The SEO Update by Yoast – September 2025 Edition

Took place on 30 September 2025 4 PM CEST / 10 AM EDT

Hosts

Alex Moss

Principal SEO

Carolyn Shelby

Principal SEO

Transcript

Mike
All right. Hello, hello. Welcome. Welcome, everybody. It is great to have you all here joining us for this SEO update. And please, yes, as Susan has already done from South Africa, she’s already dropped in the chat and said hello. Hello, Susan. And hello from Copenhagen as well, or to Copenhagen as well. I believe that’s Ula, maybe is how you pronounce that. Good morning to Wendelin as well. Hello from Flensburg, Germany, says Elona. Hello, hello. And Heidi, yes, we have a new look. Thank you for noticing that and thank you for calling it out. That’s awesome.

So welcome to this September edition of the Yoast SEO update. What we’re going to do today is have some fun. Of course, we’re going to recap and review all of the cool SEO news that was fit to print this last month. So here’s a quick taste of what you can expect in this fully edition of the SEO update. Number one, you’re going to get updated on the continuing massive impact of AI in search and SEO. You’re going to hear about some legal decisions and cases that are relevant in our space. And you’re going to learn more about the concept of personas being critical for AI search. Just a little bit about that. To quickly introduce myself, my name is Mike, and I am a support team lead at Yoast. And I am honored to be your SEO update host today.

Now, before we begin, here are two quick notes for the newcomers. Number one is this webinar is recorded, so you will be able to rewatch it later. No problem. And number two, all the links to the sources mentioned today, articles and whatnot, and on all of our SEO updates, links to all of the sources that we mentioned in all of our updates can be found on the yoast.com webinars page. Just go there, find the update or the webinar that you want to find the sources and links for, click on that, and then just scroll down and you’ll see the long list of links there on that page. Okay, so after our hosts, Alex and Carolyn, discuss the news today, we will have time for Q&A, so be sure to ask questions in the chat when and if you have them. And the first question you might be asking is, well, who are Alex and Carolyn?

Well, they are our principal SEOs at Yoast. They’re Carolyn Shelby and Alex Moss. And I’m going to bring Carolyn up on stage first. She is. And Carolyn, she specializes in enterprise, technical, and news SEO. And she’s passionate about demystifying the intricacies of search engine optimization for businesses of all sizes. Now let’s bring Alex. Let me do a quick intro for Alex. So he has a strong background in technical and structural SEO, especially through the lens of digital marketing. He’s involved with many aspects of Yoast from the product roadmap to content strategy. Okay. Intro’s done. Now let’s get to the update. Alex and Carolyn, the floor is yours.

Alex
Thanks, Mike. Thanks for having us. Hello, Carolyn.

Carolyn
How are you? I’m doing fine. I’m a little jet lagged still, but I’m surviving. So are you ready to get rolling? Oh, there’s the slides. Perfect. So I wanted to start out with, you may see there’s a little button at the top of the little banner that says SEO survey. I’m going to talk a little bit more about that in a couple slides. But if you want to click on that at any point in time, you can go take our survey. Again, I will tell you about it in like two minutes. First, feel free to ask questions. There is a little, it looks like a little thought bubble with a question mark in it. If you click on that, you can ask us questions. And at the end, we will try to answer as many as we can. Today’s topic will be available at HTTPS slash slash. I don’t even know how to say this. It’s Yoast. It’s yoa.st/update-september-2025. That’s where the recording will be. That’s where all of the notes will be. You’ll be able to find those things there afterwards. So the survey, the survey is we’re surveying all of our users about what they do, their opinions on things. It’s going to be very helpful to us. So if you could do us a favor and take just a couple minutes out of your day, preferably not while we’re talking about the update because, you know, important things that we want you to hear, but immediately afterwards would be perfectly cool too. If you want to take that survey for us, it’d be a huge help and we would be eternally grateful. Let’s see.

Next, I have some new helpful items on the top of the slide and I can point it out when we get to it. The little stopwatch means that the things that we’re talking about in the slide are immediate use to you. And the chess pieces are long-term strategy. That was like the best I could come up for strategery. But that hopefully will help you you’re going back if you want to review things and quickly identify what’s immediately useful and what’s more, you know, thought that’ll help you pick those out. So are you ready to rock and roll, Alex?

Alex
Yes, let’s do this in good time as well. All right.

Carolyn
So at the end of last month, after our update, ChatGPT announced that they were using Google search to answer all of your, all of our questions, not your plural. This was kind of important because immediately after this happened, well, first of all, everyone thought it was Bing that was being used. Then they said, no, no, no, now we’re using Google. And it was obvious from the way the answers were coming back that, yeah, it really did overlap a great deal with the Google index. The more interesting thing, though, was that immediately after that, do you want to tell me what happened?

Alex
Yeah, well, something called the recoupling happened, maybe, if you want to call it that. So maybe in a previous update, we might have talked to you about the decoupling that Carolyn also named, you know, the crocodile, the crocodile mouth, where you saw an increase of impressions and the decline of visits. What was actually happening in hindsight, that’s the way we do it, is that instead of it happening like this, it was actually happening like that and everything was going down. Now, to a Search Console report, that may look quite bad. But then there may be another argument that maybe all of these LLMs are scraping a bit too much, skewing the data of impressions, and therefore taking them away might give a more accurate outlook on what may be looked at by an LLM, but also has less visibility because these things can’t scrape to page 10 anymore and can’t see a hundred positions and therefore a hundred citations, sources of information as they’re looking at things.

Carolyn
I feel like there’s a more important takeaway and I feel like it is. So the taking away of the 100 search results means that you used to be able to force Google to show you the top 100 results instead of the top 10 all on one page. And all of the tools were using this, right, to get the search data because they don’t want to have to go through 10 pages. They just want to scrape everything off of one, which is precisely why they would only give you the first 10 pages of results. They wouldn’t give you the first 200 or 300. They would only do the first 100 because that’s all they could get Google to provide for them in a single page. Why is this important? This is important because at the, when you looked at the keyword data and the search data, you’d say, oh, we’re getting impressions for, you know, pink, but pink bunnies with purple polka dots when actually it was on like page 10 and no human being was ever looking at that. So you weren’t really getting the impression, were you? I think, I think we were getting a lot of bad data. I think it was telling us that we were getting a lot of impressions for things that realistically no human being was ever laying eyes on. It was all the scrapers and the robots. So while Google may have done this specifically to screw with ChatGPT, I think they inadvertently did us a favor because we’re going to have a much more realistic, accurate idea of which of our phrases and pages are getting actual impressions versus these are just machine impressions. So I think a great many of us are feeling less good about ourselves than we were before.

Alex
It depends, though. I guess if you look at the ranking report, that’s all going to shoot up at the same time as the impressions going down because the average rank is no longer between 1 and 100. It’s between 1 and 10. So from that moment onwards, every single average ranking is always going to be between 1 and 10, and that’s it. But again, that may make a more accurate insight for you as an SEO when you’re reporting on things. And Tim Soulo from Ahrefs, he made a good thought that maybe anything beyond page two isn’t even worth looking at. That’s just, is it indexed or not? I mean, personally, I like page 21 to 30. I like page three, especially if you’re an SMB, because you can see a bit where you can potentially climb to. But I also get from an enterprise point of view why beyond page two is not as valuable as it used to be.

Carolyn
How are we going to identify pages that are within striking distance that we should be focusing on to move them up to page one if we don’t know or don’t have good tools for telling us what’s on page two and page three?

Alex
No, that and what is a page anymore, because that’s all changing, I guess. So that’s a whole other podcast. Maybe we’ll do another one.

Carolyn
Probably true. So correlating to all of this though, Sistrix reported on a sharp drop in ChatGPT web searches, which is almost certainly associated or caused by the change to their ability to scrape Google.

Alex
Yes. And it’s interesting because Steve Payne at Sistrix’s did a bit around this looking at it. And whilst it’s kind of, it’s confirmation and what we were all theorizing what was happening, especially because Google wasn’t actually formally recognizing what had just happened. I think at most, I think, I don’t know who it was. It may have been John Mueller who just kind of said, we have never formally supported NumEquals100, which I guess is code talk for, yeah, well, we’ve taken that away because it never had to be there in the first place. And we just gave it you maybe because it was nice to play around with. And now you don’t know what you’ve got till it’s gone. Who knows?

Carolyn
Who knows? A long time ago, there was a comic called The Operator from Hell, and it was the system administrator, basically the guy who did all the Unix stuff in the basement of your university computing lab. And people would say things like, I need more space in my account. Like, how much space do you want? I don’t know. How much can you give me? And he’d go clickety, clickety. You have two megs free. Two megs? I had two megs to begin with. You doubled it. You have two megs free. And all he did was delete all of your stuff. So that answer just very much reminds me of a very typical old school Unix answer.

Alex
Yeah, but it’ll be interesting to see how in a year from now, A, whether it comes back and gets supported again. B, if it doesn’t, what happens to success metrics and were the success metrics before actually worth as much value as they are now with the change that’s happening. but something we all have to keep our eyes peeled on, right?

Carolyn
Well, and specifically, I think if we keep our eye on the referral traffic coming from ChatGPT, so the ChatGPT referrals will have refer equals ChatGPT in the URL string. So you’ll be able to see that in your Google Analytics and in GA4, whatever it’s called now, and presumably also in your Google Search Console. So if you are looking to see how many refers you’re getting from ChatGPT, that’s where you would look for it and that’s how you would find it. Let’s see. The next new bit of information, which is thankfully unrelated to the numbers dropping, is that the judgment came down or the remedy was issued by Judge Meta that Google will not be broken up in the Monopoly case, but they will be required to share search data. So they’re not going to have to sell Chrome. They’re not going to have to sell off anything. They’re not going to be broken up into a lot of little pieces, but they are going to be required to share their proprietary data that they were getting through Chrome. Now, the interesting thing to this is, number one, that this proposal basically takes away Google’s competitive edge. But this does not mean this is what’s going to happen. Google can’t appeal the judgment until the judgment is rendered. So there was no way they could fight this until this decision came out. Now that the decision has come out, now they have something concrete that they can argue against. So just because this has happened. It was very similar to they cited the Massachusetts vs. Microsoft case, which I don’t know if you remember a year ago. I said it was probably going to happen. So I’m very proud of myself for that. But now that this has been issued, Google can appeal it. So this is not going to happen anytime soon. There’s still an appeals process and they’re going to fight it as one would. But it is interesting to know that they now no longer have to worry about being broken up. They only have to fight whether or not they’re going to be required or mandated to share their data. My guess is much like with the NUM 100, they’re going to just stop collecting data. They’ll make it so that what they’re giving away to people isn’t as useful as it would have been when they were collecting it themselves, because I doubt the government can force them to maintain the same level of data collection. So it’s sort of like, I’m trying to think of a good analogy. It’s sort of like, oh, that’s awful. Maybe I won’t say that. So like, if you’re a bad person and you’re forced to pay child support, and in order to not pay child support, you just stop working and say, oh, I don’t have any income. It’s kind of like that. Yeah, we’re going to move on.

Alex
Which won’t work in the long term, right? because you still kind of beat yourself a little bit.

Carolyn
We’re going to move on from that one. Tell me about the ChatGPT experience improvements.

Alex
Yeah, so all of these LLMs have kind of been, you just sign up and go, and then it personalizes along the way. But now there’s a bit more, there’s a bit more thoughts into what goes on into who’s using it and the personalization it has. And you can do things. So I’ve always thought like, well, how would a teenager use it? And maybe the answers that it provides to a teenager should be different to someone like me or a five-year-old, you know? And that’s what they’ve kind of focused on. And they’ve got study mode, which I know has come out, but as well as that identifies if they’re talking to someone who isn’t a fully grown adult and responds appropriately, but also responds appropriately if you’re in, I don’t know, a bad mental state such as distress or something like that, or maybe suicidal tendencies. It actually reminds me a long time ago, I went on a digital detox trip for digital bods, and I ended up with a product owner at Alexa back in the day. I think she’s still there now. And she was in charge of dealing with queries that were suicidal in nature and how to deal with the answers. Because six months prior to that, it would tell you the answer because you’re asking a question and it was providing an answer to that question. Obviously, it had to do things. And by the time that she, in those six months, she connected Alexa to the Samaritans. So if you ask questions, a human would just be on the other end going, are you okay? And actually that was just in America. I tested it when I came back here and it provided you with advice on what to do. And this is kind of the LLM version of doing that, which is probably going to help a lot because without going through any of the stories here, there have been a couple of horror stories about people either committing suicide or going down a really bad path because ChatGPT gave them positive reinforcement, if you want to describe it that way. But hey, that’s a good idea because it always agrees with you and is agreeable in essence. And maybe it shouldn’t agree with you on every little thing such as this.

Carolyn
I actually ran into this yesterday. I was talking to it and I said something. I was feeling anxious. And one of the things that I will frequently comment when I’m feeling anxious is that I’m going to die. And what I mean is I’m going to expire from anxiety, not that I’m actively seeking that outcome, but it’s like, you should call someone. You don’t do anything rash, you know, and it starts trying to talk me down off the ledge that I’m not on. So I had to say, explain that that was an expression that was an exasperated expression. I am in no way interested in harming myself. I am just concerned that I’m going to have a heart attack.

Alex
Well, that’s a good thing.

Carolyn
In that case, if you feel like you’re having chest pains, definitely go to the ER.

Alex
Well, at least it’s trying to help, right? I think maybe it needs to learn more off British input, I reckon, as well for data collection, because we’re quite more cutting and dry. And maybe we can get around the nuance of what an LLM thinks we’re saying when we’re not. But yes, very good for humanity, this part of caring about the end user, which is really good.

Carolyn
Yeah, absolutely. So that’s good. Other things that have happened recently. I don’t know if you recall, but last month when we had our show, I was at WordCamp. And I said, Danny Sullivan’s going to talk. Do you guys have any questions? So Danny Sullivan did give his talk. And essentially what came out of it was things that are good for SEO are effectively things that are good for the LLMs. Though I think he did mention that he doesn’t like the term GEO as much as, just like everyone else doesn’t like that term. So the good SEO is good GEO perhaps implies more of an endorsement than was there for the naming. The gist of it, though, is that, you know, things that we have been doing all along that are good for SEO and not just that. I think the problem that we run into is many people forget to do a lot of the fundamental SEO things. And they focus on a very small subset of things because those were tactics that worked. When people like Danny or John Mueller or the more old school SEOs say things like good SEO is good for GEO, they mean all of the good SEO things, including the things that a lot of people don’t bother doing. And is particularly those things that make it good for the LLMs. So when you go and review, do make sure that you’re doing all of the good SEO, not just the parts you remember to do or that you feel like doing because fundamentals are required. You can’t have fundamentals without fun. Anyway, do it all. Don’t pick and choose.

Alex
That is true. So maybe another message, maybe don’t just leave all of those, not old hat, but native foundations behind in order to chase whatever may be happening right now. Because what’s happening right now is quite likely not going to happen in the exact same way in six months. Remember that this is, we’re in the middle of the wild west and the dust has not settled in the way that the tools are and everyone’s in this rat race of, you know, what’s going to get at the forefront and that it’ll take another year or two at least for this dust to settle and for things to be a bit more consistent in the way that we report and deliver success.

Carolyn
There’s no constraints on the innovation. So everything is going 100 miles an hour and changing super rapidly. So don’t think that you’ve discovered something today that’s going to work forever. It’s going to work until it stops and it could stop in 10 minutes or it could stop in 10 months, you have no way of knowing.

Alex
Take a break, have a quick glass of water between the new tools, you know, just let it all get into the brain and then don’t worry too much.

Carolyn
Absolutely. On September 7th, Google AI mode, we were told that it might become the default search experience soon. I do fully expect this to become the default search experience. we just don’t know actually how soon weren’t you telling me that the guy who said this the Google guy had to retract it and walk it back

Alex
Yes, well not walk it back completely i think he kind of took three steps forward and then took two steps back so i think his he definitely has this is obviously alluding that AI mode may become some primary option as the default right and that I don’t know maybe the web search the native web search that we’ve all known before the LLMs came out is going to be the secondary. But he then took a step back and said, that’s not what I said. I don’t remember what he actually said, but I think it was probably a very well-produced PR tweet that said, we aren’t doing the thing that we’re probably going to do, but we can’t say that we’re doing it right now. So we’re not doing it right now, but we are going to be doing it soon.

Carolyn
We’re not doing it right now.

Alex
Yes. He should have done a winky face instead of a full-on smile face. That would have changed the whole context of tweet, wouldn’t it? But something to look at and also think about when you’re doing your normal foundational SEO is that this stuff is generally going to happen to all of the search engines.

Carolyn
And I think when people say that you don’t need to worry about the LLMs because Google has all of the market share and Google’s market share is not going down and all those other things. But AI mode is going to become the default and that is an LLM that is running that. So you do need to worry about it because once it becomes the default experience in Google, you just cannot get away from it and there will be no alternative or at least not an easy alternative. Most users aren’t sophisticated enough to bother trying to force it into an old mode because the old mode is what they’re comfortable with. They will just get used to using the new default. They might not like it, but they’ll get used to it eventually. And that’s how these things become adopted.

Alex
Yeah. That reminds me of when Facebook famously updated their timeline. It was a good eight years ago, but I remember when they did it. Outcry from everyone. Oh no, I liked it the old way. And the answer, well, there were two things to know. A, it’s free for you to use. You are the product, right? So you don’t have any way in what happens. And secondly, after four weeks, no one ever complained again. And I think if they probably reverted it back, everyone would complain that they liked the new old one. So it’s what everyone gets climatized to after a little while.

Carolyn
This was an interesting thing that came out. So in a court document, Google said that the open web is in rapid decline. And this is despite the fact that they had recently said in two other interviews that the web is thriving. And I think there’s a difference between, I think the interesting takeaway is that there’s a difference between the web and the open web. And Google recognizes those differences and speaks about it differently. So takeaway number one is just like with before, where we said, oh, this thing isn’t coming soon, but it’s probably coming, but we’re just not going to tell you when. They are very specific in their language. So you need to learn to interpret what they say with the very specific way they say it. You said something interesting earlier that we should start calling it the human web versus just the open web in general.

Alex
Yeah, I was speaking at SEO Brein. It’s the Netherlands SEO community last week, and we sponsored it as Yoast. And I did a talk about how we should all calm down and not fear what’s happening. And actually, what we do as SEOs is probably more valuable now than it was before, because there’s even less understanding from the normies. And the open web, I just thought was, it’s open publishing, it’s open words, free information out there. And maybe the word open is outdated now because of the agentic web in general and the presence of LLMs that maybe it’s, I called it the human web. I said it on stage without thinking because that’s who it was for. The content was for the humans, not necessarily for the robots to then become the gatekeeper to then decide whether that content is for human consumption. So maybe the human web here is on decline and words direct from web site to human is now going, that’s what’s declining and having these agents in the middle acting to synthesize all of that information for you.

Carolyn
As these agents become more prevalent, I used to be always built for the humans, built for the humans. You’re going to have to start building with the understanding that these agents are coming through because they’re going to replace the experience that you used to create on your website. They’re taking that and the humans aren’t going to experience it anymore. So you have to build for the thing that’s experiencing it, which means we’re not, we are going to lose the human web at some point because all of these experiences are going to be moved to and handled within these chatbot environments where we’re just interacting with our own personal little Jarvis and our own personal little Jarvis is going to run out and, you know, buy our Amazon stuff and put in our orders at Whole Foods and things like that. So it’s fascinating to get these little glimpses that, you know, the big companies like the Googles and the Metas do understand that there is a difference here and they’re actively planning for accommodating this. Therefore, you know, the takeaway here is we need to as well.

Alex
I think their challenge is merging them both together. Maybe not the best comparison, but a good comparison is the dark web. The presence of the dark web is some area that’s a bit more private, and only humans are in there, right? There are bots running the show to be the gatekeeper.

Carolyn
Do you know it’s a human cell? Do you?

Alex
Well, true, true. But maybe, I don’t know, maybe it’s kind of our minds are split into two for the human web and the agentic web, and that some of them, like the way that I use ChatGPT, different to how I use Claude, different to how I use Perplexity, different to how I use the Google native SERP, is that maybe that’s what we split our discovery processes depending on what we’re doing between a human web and an agentic web. Because there’s some stuff I’m still going to want to do. You know, like buying a car, I’ll get some stuff, maybe the initial searches from ChatGPT, but then there’s no way something like that right now is going to inform my final decision. I’m probably going to end up on the website and purchase manually. You know, you hope in the next two years, maybe I’ll be saying something else. Who knows? But it’s an interesting thing to think, especially with the next story as well, which is about pirated books and payouts. And, you know, it’s licensing. It’s, it’s a minefield, isn’t it?

Carolyn
It is. So this is one of those lawsuits that I was keeping an eye on. So Anthropic, which is Claude, has agreed to a $1.5 billion settlement with a class action lawsuit comprised of authors who say that their books were pirated and illegally used to train the LLM. so the the important thing to note they’re all excited because there’s like 500.000 titles that are going to get three thousand dollars each so if you’re a single author and you had three books that you claim were pirated you’re going to get 15 grand and that’s great because a lot of these authors don’t make a ton of money. Authors are like actors you think that the actors are all super rich because you think of just like authors you think of Stephen King you think of JK Rowling you think of the big names, right? The thing is, there’s a couple that make bajillions of dollars, and the vast majority of them do not make enough to live on. And that’s where all of these, the majority of these authors are from. So $3,000 per title is basically a windfall for them, and they’re super excited. The thing to note here, though, is that this was a settlement. When companies settle, they’re settling with the stipulation that they’re not going to admit any wrongdoing. This does not set a precedent and this isn’t going to change the laws because it was a settlement. So basically what happened was Anthropic opened their wallet and threw $1.5 billion at these people and their lawyers to make them shut up and go away. So going forward, there might be some licensing or they might be better about proving that they didn’t pirate the books because really the only stipulation here is that they bought the copy of the book. If my option is I have to pay you $3,000 because it looks like I did something wrong or I paid $19.95 to get a copy of your book from Barnes & Noble and then rip the digital copy anyway, then I’m just going to buy one copy of your book and do that. So I think people were irrationally excited about the settlement. I don’t think this is going to do what everyone thinks it’s going to do. It’s interesting, but the important thing to note is that this is not a precedent setting judgment. This is a settlement. And this was money being thrown at those authors to make them be quiet.

Alex
It’s very interesting. Also, that $1.5 billion is a settlement amount. Like, what else is behind it? What else is in that bank account?

Carolyn
There is so much stupid money in AI right now. Like, just unreasonable amounts of money that shouldn’t even exist. like you’re so big your brain cannot wrap your wrap itself around the quantity of money that is floating around in these companies right now. We don’t even really know how they got it or for what they got it because they don’t they’re not profitable and they don’t have products yet

Alex
Crazy, well maybe the next story is maybe part of a solution haven’t looked into this with granular detail but the RSS one of the RSS co-creators has launched a new protocol called called, well, it’s the RSL, it’s the actual collective. And in short, it’s about licensing content and the bots and agentic tools out there will be able to understand what the licensing deals are and then comply with them and also give cuts of potential revenue, which I think is very interesting. So it’s Eckart Walther, who’s one of the co-creators. But I do find it interesting that this is one co-creator RSS and they’re doing something very interesting with LLMs. And then you’ve got Avi Guhar inside Microsoft, who’s doing really interesting stuff with NLWeb as well, that’s using how websites can connect directly with the human as well and kind of cutting out the agent. And this is very interesting that these two minds and that they’re in different places right now are thinking in alignment of how content and publishing in general and being an owner of a website, of any service or products or news or whatever, can be controlled in some way if it’s just thought about. And right now, I have more faith in these people than, say, the OpenAIs of the world who are clearly here to capitalize off their own platform. And this is open source, you know and so so is so is NLWeb and I thought it was very interesting that this this seems to be the good of the internet much like RSS was much like schema is you know it’s those that mindset I would i’m having confidence in what their potential solution will be it’s it’s whether it’s going to be adopted though although they have got some adoption already from some big people

Carolyn
clearly like Reddit and Quora and Yahoo and so on. I think ultimately this solution is still requiring a lot. It leans heavily on the honor system and it’s still very easy to get around it anyway. So I think they’re looking to get money where money can be gotten, but the stealing is going to continue apace.

Alex
Too fast. Is it like crime and policing that like criminals may be just one step ahead and and that it’s up to the police to close loopholes in a way that people were abusing with black hat right and you just had to wait until Google filled that hole in

Carolyn
so that it wasn’t as effective anymore and this is No honestly i think it’s more like it it’s more like You know those cafes that they set up where you pay what you can afford and and what you think it’s worth. And if you can’t pay at all, then you don’t have to pay. It’s 100% the honor system.

Alex
Yeah. I just never think that. I don’t know that we don’t do that in England because I want to go broke in a day. Right? Because you have a sense of entitlement, I guess, don’t you? And power.

Carolyn
Well, I’ll just say those cafes aren’t all over the place in the US either, probably for the same reason. Let’s roll in so we can answer some questions at the end. Here’s another lawsuit. Rolling Stone has sued Google over AI summaries because Penske Media says that the AI summaries are illegally using the reporting and reducing their site traffic and ad revenue. Well, I have news. Ad revenue has been going down steadily for many years. This is just one more dip in the altitude as far as that goes.

Alex
Is this just because an entity, you can point a finger at an entity to give them blame and that everything else wasn’t something you could really point at, but this is something you can point at with a bit more accuracy, even though it’s just happening

Carolyn
with everyone and everywhere. It’s happening with everyone. You have to be able to, in order to file a lawsuit that has any chance of success, you have to have standing to file. You have to be able to prove financial damages, which is what they’re claiming that their ad revenue is going down because it’s stealing their content. You have to, and you have to sue somebody that has the means of paying, you know, making you whole. So this is sort of a, this is one of those test cases where you’ve got someone that’s outstanding, they can prove financial damages, and you’ve got a culprit that can be held responsible and has the means of making them whole. So this will be interesting to watch. I still don’t see this amounting to much. I fully expect this to be a settlement. Google will throw money at the publishers to shut them up, but I don’t think this is going to fundamentally change how AI overviews works, nor do I think this is fundamentally going to change any of the problems that the advertisers are seeing. I still contend that if your entire revenue model is based on advertising, you need to find a way to make money with declining impressions and declining clicks. Because I don’t see how those are going to be recovered.

Alex
Maybe it’s trying to recover something while there’s something to recover in the time that there is to recover it. Okay. The next one is interesting. I’ll put two people here. One’s Kevin Indig on the left and on the right is Mark Williams Cook. And both in different mediums talked about how important personas are in AI search in general. And Kevin has gone through, I’ll actually put it in the chat here, just so you can see, has done a whole post about how personas are very, very important. They should be important and should have been important. We were talking about the foundations, what we do as a job. They should have been there the whole time. But Kevin actually showed ways and described ways and methods to use those personas to try and feel out what the synthesized answers are and adapt to those things. And it’s a very interesting read. And as well as that, Mark Williams also kind of covered it unrelated in the last day or two. And I’ve just shoved that chat. They have a similar chat, but listen to ours more than his. I’ll tell him that. But I’ve made him aware that this video is being shared. But he talked about, for the first 10 minutes at least, how he used personas to do all the testing he was doing. And this is the guy who’s created People Also Asked, which is another great tool, which helps with things like fan out queries and stuff like that. So very interesting read and in short, really important. And maybe if you haven’t been concentrating on your personas enough at the moment, maybe audit them or go through persona identification and figure out ICPs, which is your ideal customer profile.

Carolyn
Definitely. So the Kevin Indig article is in Search Engine Journal. It is very, very long, but he has examples. He has, you know, things you can copy and paste and drop right into, ChatGPT or whatever LLM that you’re using. So I would recommend we could not possibly do it justice by talking about it here. So please go read it if you’re at all interested in trying out these personas. I think it would be a good use of your time.

Alex
Yeah.

Carolyn
This just came out. So Cloudflare is offering a way to block AI Overviews. I continue to believe this is a very bad idea. I maintain that this is not going to help you. They’re going to get the information from somewhere. Google will get the information from somewhere. And if it’s not from you, it will be someone else, probably a competitor. And that’s who’s going to get cited. Your information, unless you have something so very proprietary that no one else has it, they’re going to get it from somewhere else. So I think this is cutting off your nose to spite your face if you block the LLM from your websites. And I will tell that to anyone who will listen.

Alex
Yeah. I mean, I feel like there may be a small subset of businesses and verticals that may benefit from this. But I think there may be better ways of doing so. like the RSL initiative that we were talking about in the previous slide, or some midpoint or compromise that can give some content, but maybe not do it all and provide a means to go into a site. It’s also interesting that Cloudflare have been doing some stuff which is auto opt-in. So if you’re on Cloudflare, which I assume at least a handful in here are definitely going to be on, I am as well, and it’s surprising. You know how people tell people to go into their X profile and check what connected apps they have now and again just to sort out? This is the same now with Cloudflare. I feel like every 90 days I have to just quickly check all the settings and see if anything’s changed and see if they’ve limited some of my websites to some of these bots, which they definitely did do. And I did discover some. You know, some people may, for whatever business reason, like you say, want to keep it enabled, but I’m not seeing the win at the moment. I think we were comparing it before to magazines in a shelving unit in a convenience store and purposely wrapping up your magazine with a plastic cover and sealing it shut and not it being see-through, but saying this is a magazine by me, but not actually saying anything else about it. Most likely someone’s going to pick up the other magazine that you can open, read a couple of pages at least to see what’s inside, be happy with it, and buy that magazine. What they may not do is look at the magazine that’s de-branded, doesn’t really look like anything is in there, doesn’t even tell you what’s in there. But it’s up for debate. It depends.

Carolyn
Yeah, it really does. Again, I’m not seeing the potential in this. If you’re interested in more of what I think I wrote an article in a Search Engine Journal just last it published just last week it’s about the writing style that AI defaults to and it is a long article. It is not an article that says that you should not use AI to write it explains how there is a very obvious style that AI cannot stop itself from writing. And it is because the LLMs were over-indexed, which means they were trained heavily on to the point that this is really the majority of the training data on transcribed speech. So its writing is really mimicking speech instead of mimicking writing. And it tends to be better in short form than long form. So when you try to stretch it into long form, it kind of makes you hyperventilate and feel anxious. I was particularly happy with this article. I encourage everyone to go read it, but I don’t want to toot my own horn too much. Did you read it?

Alex
Yeah, of course. Of course. We read all of our… Wait, are you not reading mine?

Carolyn
I read yours all the time. Absolutely. Here, let’s move on.

Alex
Yeah. I saw the eyes one there. Okay. So Google is going live with search. Is it called live? Oh, just called Google Search Live, which I don’t know if you’ve been in the previous ones. We’ve talked about multimodal search where you may take a picture of something and then be able to add text like, what is this? Where do I buy. This is now a bit more immersive, if that’s the word to use, where you can just walk around with the camera, not take a picture or record a video, but just walk around and talk about something that you’re seeing with your eyes through the phone, and it will give you answers straight away. Yet again, it’s keeping you inside the experience and away from the shop or the site, the transaction or the information would be at, but at least it helps. But one thing that is for sure here is that this goes down to being really concise and detailed, but not over-detailed with the content to make sure that everything’s populated in the right way so it can be discovered in this way, such as, I don’t know, the mug in this screenshot here. If you sell that mug, make sure you’re writing enough about it that it can get picked up when someone’s chatting about it without much context, other than there’s an image from a still in a video of a mug.

Carolyn
And I think this is just further proof But all of the things that are coming out are taking the users away from your website and away from the experience that you’ve created on your website. And they’re putting that experience somewhere else. So we have to make sure that we’re creating our content and writing with the understanding that this experience is likely not happening somewhere where you can control it anymore. So you can’t rely on images or something that is unique and specific to your website to convey all of the information, which means we might need to be a little bit more blunt in our descriptions. You know, something is a particular shade of white. Make sure you say white, not Chantilly lace. That doesn’t necessarily mean that color to everybody. So it’s good to keep in mind, I think. And I do have this one marked as long-term strategy. So food for thought. We only have a couple more slides. Tell me about ChatGPT Plus because I have not actually tested this out, though it seems I am eligible to do so.

Alex
I’ve not tested it out too much. And also I hope I am. I don’t know, I’m in the UK. You probably get that stuff first. But this is kind of, think of it as an autopilot for certain tasks and things. and you can let it just run tasks while you’re asleep is how it’s being advertised. So that might be just routine tasks that may be done every day. I know that part of my routine task is to read the SEO news every single morning to make sure that I know what’s going on. That might be something that it can curate so that if something happened, if I had an early night in the UK and I was in bed before the end of the business day in LA and then something happened in LA, I would know and be told about it in the morning. And it’s kind of like, yeah, I guess you can have notifications and all the other old schools, native stuff we’ve been doing, but this does it and curates it for you. And it also connects to like your Gmail and your calendar. And it also takes context from what you know and what should be interested or might not be interested in. All of those things are there.

Carolyn
You know, it’d be cool. So just imagine this and I’ll try to go quick. They’ve got these new mirrors now for your bathroom that are like wifi connected and they can take pictures of you in selfies. What if you had like a heads up display and you could say, good morning, ChatGPT, give me a rundown of all the news this morning. So while you’re shaving or curling your hair or doing whatever it is that you do, it’s telling you, it’s summarizing all of the, all of the news that you were going to have to read anyway, and just verbally telling it to you so you can continue to get ready. Like that would be in addition to telling you what, what your schedule is in the morning and that your boss emailed you at two in the morning with some sort of urgent thing. And you’ve got a meeting in 10 minutes.

Alex
Yes. Yeah. And then it can like, when, if I was to ask Alexa what the news was in the morning, it’ll just tell me about the news, like current events, worldwide events, stuff that’s political stuff. But soon it’ll, it’ll know that it’s me. Well, it should know that it’s me already talking to them and that it’ll know that I want some SEO and AI mix inside there and just curate it for me. So that’s, that, that’ll be the way that Pulse is going, which will help be a personal assistant, which means that you’ll have to rely on it and it will be harder to get away from the crux of the platform that this is. And they’ve doubled down this week, haven’t they?

Carolyn
Yeah, along those lines, they’ve got this instant checkout and agentic commerce protocol, which is going to basically make it so that you can send the little AI bot out to do your shopping. Like it will do the shopping. It’ll go through the entire process. You tell it what your credit card is ahead of time. it will purchase everything and let you know with the, you know, okay, it’s bought. It’ll be here on Tuesday.

Alex
This is very good. I mean, I’ve just put the blog post from it and it’s less than 24 hours old, by the way, everyone here. So it’s very, very fresh news. But soon you will just be able to look for shoes and then it’ll tell you which shoes and you can just buy it right from the platform, right inside the app. You connect your credit cards, which sounds like that’s not going to be somehow hacked, but we’ll see. We’ll see what happens soon. and that makes everything a bit more automated but for us seos that’s really important because it’s now using potential organic because again it’s not doing ads at the moment is it it’s not paid this is all in line with everything it takes a little cut doesn’t it. Which they didn’t say what the cut was it just says we’ll take a small.

Carolyn
So real quick before because i know we’re running out of time. SEO takeaway for this, these agents, these agents use a basically a text only mode when they’re up until the point where they make the decision to commit to the transaction, which means if the text on your page cannot be seen without JavaScript and all sorts of other bells and whistles, if it’s any in any way hidden, it won’t be available to the browser that the agent is using to make its preliminary decisions because it’s not going to switch to a visual browser that can execute JavaScript and fill out forms until it’s decided that your site is the site it’s going to commit the transaction on. So do make sure that you’re making it easy and simple for these little agents to get from your homepage to the transaction level page and make it so that they have all the information they need to make the decisions. Because if you don’t make that easy for them, they will go to another site where it is easy. That is my SEO takeaway for that. So let’s keep going because we’re running out of time. For our also in the news, these are some additional articles that we’ve provided for you that we didn’t have time to discuss. The Google AI mode now in 180 countries gains agentic features. Everything is going agentic. Ahrefs acquired detail.com and SEO extension, which is a Chrome plugin, I think. The important thing there to note is that a lot of these big companies are buying publishers because they want to get into the LLM citations. And the easiest way to do that is to control the publishers. Google drops Search Console reporting for six structured data types. Google AI mode is coming to the Chrome address bar. It’s actually already here. I’ve got that. And Perplexity has introduced a search API. So all of those are available for you to see. Do you want to tell me about the Yoast news? Because this is kind of important.

Alex
So later this week, we’re going to be releasing AI Brand Insights. And this is just a couple of screenshots from it. I don’t know if that one’s quite small, but it’s going to be a new product that we’re going to be launching. And I urge you all to sign up and check it out because it’s It’s going to have a lot of good tidbits and data on what you can do to improve the potential share of voice between you and competitors when you’re cited within LLMs and much more.

Carolyn
The TLDR is it’s going to tell you how the LLMs see your company and it will give you queries. It’ll give you what they see when people ask them about you, which I think is going to be very helpful. And I think you can do it for competitors too. but we will have more information on that. It’s very exciting. So do look for that this week. There will be a huge announcement. And I think the next thing to tell you about is that if you saw our last Black Friday preparation coffee chat, we’re going to have a follow-up to that on Tuesday, November 4th. So we invite you to join Alex and I for brewing up sales, the last minute optimizations that you need for your Black Friday success. And then our next SEO update by Yoast will be Tuesday, October 21st. Same bat time, same bat channel. And I believe that means we are now, that was it. So why don’t we do some Q&A, Mike? Yes.

Mike
Sounds like a great plan. Thank you for sharing that awesome SEO update information and let’s get going. We have a question from Meg that says, So more on this striking distance concept that you mentioned briefly, Carolyn, which is, so her question is, is this approach to identifying keyword and optimization opportunities now no longer relevant?

Carolyn
I don’t know that it’s not relevant anymore. In fact, I would imagine it’s still very relevant. It’s just going to be a lot harder to do if our tools don’t know what’s on, you know, page two or page three, because those were the areas we were looking to improve upon. I’m not even sure how the tools are going to figure that out unless we’re manually looking for opportunities. Have you heard from any of the tools, Alex? How are they going to fix this?

Alex
No, because I think it’s a bit too early. Some are saying, right, no, we’re still going beyond page two or page three or whatever. But I just don’t think that’s going to be done in the long term. I think they’re just probably soaking up some cost. So either some of these tools will increase the cost a lot to get that information. And most people, I believe, will say, no I don’t think it’s worth that extra it’s like more to get less accurate data. So yeah, I think keywords I don’t know I’ve always thought of keywords as as a good ideation starter on a broader topic. So if I was talking about black shoes for example that does it add value if I write something about black shoes. Right, it’s just a really short tale and and how can I cluster that with cosine relevance and that kind of stuff to ensure that as much information is being presented, but you don’t want to go too spammy?

Carolyn
No, the striking distance stuff is more like finding the pages that are ranking now, but on page two. And what can I do to that page to boost it up to page one? Because we know it’s already got some traction. So I think this will be a problem for me. I will find this to be tremendously inconvenient. So I wish I had a better answer for you, Meg, but my hope is that we figure something out because it’s still definitely relevant. It’s just going to be a lot more difficult to do, at least right now, and hopefully it’ll be sorted.

Mike
All right. Let’s get to the next question here. This is a real quick one. Will you be at Brighton SEO?

Alex
Yoast will be. No, I’m not there on this time around. I was at April. I was at last April and October. But I’m at Semrush Spotlight the week after. So I’m only going to one because it would have been all over the place. But there is a stand. There’s somewhere where you can pick up a Lego person. I believe.

Carolyn
So there will be Yoast people there. It just won’t be either of us.

Alex
Yeah, just look for the purple top vests and a stand. And I believe there’s a prize of Lego as well. So get involved if you’re in the area, Ben.

Mike
Cool. Absolutely. All right. Here’s a bit of a longer question from Steve. He’s asking to comment on the importance of content and the importance of the whole corpus of content held on a single domain and the interlinking of content.

Alex
The interlinking of content is always going to be still really important. Internally, the linking and referencing on site is always going to be really, really good. And it will help crawlers and LLMs and the like. The importance of content. And I know that Carolyn can’t wait to say something. You’ve got the hand. I love it when she does that. But we were talking about tokenization before and that having too much content that doesn’t get to the point and isn’t concise is a bit I consider spammy to token spam. So I call it again. This was on SEO Brein last week. I call tokenization spam a thing. It’s when you create words for the sake of words for filler, which is what some old hat, gray hat techniques were. Just let’s get this one keyword set and make 150 pages for 150 variations. That’s not a thing that should have been done in the back, but it worked. But now it’s even less valuable and it’s not going to work as much. But content itself is really important. It’s just how it’s formatted, how it’s presented. Now you go, Carolyn.

Carolyn
I think the question is asking. I think it’s more related to how the LLMs are finding the content on your site. And is interlinking still necessary for the LLMs to find your content on your site? And I’m going to say yes, because, and if I’m wrong on the intent of the question, I apologize because I’m guessing here. When the LLMs at the inference layer run out to do their grounding or to grab stuff from the live web, they’re getting those URLs to go grab from a search. So that means that all of the things that you normally do to rank in the searches still have to be done because they’re getting those search results from somewhere. It starts with search results. And to get into the search results, you have to do all of your original SEO stuff. And that includes content interlinking in order to get into that list that the LLM is choosing from to go grab the information that they’re returning to answer the question. And yes, all of that on a single domain is very important. Your interlinking is all still very important because all of the things that go into ranking in a traditional SERP are still relevant. So that’s my take on it.

Mike
Got it. Cool. All right. Going a little bit legal with this one. So are these rulings, these rulings we discussed and settlements, are these an admission that the whole AI world is just beyond governance?

Carolyn
It is beyond governance. And they’re not rulings, they’re settlements. So short of the United States in particular, I don’t think is going to allow any, they’re going to discourage restrictive laws from being passed because we don’t want AI to just move somewhere else in the world to do whatever it wants anyway, because they’ll go somewhere that we don’t want them to go like China. So I think the likelihood that there’s going to be restrictive constraints put on AI right now is very low, like near zero. And it is beyond governance. And every problem that gets thrown, every lawsuit that gets thrown at them is largely just a money grab. Because when you see deep pockets that can be sued, you sue them. And if the pockets are deep enough and you are a big enough pain in their butt, they will pay you off to go away. I think that’s what this is.

Alex
Plus, we don’t really know what to govern. Like, I weirdly believe that it will create the governance for us. Because also remember that AI and LLM, they’re not capitalists in nature. They’re just there to do a job and do it effectively. So perhaps, perhaps it may figure out something that humans won’t like, but we’ll be…

Carolyn
Skynet’s going to decide humans are the problem at some point anyway.

Alex
Let’s see. Let’s see. We’ll get Arnie on next month.

Carolyn
Yeah.

Mike
That’s hilarious. All right. Let’s do one more question. And let’s see. There’s one actually asking about JSON-LD. So is JSON-LD schema, is it still important for LLMs?

Carolyn
Um I would say it never really was. So it’s important to get into the rankings that they’re getting to make their choices but once they get to your page they can’t see they anything that’s hidden. So if it’s not visible text, they won’t see it. The JSON-LD is schema is shoved into the head and they don’t read that. That being said you know, it’s still important for the SERP rankings. I think when people say that the LLMs like schema, they mean that if you put it the schema format your prompts, they’ll eat that up because they love the organization. But it’s, I don’t know if anyone remembers Lynx. Lynx was a browser in the before times that did not do pictures. So if you were to look at your page in Lynx, whatever is visible there is basically what the LLM can see. They can’t see any more than that. So it’s important to get into the rankings so that then the LLMs can come find you, but the LLMs themselves are not going to see it because it’s hidden from human view.

Alex
But I want to add to this, that that’s the answer today, right? Because right now, my belief is that it will become more important as things go along. And I do believe that you’re also thinking about a handful of LLMs and not the LLM world in general at how it’s ingested. So yes, some don’t look at JSON-LD. Some do, like NLWeb does, that one does, but these other ones don’t because I don’t think they were looking at them as much as well. But noting that things are populated in this way, being under the hood is a bit inconvenient, but it’s still important populated information that’s valuable elsewhere, perhaps through RSS feeds as well or XML output. that can be used elsewhere that LLMs will read. So maybe it’s a bit more indirect, but I just, what I don’t want to say is don’t do it. That’s the worst thing.

Carolyn
No, I would never say all this stuff.

Alex
Do it all. And you might find that in six months, once open, you don’t know, OpenAI may say on January the 1st, you know what? We’ve just integrated schema and JSON-LD everything. If you’ve got any structured data, we’re going to be ingesting it and watch as everyone all of a sudden will find that a focus when it should have been a focus the whole time.

Carolyn
It’s one of those things. It’s like brushing your teeth and putting on clean underwear. You need to do it every day, even if you don’t think it’s necessary because you never know when it’s going to be necessary.

Mike
Truth. All right. I think that’s a good place to wrap up. Thank you all for joining us. And remember that we have these every month. So you can sign up for October’s SEO update even now to get a reminder and put it on your calendar. We’d absolutely love to see you again next month and bring a friend. You can click on the register for the next SEO Update on the screen there. Thanks again and we will see you all soon. Bye-bye.

Alex
See you soon.

Topics and sources

SEO & AI news

Yoast news

NEW! Check out Yoast SEO AI+ 💥

Also in the news…

Transcript

Mike
All right. Hello, hello. Welcome. Welcome, everybody. It is great to have you all here joining us for this SEO update. And please, yes, as Susan has already done from South Africa, she’s already dropped in the chat and said hello. Hello, Susan. And hello from Copenhagen as well, or to Copenhagen as well. I believe that’s Ula, maybe that’s how you pronounce that. Good morning to Wendelin as well. Hello from Flensburg, Germany, says Elona. Hello, hello. And Heidi, yes, we have a new look. Thank you for noticing that, and thank you for calling it out. That’s awesome.

So welcome to this September edition of the Yoast SEO update. What we’re going to do today is have some fun. Of course, we’re going to recap and review all of the cool SEO news that was fit to print this last month. So here’s a quick taste of what you can expect in this full edition of the SEO update. Number one, you’re going to get updated on the continuing massive impact of AI in search and SEO. You’re going to hear about some legal decisions and cases that are relevant in our space. And you’re going to learn more about the concept of personas being critical for AI search. Just a little bit about that. To quickly introduce myself, my name is Mike, and I am a support team lead at Yoast. And I am honored to be your SEO update host today.

Now, before we begin, here are two quick notes for the newcomers. Number one is that this webinar is recorded, so you will be able to rewatch it later. No problem. And number two, all the links to the sources mentioned today, articles, and whatnot, and on all of our SEO updates, links to all of the sources that we mentioned in all of our updates can be found on the yoast.com webinars page. Just go there, find the update or the webinar that you want to find the sources and links for, click on that, and then just scroll dow,n and you’ll see the long list of links there on that page. Okay, so after our hosts, Alex and Carolyn, discuss the news today, we will have time for Q&A, so be sure to ask questions in the chat when and if you have them. And the first question you might be asking is, well, who are Alex and Carolyn?

Well, they are our principal SEOs at Yoast. They’re Carolyn Shelby and Alex Moss. And I’m going to bring Carolyn up on stage first. She is. And Carolyn specializes in enterprise, technical, and news SEO. And she’s passionate about demystifying the intricacies of search engine optimization for businesses of all sizes. Now, let’s bring Alex. Let me do a quick intro for Alex. So he has a strong background in technical and structural SEO, especially through the lens of digital marketing. He’s involved with many aspects of Yoast, from the product roadmap to content strategy. Okay. Intro’s done. Now let’s get to the update. Alex and Carolyn, the floor is yours.

Alex
Thanks, Mike. Thanks for having us. Hello, Carolyn.

Carolyn
How are you? I’m doing fine. I’m a little jet-lagged still, but I’m surviving. So are you ready to get rolling? Oh, there are the slides. Perfect. So I wanted to start out with, you may see there’s a little button at the top of the little banner that says SEO survey. I’m going to talk a little bit more about that in a couple of slides. But if you want to click on that at any point in time, you can go take our survey. Again, I will tell you about it in like two minutes. First, feel free to ask questions. There is a little, it looks like a little thought bubble with a question mark in it. If you click on that, you can ask us questions. And at the end, we will try to answer as many as we can. Today’s topic will be available at HTTPS slash slash. I don’t even know how to say this. It’s Yoast. It’s yoa.st/update-september-2025. That’s where the recording will be. That’s where all of the notes will be. You’ll be able to find those things there afterwards. So the survey is we’re surveying all of our users about what they do, their opinions on things. It’s going to be very helpful to us. So if you could do us a favor and take just a couple of minutes out of your day, preferably not while we’re talking about the update, because, you know, important things that we want you to hear, but immediately afterwards would be perfectly cool too. If you want to take that survey for us, it’d be a huge help, and we would be eternally grateful. Let’s see.

Next, I have some new helpful items at the top of the slide, and I can point them out when we get to it. The little stopwatch means that the things that we’re talking about in the slide are immediately useful to you. And the chess pieces are a long-term strategy. That was like the best I could come up with for strategery. But that hopefully will help you, you’re going back if you want to review things and quickly identify what’s immediately useful and what’s more, you know, thought that’ll help you pick those out. So, are you ready to rock and roll, Alex?

Alex
Yes, let’s do this in good time as well. All right.

Alex
No, that and what is a page anymore, because that’s all changing, I guess. So that’s a whole other podcast. Maybe we’ll do another one.

Alex
Which won’t work in the long term, right? because you still kind of beat yourself a little bit.

Carolyn
We’re going to move on from that one. Tell me about the ChatGPT experience improvements.

Alex
Well, that’s a good thing.

Carolyn
In that case, if you feel like you’re having chest pains, definitely go to the ER.

Alex
Take a break, have a quick glass of water between the new tools, you know, just let it all get into the brain and then don’t worry too much.

Carolyn
We’re not doing it right now.

Carolyn
Do you know it’s a human cell? Do you?

Alex
It’s very interesting. Also, that $1.5 billion is a settlement amount. Like, what else is behind it? What else is in that bank account?

Alex
Yeah.

Alex
Yeah, of course. Of course. We read all of our… Wait, are you not reading mine?

Carolyn
I read yours all the time. Absolutely. Here, let’s move on.

Mike
All right. Let’s get to the next question here. This is a real quick one. Will you be at Brighton SEO?

Carolyn
So there will be Yoast people there. It just won’t be either of us.

Alex
Yeah, just look for the purple top vests and a stand. And I believe there’s a prize of Lego as well. So get involved if you’re in the area, Ben.

Carolyn
Skynet’s going to decide humans are the problem at some point anyway.

Alex
Let’s see. Let’s see. We’ll get Arnie on next month.

Carolyn
Yeah.

Mike
That’s hilarious. All right. Let’s do one more question. And let’s see. There’s one actually asking about JSON-LD. So is JSON-LD schema, is it still important for LLMs?

Carolyn
No, I would never say all this stuff.

Alex
See you soon.

Missed our latest webinar? Catch up here!

View past webinars