← The Yoast SEO podcast
The Yoast SEO podcast
The latest SEO news: from HCU recovery to Product Variants Schema, AI and more | #SEOUpdateByYoast
00:59:05
Loading summary
Transcript125 lines
- [00:01]
Ninke
Hi, everyone. Thanks for listening to the February 2024 SEO update by Yoast. My name is Ninke. I'm Associate Director of Partnerships at Yoast and I'm your host for today. In this month's update, we have a lot of SEO and AI news to cover. If you're running an E commerce business, pay extra attention to the Google Update regarding product variations. Don't worry about it. Our principal SEOs, Alex and Carolyn will talk you through it in this edition of the SEO Update. Enjoy. And I'll be back for Q and A.
- [00:35]
Alex
Thank you very much. It's good to be here. It's been a busy month, hasn't it, Carolyn?
- [00:41]
Carolyn
It has. It feels like the month went. I can't believe it's the end of February already. Everything's going so quickly.
- [00:47]
Alex
I know, I know. And I thought again, it might be a slow news month, but again, we've been wrong.
- [00:53]
Carolyn
But again, you were wrong. It's a good thing you're not a betting man.
- [00:56]
Alex
No, exactly. No, I never bet. I know. I don't want to take these risks. And essentially, it's January. Everyone's done. Everyone's done their January. Guests in there need to do stuff in February. So let's do a load of updates and make a lot of confusing language that will make us debate a lot for the next month. So, well, let's get through it because I know there's so much to do. So what are we going to discuss today? SEO news, which I reckon will take quite a lot of the time. We've then got some AI news, because there has been a lot happening in AI too. It has been a busy month for a lot of. For a lot of people. And then after that, we'll go through a couple of WordPress items and lastly, we'll have a couple of Yoast updates and of course, Q and A. So first, the Q and A always need to make sure on the right hand side, I don't know where, which way I'm pointing, but on one side, on the right hand side, there is a nice question mark below the chat icon. Make sure that if you have any questions in there, ask in there. And also if you want a question asked that someone else's asked, then upvote it. And the most out votes are probably Q skip, won't they? Which is nice to know. But of course you can chat, tell us where you're from, tell us if you've got any other interests or anything else that you might have observed in the last month of SEO yourself. If you want to learn more about anything today, you can go on to this, which is YOA st. I can't say Yoast because it's not. It doesn't spell right yoa st update-Feb2024 where you'll be able to see this replayed video that you'll be able to do in slow speed if I'm talking too fast as well. And of course you can get all the links to the different stories we've been going through today. And if you need to learn more over the month, you can learn from how to Start with SEO. And of course, that's within SEO Academy, the Oast Academy. And it's also worth mentioning that this morning I published our updated SEO glossary, which I will at some point during this course of the update. We'll share in the chat so you can have a look and tell us what you think of it. So, SEO News, what's happened first? Imp. Tell us about what you think about imp. So it's paving way in what, two weeks or something?
- [03:16]
Carolyn
Yeah, so there's been a lot of confusion about what's getting replaced and what's not getting replaced with Core Web Vitals. I know initially when this was announced, everyone's like, oh, core Web vitals don't matter anymore. It's just one facet of Core Web vitals that's being replaced or a metric that they think is more appropriate and actionable than the old one. The old one was first Input delay. The new interaction is Interaction to Next paint. Interaction to Next Paint is, I would imagine, similar to largest Contentful paint. So it's just a metric. It's not the end of the world. No one thinks that just making maxing this out at 100 is going to instantly give you any kind of ranking boost. This is still a piece of the Core Web Vitals piece. I would focus, if I were you, on getting the majority of your pages, if not all of your pages, into the good area. Green. Get them green. Don't worry about making everything perfect. Don't let good be the enemy of. Or don't let perfect be the enemy of good. I never get that right. Okay, don't let perfect be the enemy of good. You want good. Everything has to be at least good. You don't want needs improvement. You don't want poor you. You want everything green. But other than that, green is fine. Doesn't have to be 100%.
- [04:44]
Alex
Exactly. That's why there's a whole point of having a space for green. Otherwise green would just be 99 and 100, if not just 100 and then everything else would be orange. So yeah, don't be too paranoid about every single metric because you'll spend more time on that than you should do on other more natural elements of the website. So still good to know that. Don't get too scared if something goes up or something goes down. Breathe. Have a look. There might be something very easy to do and if there's something more important to do, at least you've got a little bit more clearer information sent to a dev and it isn't just going to be a six month nightmare for you.
- [05:27]
Carolyn
Just I wouldn't, I wouldn't panic too much. This is, this is fixable, this is doable and it's not. This is by no means the end of the world or even a seismic shift in your reality. Just focus on getting everything to green. Don't worry about getting things to 100%.
- [05:43]
Alex
Yeah, yeah. So talking about getting things to 100%, we had a very, very long post by Jano Van Driel, who I'm sure would be a fan to hear that we're mentioning him here. So what you can see here is a schema entry on one URL. I think there was 700 in that, but I think he actually found something with 7,000 entities and trying to take Jono for his money as the most complex schema entry ever. But anyway, he went through a very big rabbit hole. So if you're in the mood for having a very complex read about structured data, EE80 and any if at all correlation you can do here. And he does it in a very nice way, bit of a storytelling way. But in short, no, it doesn't influence eat. Structured data and EAT whilst they have a connection in code, don't have a connection to do with actual more visibility. Maybe you can elaborate more on the connection or lack of restructured data in EAT and what people should think about here.
- [06:47]
Carolyn
Oh geez. So I think what we've discovered from reading all of this is that it's complicated and that everybody. The way people's brains make the connections from point A to point B when it comes to structured data and EAT and rankings, they go in different directions. I think the important thing is that we all kind of arrive at the same place. The things to remember about E A T is that it's not a ranking factor directly and fixing your structured data is not going to necessarily improve or change what Google thinks about you as a person. Actually, the more I think about this, the more Confused. I'm making myself. I guess the short story is you need the structured data to help the search engines understand what's on the page, what you're talking about on the page, but mostly to make it easy for them to take pieces out of the page that they want to use and repurpose for other things. It is not going to necessarily help with your rankings other than it makes the engines understand what's on the page better. And if what's on the page is worth ranking, then they will rank it. You're not going to get, you're not going to accidentally tell them the page is about something. It's not because the structured data is there to make it clear. Eat is is so subjective. It's not. You know we talked about that. What was the thing we talked about about Tucker Carlson having his name on an article isn't going to make that article rank better because he has more authority. It's because he is a more widely searched, higher search volume thing and that his thing, his entity is going to be attached to or associated with that article because his name is mentioned. So whether he wrote the article or the articles about him, it will show up for his name. It's not going to magically provide fairy dust that will make your entire site have more authority.
- [08:55]
Alex
Yeah.
- [08:55]
Carolyn
Which I'm still not sure is entirely clear.
- [08:58]
Alex
As mud maybe. I think we were trying to. Trying to describe it in the simplest way and using Tucker Carlson was a good example because he's such a well known. Sorry. He's got a lot of visibility online. Right. So scenario was would Tucker Carlson do write a post in a very low authority news outlet and not talk about it anywhere and the only thing connecting him to this post would be structured data. The point is would that, would it be influenced because of who took Carlson is. And the answer in short would be no. The result of him publishing it and other people seeing that and then linking to or citing that is the result of him as a person. And. But that is, but it's the factor that goes into the ranking is what happens as a result, not because it's him.
- [09:53]
Carolyn
Yeah. So he is a person, is popular and it will attract links because of who he is. But I mean that's not really, that's not really something you can affect through structured data. That's not a thing that you can affect through any machinations of your own. That is, that is a thing. That is a fact. And he is as a per, as an entity attracts links. So it's sort of like Newsjacking. I don't know if newsjacking a long time ago would be where you would write a news story about something and you would mention something that was related to it because you could work it into the article, but the article really wasn't about that, but because you mentioned it, it would attract links and get more, get more publicity. So long time ago, when Hillary Clinton was running against Barack Obama in the Democratic primary in the U.S. hillary came to West Lafayette and did a rally. Well, it never mentioned Barack Obama, but at the time he had a much greater search volume and more velocity in the searches and in Google News than Hillary did. So I added a line to the article that mentioned that her primary competitor in this race was Barack Obama. All I said, I mentioned his name in the first paragraph. The story got so much more, so much more traffic, so many more views because I sort of hijacked that mention. And I feel like this is very similar, but again, not a ranking signal. You're just sort of taking advantage of the notoriety of something else.
- [11:31]
Alex
Exactly.
- [11:32]
Carolyn
Like this is very clear as mud.
- [11:34]
Alex
Yeah. Yeah. Well, it's good to know, right? Good to know. But if anyone wants to go even down an even deeper rabbit hole, do read that. You'll get, you'll get a link in the next email about it. So what do we have next? Google have revamped the SEO starter guide, which is interesting. They've halved it in size and they've made it much simpler for a beginner to understand or a beginner to intermediate, which I think is really good. I mean, there's not much to cover in there that you know, it's not a news update, is it? It's just repurposing the same content. But the interesting thing is getting rid of that half because of a few reasons not to over complex, over complicate things. And also because some things with technology are kind of being dealt with, whether it's through plugins like ours or whether it's on Google side with how to read that stuff. But it was good to know that it's done. I know that advanced SEOs are like, why has it gone dumber? But it hasn't. It's just not for them.
- [12:38]
Carolyn
I think sometimes you have to write, you have to not get too complex or detailed or even nuanced with your instructions to people at the beginner end of the scale because it's a lot to keep track of. And if you, if you mess up any of the foundational points in your head, the more advanced stuff will start to not make sense, or if you skip over a foundational part and go straight to the advanced and you don't understand the mechanics of why this is what it is, then I think the problem that you run into is people start making leaps of judgment and logic, leaps that land them on not solid ground. And things get messed up and there's weird interpretations. It's best if you say less. There's less to misinterpret.
- [13:34]
Alex
Yeah, exactly. Which is, well, talking about less to interpret, you know, that's what Google have been doing this month. Which leads us to our next story of what's a factor, what's a signal, what's a system? And I don't know if anyone's been looking at discussions on Twitter or not all months. And you've got people like Ryan Jones who, you know well, who's been getting into this discussions as well. Well, it first kind of started from a tweet, I think it was by Danny Sullivan or from the search liaison that said core web vitals aren't a ranking factor. Now other people may disagree with that and go, how can you say that? You've been telling us all this time about speed, performance, introducing cwv, making sure. Telling us make sure everything's better, faster. We were just talking about INP trying to get into green just then. Why should we care if CWV isn't a factor? Well, he kind of answered that in there. We look at many things and not one thing and it might not be a direct ranking factor, but then they kind of again, add fuel to the fire. They say that it's not a factor, it's a signal. And it's kind of goes into like lawyer talk at that point. Do you know what I mean? What's the interpretation of a word? What is the word that you should use? And I even, I even made that gif myself when I was talking to Yarno, funnily enough about this. You know, everyone's pointing at different things of what is a signal, ranking and system. Again, overthinking, right? Saying too much, trying to interpret every word. Just.
- [15:09]
Carolyn
I think when Google talks about ranking factors, they mean the actual algorithm, which to be fair, they don't even know what that is at this point. The contributor, the signal that core web vitals CWV is sending is all of those things make up fast page load and good user experience. So while the core web vitals themselves individually are not part of the algorithm, officially, we all know that your page speed and the user experience are ranking factors. So core web Vitals explain to you the different things that go into a good user experience so that you can fix that. That's why it's important. But once you start, Google has always spoken in a very Clintonian sort of way. You have to semantically dissect everything they say and look at the words that are used in the context that they are used, and that is really the only thing that those will be applicable towards. They don't. They're not in the business of giving away the secret sauce, assuming they even know what the secret sauce is at this point. So are core web vitals signals or are they ranking factors? Google says it's not. We all know that it is clearly important because we can see a change in how we rank when your core web vitals go in the tank, because that means there's something wrong with the site. It's slow, the crawlers are having problems with it, the users are having problems with it. And when the users have problems with the site, Google is not going to recommend your site to the users because they want the users to be happy. So it all goes into this mismatch of how they're going to rank things. Just because Google says this is not officially part of the algorithm doesn't mean that they're not looking at it. And it doesn't mean you can ignore it.
- [17:11]
Alex
It's weird because as you were saying, secret sauce, in my head, my analogy was going to. I was thinking about McDonald's fries, for example, and the secret. And the actual source, the Big Mac sauce. And it is. People try to replicate it and you can't get it exact. But does that mean that the source isn't.
- [17:26]
Carolyn
I'm pretty sure it's just Russian dressing with relish.
- [17:29]
Alex
I'm sure it is. But like, for example, if you close your eyes and you eat McDonald's fry, you know, that's McDonald's fry. Quite. I would be able to. I think I'm that confident I'd be able to distinguish that fry from another fry. But the other fry is still a fry. It's. It's still a fry and it's still edible and it's still very tasty. But McDonald's have that specific recipe on it. But that doesn't mean that. No, fries aren't good fries. Maybe it's a bad example. Maybe it is. I don't know. Maybe I just like your fries. But you know, it's one of those. You can't. Unless it's clearly, this is just a raw potato and you've not done anything with it. You've not put it in the oven, you've not put anything around it, you've not put seasoning on it. It's not a proper fry, but I guess it's a potato. They're both potatoes. Now we're getting into entity asio. We should get away from this, shouldn't we?
- [18:17]
Carolyn
I feel like we could go a lot deeper on the potato analogy. Maybe we'll work on that.
- [18:22]
Alex
So what else have we got? Well, Google's taking stuff away again, aren't they?
- [18:27]
Carolyn
This is an interesting one. So web stories evolved from amp, and I know everyone likes to hate on amp, and AMP is being kind of lost to the sands of time at this point. That would explain why they're taking away web stories. Web stories are a complicated pain in the butt way of generating effectively a text based, HTMLized kind of reel or Instagram story. Except it's on Google and it's kind of on amp. And Google's putting it in their cache so they can serve it faster, which is what they did with amp, which is why AMP was amp. The point is though, it was hard to do. There was no easy way to do this and it wasn't generating. It wasn't generating enough revenue. I looked at Google's development site and I remember when they launched this, they were kind of twisting the arms of big news publishers to go into and invest heavily in these web stories. And we'll link to you, we'll use you as a case study. One of the primary case studies was vice. VICE is shutting down this week. So clearly that went well for them, I think. Long story short, is it bad that they're going away? No. Is anyone going to miss them? No.
- [19:45]
Alex
It's a shame. It's a shame. That's what happens, doesn't it? But yeah, web stories, it's like, do they like the news or not like the news? But we'll see. Probably next month they'll brought it back like FAQ snippets, right?
- [19:56]
Carolyn
Yeah, I don't think they will though, because I don't think there's. I think they're going to just lean into the reels and the Instagram stories and the things that already exist that people are actually using and monetizing. I just don't think enough, I don't think there was enough adoption of these web stories to make them viable.
- [20:15]
Alex
Which is, which is good though. A little more. A little more real estate for everyone else as well, hopefully. Yeah. So what's next? HCU haven't Heard that in a couple of months because nothing's been updated or nothing's turned. But one thing that was interesting this month was a lot of debate about classifiers and timings on do I wait? What do I do if I get. If I got hit by the helpful contents update, and I do, and I did everything I needed to, theoretically, the next day, do I have to wait until Google does something again, updates the algorithm? And in short, the answer is no. If you were to do something, everything you needed to do within 24 hours, well, at least. What. What search liaison saying you should be able to recover without waiting for what's going to happen next. And if you continue to do stuff in a positive way, the classifier runs continuously and therefore the quicker you do things, the quicker you'll recover. Theoretically.
- [21:19]
Carolyn
But you have to do the right things. You have to do things that will actually make your content more helpful, not just rearrange the deck chairs on the Titanic.
- [21:28]
Alex
Yeah, well, what's the right thing here then? So there's no interpretation.
- [21:32]
Carolyn
I mean, your content, not crap.
- [21:35]
Alex
Exactly. So it's interesting to see a lot of people who are saying, I've not recovered yet. I've done everything I can. But part of me also says, well, an algorithm is a computer and it's checking on you, so clearly you've not done what it wants you to do, or have you just covered.
- [21:55]
Carolyn
When I've done everything that I can. Sounds an awful lot like my children when they say, I can't find my shoes for school. I looked everywhere, and without fail, I will say, if I go check in your room, under your bed, will I find your shoes? No. Did you look everywhere? Yes. Did you look there? No. Okay. It just. It seems like a thing a kid would say. Clearly, clearly, if you have not recovered, you have not done the right things. So you have not done everything. The HCU is about helpful content. If you got hit, your content was not helpful, therefore make your content helpful.
- [22:37]
Alex
Yeah, I think one thing he didn't answer, though, or they didn't answer, whoever was behind the tweet that day, was if you did everything you needed to and you completed your list today, how long it would be until that recovery happens? There's no. There was no set timeline.
- [22:54]
Carolyn
So once that's running all the time, then I guess it would be how long will it take the crawler to grab your data, recalculate it, and then determine that you've done enough to start seeing a change? So if they're looking at. Let's pretend that we know what they're doing. Let's pretend they're looking at the average of the content throughout the entirety of your site. That means it'll have to crawl all of the content on your site, or at least enough of it to count as 50% plus one, or whatever it is their magic number they're looking for. And everything that they've looked at has to add up to the right number. I'm sure they know exactly how many, how many pieces of content they have your site has submitted to the index. Right. If they know the total, they know the magic number for that percentage. They've got to recrawl everything and you've got to hit the magic number. Do you know what the magic number is? No. Are they going to tell you also? No. Just go through all of your content, all of the content that they might be looking at and make sure that it's helpful. If you have not touched every single piece of content to ensure that it is meeting that, that, that standard, that helpful standard, then you have not in fact done everything that you can do.
- [24:16]
Alex
Yeah. And things like, oh, well, I've connected them to a profile that's now real and I've done the bio and, you know, I've connected all the dots and.
- [24:25]
Carolyn
All this, connected them to a profile that's not real, as opposed to the fake one that I was lying about 10 minutes ago. So therefore you should totally trust me.
- [24:32]
Alex
Exactly. Exactly. They know all of that. You know, it doesn't even. It doesn't take Google's algorithm or much to figure out that if there was something written three years ago and now a new person has written this same piece of content, you're probably not helping yourself there.
- [24:48]
Carolyn
Just for the record, Google does not look at your site in a vacuum. What they know about your site is not exactly what is there right now. They know everything ever. It is like your permanent record from grade school. They know that you ate paste in kindergarten. They know that you had chickenpox and were out of school for three weeks in second grade. They know all of these things and they're not going to forget. And they will probably hold it against you.
- [25:14]
Alex
Yes, they will. So don't, don't do anything online is the lesson, which we've all failed at already, haven't we? So, next, new search experiences, but only in the eea, which is the European Economic Area. Maybe. Let's go with that. I mean, it's pretty reasonable. Yeah, it's pretty close in Europe. Right. They have changed the rules on what some Comparison sites are doing and comparison based searches in search results. This was via a court ruling, was it? Dmacc, I think dmac, something like that about you know, monopolizing results on search engines. So in short, if you are an E commerce person or you do holidays and comparison, that kind of stuff, all of these things are going to be better for you if you're again, if you're in Europe. Oh, it's the European Economic area. I was right. Thank you, Axel. Yes, that was a good guess. So yeah, that's good. And it makes things more fair and a bit more democratic on the serps. Nothing you as the audience will probably need to know, nothing you need to check change. Just that if you're in Europe it benefits you because of this ruling. So thank you judges. The next update is quite a big one from a techie point of view. So Google have now added Support from a schema.org rules around product variants. In short, if you've got a T shirt in different sizes, different colors, you can now have better control over the way that that structured data is provided to Google. In turn, that helps you with things like Merchant center and any other feed that goes with that and it makes searches much more of a better experience. Right, so you said you were searching for a red dress and it brought up yellow dresses and even though it's a small thing, it, I'm sure it frustrated you no end just to have that and puts you off a search and you may not go back to that shop as well because you saw that brand that puts you off.
- [27:24]
Carolyn
Well, it was frustrating and it was actually shoes and I was looking for a specific color, a specific variant of a specific type of shoe in my size and I provided all of those things and it pulled up a lot of sites where they had my size but they didn't have that color or they didn't have any of the things that I specified except they carried that shoe and they didn't even have them in stock, which was also frustrating. There are, you know, again this all goes back to Google wants to make a satisfying and happy user experience. And as a user I am happier when they show me pictures and take me to places that have the thing that I am trying to find.
- [28:06]
Alex
Yeah. And I think the other thing to think about is that as time moves forward, because this is being supported, if you don't do it, you won't be visible in those places. So if you are selling red shoes and you do not put red, you may not appear in results that are filtered with Red. Even though the technology can understand that it's a red shoe from reading the picture and all of that kind of stuff, it still may not because it's not. The structured data you have spoon fed to them may not be enough to get you in that search result. So that's very important to do.
- [28:40]
Carolyn
You didn't provide that specific information even though the picture matches, but somebody else did. So it's not that you're not necessarily in the results, but you're far back because other people did it better. So this isn't a case where you just have to be faster than the slowest gazelle to be eaten by the lion. Anyway, the point is you need to do it right because somebody else is going to do it better. And if you want to beat them, you have to at least be as good as they are, if not a little bit better and faster and more what the user wants.
- [29:14]
Alex
Yeah, yeah. And before we go on to the next one, we did have user 8c7b99a8b925b2c8 ask a question. What about service variants? They didn't say anything about it as very quick answer to that, but I'm.
- [29:28]
Carolyn
Sure when those come out, that will be a wonderful thing to incorporate.
- [29:32]
Alex
Exactly. If you do have service variants, my advice is maybe to just adopt the schema.org stuff now and then when they do click on the switch like they have with product variants, you're already there. So that answers that question, even though you didn't officially ask it in the question. I'll let you off because you have a good username. So the last one in the SEO news. Oh, you're going to want to talk about this, aren't you?
- [29:58]
Carolyn
I'll show you. Yeah. This is why I contributed. So Reddit and Google are expanding their partnership. Reddit is selling Google access to the entirety of Reddit's database. All of the user contributed information for $60 million per year. And Google is going to use this to train their language models. So they're going to train their AI using content from Reddit. This is incredibly frightening because the intellectual average of the, of the content that's on Reddit is interesting. We'll call it interesting. My concern is that if the AI is learning about humanity from the. I don't want to say garbage. I'm sure some of the comments and posts are very intellectually, you know, wonderful, but a lot of them aren't. A lot of them are silly memes, A lot of them are rude. There are, there are Entire groups that are dedicated to hating people or hating companies. And if the content that you're getting is all, is all that negative bile, what is the AI going to start saying about things? Okay, so let's look at celebrities. There are, there are hate groups about certain, certain, certain royals. Let's say. I know I wanted to say one off top of my head and I'm trying not to say the name. So let's say there is a royal adjacent person that everybody likes to hate. Everything that people say about that person that's in Reddit is going to get learned by this AI. And then if you ask the AI in the future, why does everyone hate so and so, it's going to tell you as though it is fact. All of these nasty things that people have said, whether they're true or not. And that can be applied to companies, that can be applied to politicians. If the prevailing political bias on Reddit leans one way or the other, the AI will eventually adopt that political bias. It's. I just don't know that this is a great. I don't know that I want the robots learning about humanity from Reddit. That's like having the robots learn about humanity from 4chan.
- [32:28]
Alex
It is. I don't know, I see the, not the flip side, because I get what you mean. Like, I think the term you're looking for is degen. It's degenerative language everywhere. And I get it. But unless Google is on a challenge to understand degen behavior and know what not to do and learn from it in order to provide a more objective view and understand what opinion is, I don't know. Because we are complex to learn, aren't we? Especially when we're hidden behind keyboards. So it's a very big challenge for them.
- [33:01]
Carolyn
Lisa just commented, AI is Cyberdyne 1.0. And the 80s kids have warned us. They have. And there's so many times that the politicians now are proposing things that either have to do with AI or controlling the weather that I just think to myself, like, I have seen this movie and it went badly for us. Why are you considering this? Somebody said something about we're going to shoot a rocket into the, into the sky and we're going to figure out how to block the UV light so that global warming gets stopped. That is literally the premise of Snowpiercer. And the world is plunged into an ice age and everyone dies.
- [33:37]
Alex
So yeah, and then you get arguments as well. You can't even, you can't even live peacefully in a small space. There's Only how many were there of them? A thousand. And everyone gets in arguments automatic, automatically.
- [33:51]
Carolyn
Horrible.
- [33:52]
Alex
Yeah. It's going to be interesting what they do with it. But then obviously this has had snowball effects on the fact that Reddit's having huge uptick in visibility. It's ranking for anything. And like, people are laughing at. One of the comments was like, just post something on Reddit if you want to rank. But like, away from that joke. That's literally what happened this month. Someone. It was. I forgot the brand.
- [34:12]
Carolyn
Literally been telling people that.
- [34:14]
Alex
Yeah. Was it hello, not hello Fresh. That's a brand in the uk. Hello, something. They wrote a post about how they don't like how Reddit is ranking above you, you know, independent sites and. And then someone posted it on Reddit and then the Reddit rank higher. It was like they were living their own inception. It's crazy.
- [34:35]
Carolyn
Well, but you're kind of at the mercy of the mods in those subreddits. The moderators can block people that they don't want in there. See, it's not like as a business, you can go into a subreddit that is dedicated to hating you and defend yourself because the moderators could very easily block you.
- [34:53]
Alex
They can be degens as well. I mean, we discussed it last month, didn't we, with taking away half of the SEO content. Right. But yeah, we'll see.
- [35:04]
Carolyn
Basically, we're speeding towards apocalypse and we're all going to die.
- [35:07]
Alex
We are, we are. We'll talk about speeding. I know that we can chat for ages about everything. So I'll go in other news very quickly. Schema has launched version 25. It adds support for certification. Google officially retires the cache link, which I actually found really, really useful. Just use Archive Dog instead. Google Search also having issues with recipe sites. I don't know if they're back. They kind of went away. That's part of the schema. Well, schema issues in the SERPs, a sitemap bug with hyphenated file names got found. I think it's been plugged and it had nothing to do with WordPress or Yoast XML sitemaps. You didn't need to worry. Yandex got sold for 5.2 billion, which sounds like a lot, but I think it was a lot less than what it was valued a bit ago or something. I don't know what happened, but it was.
- [35:55]
Carolyn
Compare that to the market cap for Google or one of the other search engines, that's really kind of peanuts.
- [36:01]
Alex
It is.
- [36:02]
Carolyn
I would like to possess but yeah.
- [36:04]
Alex
I don't think anyone needs to care much about Yandex unless you are based in or operate in Russia because I think it got bought by a big Russian player or something. So if you're. It's good if you're in Russia. Operating in Russia is probably, is the only thing I know. Google have also said they're going to clarify eat and quality rated guidelines documentation which will be interesting. I'm sure a lot of techies will have lots of complaints or interpret one sentence in the 79 page document differently and cause a massive Twitter explosion. Google confirms a fix to a week long search and indexing that happened at the very beginning of the month. It happened for about a week and I think it affected 1% of sites or something, but that was all plugged up. Machine learning is now removing fake reviews. I would have liked to have talked about this longer, but there's not too much to cover really. They've been able to get more sophisticated, identify fake reviews, fake reviewers, been able to deal with stuff on a very big basis. I think they've got rid of 43% of fake reviews in the last year or something.
- [37:11]
Carolyn
Good. Because a lot of them were fake.
- [37:13]
Alex
They are, they are and they were. And good, good riddance. And lastly, Google will clarify the use of AI to generate content for SEO purposes, which has been basically them thinking since they removed it away from the definition, remember from what's good content, make sure it's human, it's human produced and they removed it. And this is going to be the results of that. I think I got through all of that in two minutes. Let's crack on to AI news. Look at this video of three puppies playing in the snow. Isn't it lovely? And also not real, which is crazy. This is Sora that was introduced by OpenAI. It's not available to the normal people yet, but some high end influencers, videographers, designers, that kind of stuff to understand everything about it and see where the weaknesses are. But it's been interesting to see some of the examples. I don't know if you've seen any of them like beyond what OpenAI put in that page, you know, with the woman walking through Japan or something, were frighteningly good. They are. And a thing to note here is that this is, this is version one. Like this is the worst that you're going to see it, right? So if you were, they're, they're cherry.
- [38:34]
Carolyn
Picking what they're showing us. So it could be, it could be highly dependent on the inputs. It Receives, but we don't, we don't know that yet. But what they are choosing to release is clearly very good. But remember, Google faked one of their product launches, so we don't know that, that OpenAI has done that. But we also. I've become a very distrustful person, but.
- [38:59]
Alex
It'S interesting to see what people will do with these videos because you could actually now produce videos that are helpful, right? If you want to go back to Google and how it interprets them. You know, if you are talking about taking a dog out for a walk in the snow, why wouldn't this video make sense? Even if it has imperfections from a machine? I read a post about a video, a video that a videographer just dissects all the videos that they've released. And if you look at everything closely, you can see things like people don't walk exactly on the floor. You know, some people just have three arms. You know, as they're walking along, it's lots of imperfections. But again, this is the worst. And it's not even open to the put. So they're working on that right now. And remember that chat GPT only came out just over a year ago. So what's this going to be like in a year? That, that's, that's the crazy thing.
- [39:50]
Carolyn
The speed of change and the speed of developments is just. My brain cannot even.
- [39:57]
Alex
Yeah, it's, it's, it's, it is weird. And I'm, I also feel for video producers as well, because this will cut a lot of stuff out, but hopefully provide an opportunity for them to be so. Because they know all the terms, they'll know the perfect, the perfect command to put in. Whereas I would I just say put three dogs in the snow. But I know that a videographer will put in a small essay to describe the same scene, so that might be something to think about as well. What else has happened? Bard has now renamed itself to Gemini, which should have been done on day one, because I don't know anyone that thinks that Bard is a good whoever. I'm sure that the person who made the product name is no longer there, but it's a new name now. It's part of a new name. And this was announced the same day Asura getting released. And they also announced that 1.5 is coming out and there's a nice little graph showing off how much more powerful and faster and cheaper they are than OpenAI. So it's going to be interesting once that all gets released, how that's going to play. Out, but they're basically claiming that it wipes the floor of ChatGPT4. I feel like this is always going to be a case of, you know, upping each other, which is probably good for development.
- [41:19]
Carolyn
I mean, having a competitor does tend to push you to innovate and make your product better. So yay. Yay for competition.
- [41:27]
Alex
Yeah. I'm going to be interested to see what they do and how fast they do it as well, because obviously they've had to become part of the race. And talking about becoming part of the race, I like how all of these stories are connected. It's quite, it's Quite, quite good. OpenAI are developing an AI web search that makes sense.
- [41:46]
Carolyn
I mean, if Google's going to come after OpenAI's flagship product, it makes sense that when you get big enough and strong enough, you go after theirs. And if they believe that their technology is better and that they've got a better mousetrap, then they should enter, they should throw their hat in the ring and become a competitor. Because, let's be honest, Google doesn't really have. I know Bing tries, you know, bless their hearts, they do try. Everybody else tries. But Google still has like 97% of the. Of the market. So let's get somebody else in there. Let's see if we can disrupt this a little bit and make Google defend their position rather than just ride on their laurels.
- [42:33]
Alex
They should, because search volume. Barry Schwartz did a bit of, a. Bit of a survey on Twitter and I think on LinkedIn, I think he's combined a bit of data together that less than 20% of SEOs or people who participated in this think that nothing's going to happen with search results. And actually nearly 40% think of SERPs as changing by over 25%, which actually goes back to OpenAI making the web search, making it more interesting. They're probably reading this stuff as well, going, if I was in Google, I would be, I'd be. My heart would be racing a little bit faster at the moment, deciding what to do. But that's the search team, right? Not. Not the AI team. They can, they can, they can do their own stuff. But it's going to be very interesting to see how accurate this is in a year. I mean, we'll be talking about it, hopefully, and December, talking about what actually happened over the year.
- [43:29]
Carolyn
So I don't think Google's actually going to sweat about this because I think the, I think the way the serps are changing is going to make it so that the only way you can appear above the fold in a premium position is ppc, which is only going to drive up revenues. So while this is maybe bad for organic marketers, I think for from Google's perspective, this is going to be revenue enhancing.
- [43:57]
Alex
Well, yeah, I mean it's going to be very, very interesting. And again now, I mean the last AI is again about Google doing something that made something called Gemma. Oh, I'm just seeing that there's a play on Gemini and Gemma maybe, I don't know. This is like a lighter, laptop friendly, open model AI. I don't want to define the term open model as opposed to open source because there is a kind of. It's not open source. They've released an open license model. An open license model which has its own meaning, which is this is what I copied and pasted from their own site. It allows the free use but restricts it from being used in harmful ways, which is I guess that's nice kind of open source. But don't. You can't make adult like I guess.
- [44:44]
Carolyn
With WordPress.org that feels subjective. Like who defines that?
- [44:50]
Alex
What's harmful? I'm not getting into that now. Okay, we've got WordPress needs to do. There's not too much except for WordPress 6.5 beta 1 is ready for testing. And for anyone who doesn't do beta testing, this is probably a sign that you've got a few weeks until 6.5 is actually going to be released. Maybe a couple of months depending on what those bug fixes are. But. But there are some cool ones to look at, such as you can upload your own fonts really easily. There's plugin dependencies, which I think is very cool. In short, if you're using. Actually we can use it with Yoast. If you're using some of the premium products, you do require the core product in there as well, like free and premium at the same time. And they use dependencies. But some plugins can't even work without another plugin being there. So this creates and closes the hole on that. And as well as that we've got like cool overlay things for all of their themes and stuff like that, making all the blocks much easier, more intuitive and much better performing. Lastly, we'll end up with Yoast News. Well, Yoast News, we can tell you that there's a lot of improved AI results in the next version, especially in non English. There is a release somewhere on our blog, on our releases blog that tells you all the specifics of what has been improved. There'll be even more in our next release. And lastly in the Shopify app, there's some minor bug fixes and stuff, but it might be worth noting, even though it's not on a bullet point here, that product variant stuff that we did talk about before is coming in both our WooCommerce and our Shopify apps. I can't confirm the date, but it will be soon and we'll sure to let you know on X or Twitter or LinkedIn or whoever. So with that, where can you see us in upcoming events? So Carolyn, you're at Pubcon on the 4th, which is only a few days away. You're speaking, aren't you?
- [46:53]
Carolyn
I am. I'm Speaking on the 5th. I'm talking about WordPress AI empowering your SEO through WordPress plugins that are powered by. Wait. Empowering your SEO with AI powered WordPress plugins. I'm pretty sure that's the title.
- [47:09]
Alex
That's the title.
- [47:10]
Carolyn
It's a long title.
- [47:11]
Alex
That's a tongue twister. Slides, I'm sure. I know it's going to be good. It's going to be good though. And very useful, right? Very useful. Are they going to be shared publicly or is this an in house PubCom thing just in case?
- [47:24]
Carolyn
No, I think eventually they do release it, but it takes a little while to be released. But I will write up a a summary of my talk and provide my slides later probably on the YOST website. So I'll make sure every. I'll make sure you guys are covered, don't worry.
- [47:41]
Alex
Okay. And we'll Both be at WordCamp Asia representing if you're in the area. I'm also going to be at SMX munich just after WordCamp Asia. Neither of us will be at CloudFest, but I know that Carol will be there and I believe a few others, maybe T will be there. And I'll also be at Brighton SEO in Brighton in April. One that got missed in the middle was Women in Tech SEO. There's an event in London and I know that there's several of us there representing women in tech SEO.
- [48:15]
Carolyn
We have one in Chicago that's also not on the list at the Tastemakers Conference. And I want to say it's the 15th and 16th of March. It's in Chicago and it's food bloggers, so that should be fun.
- [48:28]
Alex
Okay.
- [48:28]
Carolyn
Hopefully I'll get a lot of. A lot of fun food.
- [48:30]
Alex
Well, let's chat. We'll chat about it again next week. So the next one is the 26th of March at 4pm Central European Time. We've got nine minutes because we always blabber or do we blabber or. There's a lot. There's a lot on right, isn't there? I'm sorry, Ninka.
- [48:48]
Ninke
It's okay, it's okay. I think we were all distracted a little bit by all the McDonald's and fries things going on and very hungry. Getting deep into all the AI troubles that are coming our way. No, it's fine. I think it was a. It was a great, a great update and I'm sorry about everyone experiencing issues with the stream. In the beginning of this update I was celebrating that we had the highest registrations. In a few months. It might be that Crowdcast can't handle our popularity on this, on this platform. So I hope everyone will be able to.
- [49:25]
Carolyn
I haven't heard any buffering, so I think the recording will be clean.
- [49:29]
Ninke
Recording is mostly. Is usually clean, so please watch the recording. If you missed a few important bits. Talking about important bits, we have a lot of questions and we have one question with the most, far, most upvotes and obviously it's about AI, because that's what people are worried about. The question is, will Google punish ChatGPT generated pages?
- [49:56]
Carolyn
No.
- [49:58]
Alex
Maybe not today.
- [50:00]
Carolyn
I don't think. I don't think they're going to be able to tell.
- [50:03]
Alex
Not yet.
- [50:04]
Carolyn
I think if you're good about how you write it, I think no one can tell.
- [50:08]
Alex
I would also maybe say as soon as they release that using AI for SEO content thing that Google did mention that they were going to do, read that with a fine tooth comb. If you are generating pages with ChatGPT and that and you are guilty of using AI content 100%, then read it. And then if there's anything they're telling you against, just update it a bit and make sure that it's got enough useful stuff in there to be helpful for them.
- [50:37]
Carolyn
So I think this is like using stock photography and photos that you shouldn't that, you know, don't use stock photography. If you take it and run it through a filter, it changes it a little bit. So be, be smart about how you use AI to write your content. If you're generating something that is bad or inaccurate or has silly mistakes in it, that is ultimately what's going to give you away. And the fact that it was generated by AI isn't why you're being punished. The fact that it's not good is why you're being punished.
- [51:15]
Ninke
Of course there are tools like Yoast SEO. If you Want to do some checks on readability and SEO and yeah, be helpful. Make sure you at least check all the boxes, I think. Okay, let's go to the next question. We have a bit of time. This is about cornerstone content, one of the principles behind also the Yoast SEO plugin and how inside structure. Of course, the question is how do I choose which web pages should be my cornerstone? Content.
- [51:52]
Alex
Go for it.
- [51:53]
Carolyn
I would say when you're evaluating articles or pages, look at the content is you're looking for foundational content. If a reader knew nothing else about your site or your topic, what is the most important thing that they should read to understand what you're about? That is a cornerstone. The cornerstone is foundation. So someone's starting from scratch. They don't know anything. What is the piece of content they need to know? What is the piece of content they need to know about this particular facet of your of your topic. So let's say you sell candy. They need to know about your candy making process. They need to know. Let's say you make jujubes. They need to know about your core jujube production process and what jujubes are. And they need to know maybe what your recipe is or what you include or don't include, but they need the core stuff. They don't need fluffy things that are there for entertainment purposes. Foundational knowledge. That is cornerstone content.
- [53:03]
Alex
I have nothing to add to that. That was pretty good. That was impressive. Cool.
- [53:10]
Ninke
Thanks, Carolyn. Let me see. We have one question about the helpful content update. I think a topic that is here for a long time already. Can we give a top three of things that people should do to recover from the helpful content updates?
- [53:30]
Alex
The first one's in the question, which is, which is the content itself? Like just try and think about it as you're trying to introduce someone you've never met in a restaurant, you're out with friends and you're trying to explain everything on your site, like would you go as far and if you were to read that page that you are even considering may not be helpful. If you read that and then said it out loud to that person, would that person actually learn something much? Would they find it helpful? If the answer is no, then it's not helpful. And also old content may be helpful right as well. So it doesn't mean just getting rid of old content. You need to understand what's actually not actually helpful at all. And if it isn't, merge it with something else or remove it completely. The second thing I would say is don't put too many ads on your page. So that I know we've all been to those sites. I'm not going to say any specific brand, but you know, you can only read like this little part of the screen and because everything else is ads and you don't care about anything else, that's not helpful as well. But yeah. Carolyn, you're going to add.
- [54:38]
Carolyn
Well, I was just going to add a while ago in Google News. So this is for the news algorithm, not the general algorithm. They had a metric where they looked at the amount of uninterrupted text that you had. So if we didn't have a certain amount of text that was not interrupted by ads, they would not include that article at all. So I know that they are capable of looking at spans of uninterrupted text and they like spans of uninterrupted text. So if you're writing two sentences and then have a couple big ads and two sentences and have a couple pig ads, Daily Mail, you know, that's not going to rank well.
- [55:21]
Ninke
That's a, that's a really practical way of looking at it as well. Look at the structure and what else is on your pages besides the content you put there on there on the topic.
- [55:32]
Carolyn
Cool.
- [55:32]
Ninke
I think we have time for one more question. I think we have a few questions with seven upvotes and I think it is nice to look into schema for a bit because this is nice topic, I think for also the people worried about technical SEO and all of that. The question is if writing content about some latest information, which schema would be more suitable, blog post or news article?
- [56:01]
Alex
Well, I know you're from news, Carolyn, but I don't know. I'll go from a more, not dumb but less experienced place. Is it worthy of being in the news? Right. If the answer is no, then it's a blog article. If there's, if you're writing about, oh, here's, here's a new, Here's a new attraction that's opened up in Paris and I'm writing about travel. To me that's a bit of a blog article. And the news article may come from the actual source of the building. But what would you.
- [56:30]
Carolyn
I think the difference between a news article and a blog art a blog post comes from the source. Are you a news outlet? Do you regularly write news? Does Google see you as a news outlet? If the answer to that is no, then it's a blog post and you are a blogger. I think if we, if we phrase the criteria in what do you think and do you think it's newsworthy? I think we're going to end up with a lot of things that are mistakenly marked news because especially public relations people and certain marketers think that everything is newsworthy. My store is having a sale. It should be in the newspaper. No, it shouldn't. So look at what kind of publication you are. If you are universally recognized as being a source of news or a news publication, then by all means make that a news article. If you just randomly, occasionally write about things or you're not considered a journalistic news outlet, I would call it a blog posting. Cool.
- [57:36]
Ninke
I think that's clear. It's helpful that people at least can start with themselves instead of the news article. But just think of your own role on the web and how you provide that.
- [57:46]
Carolyn
If you create a new article, it will. Especially if you install our news plugin, it will create a news sitemap and that news sitemap empties out every 48 hours. So unless you are making new news articles within that 48 hour span, eventually that will empty and it will become blank and the basically useless will not come and look at it.
- [58:08]
Ninke
Yeah. So it's basically useless to add news article schema if you're not actually publishing more news and updating stuff.
- [58:16]
Carolyn
Yeah.
- [58:17]
Ninke
Okay, that's it. We have more questions left, but we don't have any time left, so.
- [58:23]
Alex
I know a disagreement. I want to answer one more. Right. It's what.
- [58:30]
Ninke
You can pick one.
- [58:34]
Alex
This one. I'd like to delete some pages. Is it okay to use the 410 code for those deleted pages? Yes.
- [58:41]
Carolyn
Cool. Yes.
- [58:42]
Alex
Yes. I just asked 410 upvotes. I felt like an obligation to make sure I answered that before the end.
- [58:48]
Carolyn
Consensus.
- [58:49]
Ninke
Great. Thanks for that, Alex. Well, this was the SEO update of February. I hope you all enjoyed today's update.
- [58:58]
Alex
Thanks for coming everyone.
- [59:00]
Ninke
Bye all. Thanks for being here.
- [59:02]
Alex
Bye.