [00:30:00] I want to say something. I want to say something because I was already a fan of you before I met you as well. Like I was all over your blog for all sorts of reasons. And so I met you in person, I was like, yo, this dude is cool too. Okay. Guess what's up? You know what I mean?
Yeah. We found common ground really fast. But that's one thing that I want to talk to you about because you also mentioned your rap career. And so you've been doing a lot of stuff like, and you have a family as well. You have two, two wonderful kids. And, and so. How do you find this is the thing that I've been fighting with for a long time how do you find today like 10 years into agency life and God knows how long it into into music and and and being a you know technology research how do you find the time to be on the cutting edge like how do you do you actively set time aside do you spend all night reading stuff.
No, everyone asks me this as though, like I've solved it. I have not, right. Like my whole life is basically triage. It's like, what is the most important thing to do right now? And, um, you know, the, the cutting [00:31:00] edge stuff is just a function of my curiosity. Right. Like when something pops up, I'm like, I want to know everything about that.
And so it'll be like, cool. Well, I have this other thing to do. I'm gonna put that to the side today. Cause I want to learn how retrieval augmented generation works. You know what I mean? And also with me, um, Finishing my book. I've been so deep in research anyway. And so I just have a much better understanding of things.
And it's easier for me to just add on. And you know, if I'm sitting around like on archive reading white papers anyway, and then I find like, you know, here's a new paradigm. I'm like, Oh, let me learn how to do that. And it's just, It just fits in naturally with what I'm already doing, but it's not because I have, you know, mastered time management.
It's nothing like that. It's just because I, I just go after what I'm passionate about at that given moment. And you're kind of lucky that many of those things that you're passionate about are also very central to your professional career. So you're kind of converging those. That's, that's pretty cool.
What I will say to that, [00:32:00] because of your rap career, a lot of people don't know, but my husband, actually, that's how I met him. I was managing him. He's from us and we, yeah. So we gave up on, I gave up on my career just to do music for two years. And I will say to you that breaking into the U S hip hop is one of the worst shits.
That I've ever experienced in my life. Everybody wants money. Everybody messages you like, Hey, let's work. Let's call up a hundred percent of the time. It's bullshit. And, uh, I produced a lot of albums as an executive producer for UK hip hop. And so I will say that if you don't have the job. To fuel that is very hard to make it in music.
And I seen, and I know so many talented people like you're from New York, right? I'm from Philly. I live in Philly. Yeah. Okay. So in New York, there's like a bunch of dudes that used to be in Sean Price's group, it's like very, very hard to make it in music. So like, I did listen to your stuff. I saw it on Twitter. I think it's [00:33:00] cool, but I don't think like, if you didn't have the job and like the company, it would be very hard to kind of like invest. Cause me and my husband got so fed up that he doesn't even write anymore.
He's like, we're just watching, you know, battle reps. So we see, and then we talk shit every time the song draws or like, it's like, it's, it's like we, we've been burned, like, we've been really, really burned. So if you still have like the power and creativity to be able to pick up the pen, that's really cool.
Cause it's hard. Yeah. I mean, on that point, you know, we're, With my rap career, um, I got to the point where I was touring over there a lot and um, you know, like around that same era when like Sean P was still alive and all those guys were out there a lot. Like, you know, we would do shows like Afu Ra and all those sorts of guys.
Um, And it was going well, but you know, then, then something happened. I'm not going to say it on recording. We [00:34:00] can talk about it afterwards, but, um, it got really difficult for me to get shows out of out there as a result. And so that's when it kind of, it overlapped with my SEO career. In fact, the first SEO job I got, um, was right before my first European tour.
And what happened was, uh, I got into an accident and I needed to get a job to pay my medical bills and the first place to hire me was an SEO agency. And I was like, cool, I'll just. Keep this job until November when I'm going overseas for this tour. And I tried to quit and they were like, you don't have to, you can just work from home.
And I was like, what, what you mean? I just take my laptop on the road and keep working there. Like, yeah. I was like, what? And that changed my life. Like once, once I could make. You know, consistent money while also wrapping. I was like, Oh, this is, this is everything, you know? And, um, even when I started working at razor fish, uh, in Philly, um, you know, I was, I was only a [00:35:00] contractor there and they only had me working like 24 hours a week.
And so I was like, cool, let's just make my days Tuesday through Thursday. And then I can do shows on the weekend. So like. I would just like, you know, go to Dublin real quick and do a show. And my DJ flip or, you know what I mean? Just like random stuff like that. And it was cool. But then I was like, you know what, if I stick with this, maybe I can make some real money.
And my, my, one of my really good friends who I used to tour with, we actually on the train in Sweden on the way to a show, and he was like, Mike, I know you're doing SEO. My cousin has an SEO software company. I'm like, who's your cousin? He's like, yeah, his name is ran Fishkin. I was like, what? And so like my worlds were like colliding.
And then I was like, all right, maybe this is a sign. I need to like do more of this. And yeah, so. Um, yeah, it just worked out for me in that way where it was just all connected. That's awesome. I have to ask a LinkedIn question, like, Oh, or LinkedIn style question. So how have [00:36:00] your, your marketing skills informed you as a musician or vice versa, because you released an, you released an album a couple of years ago.
Right. And you did a lot of stuff around, like a lot of content, a lot of like. You didn't pull any punches with the marketing. So do you feel like the time was right for the album because you were in a certain place in your marketing career as well, or with your marketing skills? Yeah. I mean, it was, it was very valuable what I do in my day to day for focusing on the music.
So two ways, one creatively, I looked at it as though it was like an agile project. And so, you know, when I first released the record, none of the features were on it. And I was like, this is the minimal viable product. I can just keep updating it. And so then I got a feature from music soul child. It was like this platinum R and B singer.
I got a feature from a little fame from MOP and then I just re released the records. And so every time I did [00:37:00] that, it like increased the um, the listenership. And so what I, and then I also staggered it. I released it on Bandcamp first and then I put it on all the DSP. So that way I can make like a couple dollars and then it could be available everywhere.
But then beyond that, it was also like, cool, what do I do from a content perspective as well? And, uh, also what do I do from like a traditional marketing perspective? Because, you know, when I was coming up, it was all about traditional. Like if you were doing things on the internet, you were considered not a real rapper.
They were like, Oh, you're an internet rapper. You don't have vinyl. You only do ads online like they didn't take you seriously. And so There was still this aspect of me wanting to feel official. And so I did a lot of traditional stuff. So like I did billboards in times square, I did, um, you know, I had my videos on MTV, like I was doing all that stuff, but none of that stuff did anything like what really did it, it's the playlist and whatever ads you run on like Instagram and [00:38:00] Tik campaigns.
So we tried everything and really like, that was my. You know, testing to see what's going to work now, because I hadn't put out a record in like. eight years, 10 years, something like that. So now I know how it works. So my next record is going to have more features because features is what gives you more, um, you know, listeners.
Yeah. And then also, uh, I'm going to focus, you know, very heavily on playlists and I'll run ads and that'll be it. Like I'm not going to do any of the traditional stuff. Uh, but for me, it's going to be like, have fun first. Because, you know, I know it's only going to go as far as like what I spend money on.
Um, it's not the sort of thing where I put a record out and it just like goes viral. Like I don't, I don't even think that's true. Like no one does that. It's always a function of, uh, you know, did you get influencers to do it? Is it one of those songs that people are going to use as a sound on TikTok?
Things like that. Um, And again, that's not hard to do. Like you could do that for 7, 500 bucks, but you've got to have a song [00:39:00] that's going to be catchy in that way. And that's not necessarily the type of music I make. So, um, you know, yes. In answer to your question, it was very helpful. So I can understand like what was possible.
I also found like all the right analytics tools to see what was going on. And, um, it just helped me have a plan. So I can understand like, What is going to be the return from this? Whereas before it was always like, I'm just going to put this record out and then what happens, you know, like I go on tour, I sell a few thousand copies.
Cool. You know, like it's not, it's not this big splash that you imagine it's going to be. But now with marketing, I know precisely what I need to do to be the biggest splash possible. That's awesome dude, like that you're still doing it. Like for me, the biggest win that I had from hip hop is that I met Yeezy Moby.
Oh wow. Oh my god, like I love, I used to work for a Finnish company called Music Info that distributes music in China and you get royalties. So I worked there, [00:40:00] uh, cause you know C Mobile Music Info, I worked there, I was my first touch with Finnish culture. So, I met, uh, met him on Instagram, and, uh, I talked to him on the phone, and he was always like, so And I, I was shocked when I was talking to him, like, I was starstruck.
I was just listening. I was like, yeah. My mouth was shaking and my husband was like, my husband was like, damn, like what's wrong with you? I was like, it's fucking easy. He's so big. He's a nice guy and he's all into the sustainability and stuff. Yeah. Anyway, I digress. I would talk about hip hop. All day. If you want to do that too.
But so I live in Brooklyn now and my cousin who's from Brooklyn where he gets his haircut. Easy Mo B is in there all the time. You know, so like, like that's the thing living in New York, you run into such a New York story. Like large professor used to live in my old apartment building. I would see that guy all the time.
You know what I'm saying? So like, and, and when you go out to parties here, you run into [00:41:00] people all the time, right? Like one of my favorite rappers is Feral Munch and, um, You know, I see him regularly, right? Yeah. We'll just like run into each other and we'll just like catch up real quickly. Or for instance, he, um, his Twitter got hacked.
And so, you know, his people reached out to me, his, uh, assistant Zoe. She was like, yo, can you help? I was like, say less. I got you. I love being able to help the, you know, the rappers I love with the skills that I have and the network that I have and so on. That's the, we need to have you just to talk about hiphop.
It's like it would be a long one. 'cause I, I have a lot to contribute. . We can talk about that. We can talk about. And don't forget, my first website was, my first website was about big is small, so let's not forget that. Nice. There you go. We can talk about take nine. 'cause Take Nine is a good mixture of metal and hip hop and I've seen is amazing.
You don't have to throw bone, you don't have to throw me a bone. I've seen tech in live. He came to Bucharest. I've never seen in my life [00:42:00] such performance. Like that dude can rap his fucking ass out, but still dance. And he was with Chris Calico. I have a picture with them and I was so starstruck that my face was like this.
Like, I'm, I'm I'm going to stop. Simo, save, save this, save this podcast, because I will go. I love, I love the vibe here, but we are, we only have 20. Let's talk about vector embeddings. No, before we go to vector embeddings, and I know you want to go there, let's do a quick recap because you said you've had a busy three months and it's obviously because of the algo leak.
And so can we do a quick, like, give us the, the elevator pitch about the leak I want to know where we are today. Like what's the status? Uh, do you see the industry having any impact? Because I see a lot of people being like, you know, this is stuff we already knew, but now it's just in the league and trying to kind of float above the discussion.
But sure. Everyone wants to be like, Oh, yeah, exactly. I knew that already. Yeah, that's people being petty because they weren't involved. Frankly, they discovered it, right? Um, so for me, I mean, first [00:43:00] of all, it's it's the ingredients for the algorithm. It's not the recipe, right? So, like, we learn more about the features that Google is tracking APIs and how they're potentially using them and things like that.
There's a lot of little blurbs that says this is used in this way for this thing. And also validation and verification is valuable, right? Like, even if we thought we knew it. We didn't have validation of it before because, you know, the way the algorithms work. First of all, it's not just one algorithm.
It's a series of micro services that get triggered differently based on different contexts of the user, their query and so on. Right? So even if you felt like you knew everything about how search works, Well, you don't know how it works in video search. You don't know how it works in, you know, core web search.
And even in web search, there's a variety of different ways it can work. So, everything you thought you knew anecdotally, you were sometimes wrong. And you weren't really sure why. So I think that there's a lot of value in us understanding Where [00:44:00] those edges are. What is Google actually looking at versus the things we thought they were looking at?
Um, and then there's a lot in there that we didn't explicitly know, right? So as an example, one of the things that stands out to me is they have this a feature called source type and it talks about your links being in different segments of the index. And even, even though that there's different segments of the index, I don't think was a widely known feature.
Right. Like we always looked at the indexes, like this one giant database or this giant big table table, or, um, you know, just like this one box where everything goes, but really it's broken down into different layers and where your page, where the page is linking to you live in the different layers impacts the value of the link.
That's not something that's ever been contemplated in any of these scores. Like. Page authority, domain authority, any of that stuff, right? So understanding that tells us that link building being a [00:45:00] volume game was always the wrong way to do it. And so, you know, that for me fundamentally should change how you do SEO.
In addition, we also have Complete verification that they are using click data to inform what should rank. So, you know, it always made sense to me that they did that. It's like a best practice and information retrieval, but Google, yeah, they always said, no, it's not. It's, it's a noisy signal. We can't use it.
The reason why is because they need it so much. It is, this is the only way it's. Scale to verify that a user is getting what they want. And so it's not just like, Oh, is this being clicked on? It's also, is it being clicked on and the user stays on the page or did they click on it, go back to the SERP, perform another query, and they end up clicking on that result again for the reformulated query, or is it that they, you know, went to your result, came back and then left Google entirely like these are all signals that they're using.
And so. What that tells me is that UX and [00:46:00] SEO need to work better together. Like they're, they're obviously completely different disciplines, but you can't do SEO longterm effectively without thinking about UX. And I think that people have looked at those things in isolation for too long. And, you know, I think you should really have like a UX person involved in your SEO teams moving forward.
So those are like high level things, but there's a lot of minutia in there that you can. Change your strategies around as well. But I would say those two things are the biggest that people should be focusing on. How do you like, how, how do you dare to act on those signals when you don't know their relative weights or which Google actually ends up using or which ones are like just collecting dust in the corner of the, So I think you've got to use the context for other things that we've gotten.
So as of late, um, you know, Google has been in the antitrust trial where there's been a lot of testimony about things. And so, um, a [00:47:00] lot of what we found in the leak reinforces the things that were said in the testimony. So they have a system called Navboost. Navboost is the system that drives all like the click information and user activity that's used to reinforce rankings.
And In the testimony, they said like, yes, this is the biggest. factor in our ranking algorithm. So while we don't have the individual weights for each of the features, we do know that they prioritize that. We also have like the history in SEO where we've seen over time, um, different factors appearing higher in correlation studies and so on.
And, you know, in some cases what we were measuring in SEO were the wrong things. But they were, they overlapped with the things that Google is measuring. So as an example, like we always looked at TFIDF and SEO, but Google has been using what's called BM 25 effectively since the beginning. Right. And so to some degree, they measure very similar things, but as the [00:48:00] BM 25 score of a query versus a document.
That you need to be thinking about. And then also because, because, uh, you know, Juliana mentioned, um, uh, vector embeddings. We have definitive proof now that Google is vectorizing all pages, and they're also comparing it to a site level vector. And that gives us the sense that they're saying, like, Hey, if your content is too unrelated, So what the core site is talking about, then we're probably going to say you're not an expert on this subject.
And that content is mostly just like diluting your expertise on your core subject. So that's another thing. You can't just be writing about random stuff and expecting to rank because Google is looking to see like, does this align with what you normally talk about? So, yeah, and answer to your question, I mean, we're never going to know what all the weights are, but we, we can use context clues to understand like what is considered more important and focus more on that.
So you're, you were the, [00:49:00] you and Rand and then, um, Dan Petrich, right? You were the, like the, the people who were first exposed to this, this leak. And so how does it feel to have the responsibility of translating something that's this incredibly huge and complex into a language? That the 99. 99 percent of, you know, marketers that don't understand the technology or the kind of the computational linguistic underpinnings that are here.
Yeah. How did you take that responsibility? Because you wrote a lot and you were in many podcasts and interviews and all that. Like, did you feel the weight? Yeah, I mean, I feel like I've been training for this my whole career, you know, like, like I've been the person, I've been one of the people that's been so really trying to figure out how does this thing work.
And so that, that's the core thesis of my book, that You know, how Google works isn't magic. It's just a series of different computer science things, right? Like it's information retrievals, NLP. It's, it's, you [00:50:00] know, computer vision. It's a variety of different things just tied together. And, you know, everyone that works at Google, they go to conferences, too.
They share what they've built. They share like different components of it. And when you go back to when Google first came out, they were a lot more forthcoming about Yeah. What it is that they were doing, right? If you look at, uh, papers from a meeting, God, the guy that used to run search like he he rewrote Google, you know, because because when it first came out with Sergei Brin and Larry Page wrote like they weren't good at coding.
They wrote something that was like passable. And then And then a missing gal, he studied under a guy named Gerard Sultan, who he's considered like the godfather of information retrieval. He came up with what's called the vector space model, which is what all search engines are based on, where you're basically taking a document in a query and you're plotting it in multi dimensional space.
And you're figuring out which documents are closest to the query. And then based on that, that's the most relevant [00:51:00] documents, right? So the guy that built. The beginnings of modern Google studied under that guy. And during that time that he was an academic, he shared a lot of papers about like, Hey, here's how search should work.
And then when he came to Google, he was like, here's an overview of how modern search works, which he wrote right after he rewrote Google. So if you read all these papers, you'll learn how Google works. Right. And then you've got people like Jeff Dean from Google, who's like considered the Chuck Norris of computer science, who is giving Uh, talks at Stanford and like, oh, six where he's like, here's how we built Google.
The architecture was broken before I fixed it. And I invented MapReduce and, and, you know, BigTable and all these things. And so if you stitch all that together, you'll know precisely how things work. And with all this generative AI stuff, like everyone is very loud about the things that they're coming up with.
And a lot of those people are Googlers, right? Like generative AI. Started with eight Googlers, [00:52:00] guys like Noam Shazier and all these other folks who were integral to the building of Google search, like generative AI works on that same model of that vector space model. And so they just shifted it to another thing.
And so you can learn everything about how this stuff works. If you just pay a lot of attention to really nerdy stuff. So my book is all about taking all that stuff and then synthesizing it and making it accessible to most people. So I say all that to say that once this leak came out, I was ready. I was like, Oh yeah, I've already been doing this.
And I have all the context required to make sense of this stuff too. So I was, I was excited and it was a holiday weekend in the States. And I was just like, yeah, let's go, let's do it. I'm getting after it. And I went through all the documents over the weekend. I spoke to Dan a bit cause he had already found it like three weeks ago and he had already started to make sense of it.
And yeah, I was like, the world needs to know how this [00:53:00] works. Here you go. That was, uh, that was awesome. And I finally now, cause I know we don't have, I have two questions, like very important cause I'm working, this is selfish questions. So if you're listening to this, I'm sorry. And you're welcome. So first question I, I had this hypothesis.
I, so, uh, context, cause you know, people know, but you, you don't know. I work in CRO experimentation, data science. That's what I do. So I came up with the hypothesis last time. You know, this summer in June, after reading your stuff, after, you know, doing a bit of research on Artvix myself, because I do like Google Scholar and Artvix, I, I found this thing called textual entailment.
So I said to myself, what if I analyze search intent, I then do a topic classification using natural language processing, and that's better than SEMrush or Ahrefs, you know, keywords ranking, it's a total different thing. Yeah. So, Now that you're saying that CRO and UX should be embedded in SEO, this [00:54:00] gives me a lot of like good vibes because lately that's kind of like what I've been trying to, uh, So my selfish question is I'm speaking at SMX in London.
I've never been at an SEO conference like I, the only thing I need, it's a t shirt to say that I don't belong. I just, I just need advice for how to go and talk to SEO people about this from a CRO perspective because my message is, I love y'all, but y'all need some CRO, y'all need some UX, like let's, let's, let's test something.
So I want to present, I, I built together with my colleague Rassimir, we use the specialized language model like BERT. And we are doing textual entailment on directly plugged up from the search console goes in GCP and then we analyze the queries and we do topic classification and then we kind of created, um, this metric that is kind of like a content predictive content consumption that looks at clicks, scroll average time on page, but that's [00:55:00] because I read your article.
Yeah. That's amazing. Just wanted, you know, to put that out there. Like, I got very inspired from your articles. I don't know you Mm-Hmm. But I was like, shit you do now. Yeah. No, but I, I was like, you gave me hope that there's other data sources that we can look at because for CRO we're very much using old dated ways of coming up with hypothesis, but search intent.
Cause a lot of people say, okay, but, uh, you, you can use SEMrush and you get top keywords. And I just wanted to, I told them that it's not the same, but maybe, you know, maybe you could share some thoughts about why is it not the same? I'm selfish. Okay. I'm done. Yeah. Uh, I mean, first of all, I think your approach sounds amazing and I would love to know more about it, but to your point, uh, SEMrush.
Their data is like the best you can get from a third party perspective, but it's still third party perspective. A lot of it is like modeled based on a variety of different things. And it's difficult to tie it to, you know, real analytics [00:56:00] data because it's so far away from what you can get from GSC. And that's not to say that GSC is perfect.
It's a completely different methodology from any other, you know, um, uh, analytics software, but nevertheless, it allows you to like, Plug it in a bit better than what you can do with SEMrush. And I think your approach of like building kind of like this composite score or what have you to determine like user satisfaction and so on is a great approach.
And in fact, there was a white paper that Google published, I want to call it like 10 years ago. Um, uh, I think it was called like the customer attention score or whatever. I'll send it to you, but they effectively did something similar where it is combining clicks, combining, um, you know, like user actions on the page, a variety of things, uh, to determine user satisfaction.
So, yeah, I think your approach is great. And I think it's definitely something that SEO should. Should so I won't get lynched, right? I won't get lynched. That's [00:57:00] not at all. I think I think again, especially because of, you know, what everyone has learned about Navboost and so on, especially if you tie your talk to that, I think it will resonate a lot because I've always been a proponent that you should do SEO and CRO together because, you know, SEO, if you're not thinking about a target audience and how it resonates and how they convert, you're not driving qualified traffic.
And so, um, My approach has always been what we call like persona driven keyword research, where we're aligning keywords with user segments. They're, they're, uh, there's stages in the user journey and so on and using that to inform things as early as like, you know, metadata and so on. So we can get people that we actually want to click on our results.
So I agree with you and I would love to see more metrics, uh, put in place for this. And in fact, we're working on a study with similar web. Cause we're trying to see like, what are some of the features of pages from a UX perspective that [00:58:00] yields higher time on site yields, you know, um, better user behavior from a search perspective so we can identify, um, you know, a series of things that everyone should be thinking about.
as they're optimizing their pages. And that's not to say like it replaces what you ask people do. But I want SEO is to be more educated on how you X impacts things. So as an example, if you think about like recipe pages, No one wants to know about the story behind your grandmother's apple pie or whatever.
They just want the recipe. And so my hypothesis is if you have like the jump links on the page, so it brings them directly to the recipe, then that's going to be a better experience for that user than like, Waiting through your grandmother's life story. And so I think it's, it's us understanding those elements from an SEO, from a UX perspective as SEOs is going to allow us to like, you know, [00:59:00] come to the table and say like, Hey, we need a UX person that can like do this or a UX person.
That's going to bring us better ideas on how this page can be better. Um, You know, in alignment with a user's expectations, and if you, and if you only consider SEO as like from the acquisition point of view and not from retention, it's just a throwaway traffic, right? Yeah, you're getting them in, but you're not keeping them there.
So what's the point of all that effort? No, and I know I hijack. I have one more question. I'm done. You were completely allowed to hijack. So my last question is, I'm more, I want to ask you, is there any way you can influence the search results of, uh, GPTs and other chatbots? So I read a lot about prompt injections.
I read a lot about Reddit and YouTube. So right now I want to ask you from, I know everybody's trying to do this right now to change the results in, you know, in chat GPT. So for, I, the, the, I read some good articles on search engine [01:00:00] journal, I, it's dope, but I want to know your, I don't want to say my opinions, I just want to know your opinion, so can you change the way, uh, GPT or others, you know, Answer about your company.
Is this a branding move? Yeah, I definitely feel it's a branding move. In fact, I have an article I'm finishing up right now about how these are branding channels um, and I think it's such a mistake for people to block these bots from ingesting their content because You know if you're looking for like basketball shoes or whatever And you're Nike and you're not there because you blocked your site.
And someone is like looking at chat GPT, they're going to be like, okay, like they're, they're not going to think about you in their consideration set because you know, you're not meeting the expectations in chat GPT. So as far as appearing in there. One, don't block your site. Two, um, you know, there are ways that you can mark up your content so that it is better [01:01:00] suited.
So the way that these things work with Retrieval Augmented Generation, and if y'all don't know, I mean, I'm sure your listeners will benefit from this. Um, What happens is the pages get broken down into chunks and the chunks are different sizes depending on the methodology behind it. But what happens is the, the search engine that's associated with the large language model returns the most relevant chunk to the query or the prompt, and then it feeds that to the language model to generate the response.
So what you need to do is make sure that you have the most relevant chunk on the web. For that query. And so it's a combination of that. And also, um, you know, there's a whole paper on this called generative engine optimization from a few researchers. And they found that they tried all the different, um, you know, methods that we do in SEO.
So they did like the keyword stuffing. They do all the things. And then they found that there are three things that yield better performance in these engines. One. It is more [01:02:00] authoritative language to exciting sources and three. It's using statistics. The statistic didn't even need to be real. It yielded better performance.
And then the last thing I would say is that use all the structured data that you can. So typically or historically in SEO, we only use the structured data that gets us rich results in Google. But large language models will use all of it. No matter what you've marked up. Um, and so what I would say is like, whatever vocabularies are relevant to your page, mark that up.
And then when you're your content gets ingested by these language models, you're giving more reinforcement that you should be related to whatever those queries are and you're preparing for a wider. And you're preparing for a wider array of search engines that consume the different types of, you know, absolutely.
Okay. I'm done. We have a lot of stuff in the show notes today. Please send me this paper. Cause I'm all of that. I'm actually looking at this right now with, uh, I [01:03:00] will shut up. Cause I'm, I'm, I'm not, I'm not, I'm not trying to bully you into stopping anything, but, but we do have to start wrapping up, but we left a lot of stuff on the table.
So we are of course going to have my comeback. Uh, maybe, so when, when's your book coming out? Do you dare to say a date? Okay. So we'll have Mike back soon. And, uh, in the meantime, so let us know where we can find you and, and follow your incredible journey with iPullRank of where we can read your research and everything.
Yeah. I publish most things at iPullRank. com. You can follow me at iPullRank on all the channels. And yeah, that's pretty much it. Awesome. Thank you so much. Thank you, Mike. It's always a pleasure.