The State and Future of Search Q&A with Danny Sullivan, Google Search Liaison

Welcome, Danny Sullivan, to Crawling Mondays. I really appreciate your openness to do this interview. Looking forward to go through all of these questions with you. In case you don't know Danny, Danny is Search Liaison from Google, 7 years already in Google, and before you have a lot of background.

I started off as an SEO in 1996 and started a site that was called Search Engine Watch, and then later on started a site that was called Search Engine Land. As I joked earlier, after about 21 years of doing all that, I kind of wanted to retire. And so I did. And then I didn't really expect to be getting involved in search again, but, um, I got asked by Google if I'd be interested in what became this role. And I gave it a lot of thought and I thought, yeah, actually it'd be interesting to see how things work on the inside to see if I can explain things better to people outside about search. Been in that role about 7 years now.

Amazing. So you have all of this context and experience from different angles in search, and I believe that this also can bring a lot of different perspective on the challenges, the issues that you see in the day-to-day and what people tend to share with you to get feedback from. And one of the things that I see most of the people struggling out there is the challenge that many are perceiving that in the latest Google Core Updates, where you have tried to give more visibility to more helpful content, firsthand content, that is created towards the user, not necessarily just for the sake of ranking and getting more views and traffic, but because there's an actual value and benefit for the user and an actual product and service to offer that will benefit the user. But many have seen that as a consequence, affiliate-based websites or ads-based websites that are independent, that haven't had the capacity to develop Established brands have been hit, right? So there are a lot of these conversations going on in search about, oh, is it still worthy to start a new website, independent websites? And also, what does it say? And if it is really beneficial for the web ecosystem to not get visibility to this more of like independent websites that can provide more diversity and a different type of perspective in search results and to the final user. It will be great to hear. Your thoughts on this?

Gosh, where to begin? Because there's a lot in that. Our ranking systems aren't saying, are you a big brand, therefore you rank. And there's, there's not stuff in them as, to the degree that I understand, that I know, and I talk to lots of people and I see lots of things that says big brand. It would be a really complicated thing to even determine that, right? And also wouldn't necessarily be a useful thing because you could have something that is a big brand. but not necessarily be a good brand, right? So that's, that's not what's happening. But there's definitely this sense then, and I understand it and I'm not going to take away from it, that these large sites or these well-established sites, in some cases people feel like, oh, well, they just can rank for anything. Or I have a small site, which will be a small brand, by the way. Your small site is a brand. You are a brand in a lot of these cases as well. And I don't feel like I can compete. And is it fair that only the larger, well-established things can do better than maybe the smaller, um, well-established things? So, no, we don't, we don't want it to be only the very big things rank well. And I think in the last update, we did talk about how we were taking in a lot of these concerns and trying to understand how we could do more for some of the smaller sites, smaller, the The so-called independent sites and things that are out there as well. But the core of it isn't really whether you're big or you're small. The core of it is whether you have the most useful, the most relevant, most satisfying information. And our systems sometimes are not going to perceive that from some of these sites that are smaller as they should. And that's stuff that I hope we correct and that we've been working on. And I hope that we will improve working on. And certainly sometimes they can Maybe overestimated or over-rewarded some of the larger things. And then those extremes aside, there's a big chunk where everything works perfectly fine. I have looked at cases where people say you don't like small sites, and I am not taking away from any of the real concerns that are there because they are there. And I have been to, at this point, hundreds of these sites that people have submitted feedback on and looked at them. And I've said, you know, these sites are where hearts are in the right place. I see really good content there and wish they were doing better. Um, but I can also see co-occurring in some of the same queries that I'm given other independent sites that are doing well, but that, that are small sites that are doing the same kinds of things and that they are being successful as well. So I don't think it's the case that independent sites as a whole simply can't do well. Small sites as a whole can't do well, but we are certainly having some issues in some cases with some of these sites that I would like to see us improve on. Um, but then I'd also go back and speak to what I was saying about the whole brand thing. Brands are like an important thing. They, they, they say to people, especially if you are a good brand, whether you're big or small, that you kind of stand for something, that people recognize you stand for something. So this idea that, well, Google shouldn't show brands, that doesn't really make any sense. You would expect us to be doing that.

Yeah.

You wouldn't expect us only to want to be showing big brands, right? And in fact, you can meet any examples where this doesn't happen because you can do local searches and you get local companies. And those are not big brands. A lot of cases you can find local businesses that if you're in those local places, you recognize those places, you know that those are good places and they've done that. So what I would say is, if you are a smaller site that feels like you haven't really developed your brand, develop it. That's not because we're going to rank you because of your brand, but because it's probably the things that cause people externally to recognize you as a good brand may in turn co-occur or be alongside the kinds of things that our ranking systems are kind of looking to reward. The whole EEAT stuff that we've talked about, and I won't repeat all the words, I'll get them wrong. All right. Expertise, experience, trustworthiness, and authority. I think I've got all of them with it from there. That those are the kind of characteristics we're trying to see. And those are the kind of characteristics that you can often find with, with brands that are well recognized and trusted by consumers and people and so on. So, um, so I hope that helps, but, but like I said, I, I don't want anybody to think I'm saying, yes, everything's perfect. And no, some of these concerns that are out there are not valid. Um, I think one of the disheartening things that I've seen is that sometimes on social, people will share some of these stories of the site. And then others start to come in and pile on top of them and say, well, it's because you haven't done this and because you haven't done that and this and that. And like, I'm all for anybody getting constructive criticism and things that people can do that are productive. But in some cases, there might not be anything wrong with the site. It's us. As I've said many times on social, it just might be that our ranking systems need to get better in these kinds of cases. And I hope that they continue to do as our As our blog post said.

Also, one of the challenges here, a more disheartening type of scenario besides seeing your traffic sink is seeing which are the websites that have taken your spots. And that this might be, and many times you might end up like still not nice, but understanding the why behind. But in others, these are more generic type of publications that have very well-established brands. That are recognizable, as you mentioned, but their domain of expertise are not really about that particular topic. They also write about that topic, but they, they tend to write about everything lately just to get that type of traffic. And they piggyback on being a very well-established brand to, let's say, a little bit like get away with that a little bit, right? And yeah, I understand that you just said that, you know, that not everything is perfect, but I believe that that particular angle, right, is what is really disheartening for many, seeing which are the players that have taken that spot. Not necessarily your bigger industry competitor that you recognize that actually offers something even better than you, that you realize is, I need to become more comprehensive or add this layer of information or this better features, right? But no, it's this other more generic website that is not necessarily better.

I know what you're talking about. And I, same thing, I understand those kinds of concerns. That said, if you are a small independent website and you start branching out into other things, you wouldn't want us to be saying, and you did well and you were doing really good work, you wouldn't want the ranking system to say, well, I'm sorry, you started here, but you can never go there. Right? And that's something that happens if somebody says, well, this is this kind of publication. That's how it started. And that's how it always has to be. And nothing is like that in the world. Things don't stay. always static and be in that sort of way. So that's not really probably a, a good thing for a search engine to be coming at. Well, you can only ever be successful for that thing and never, ever, ever more. Again, that said, because that'll be the only thing people will repeat, he said everything's perfectly fine, and that's not what I'm saying. I struggle sometimes to help people understand some of the nuances that are out there in a world where we don't want to have nuances. We just want it to be one side, one side, and that's it. So, and I see all the nuances out there and I try to understand all that. So with what I've just said, there are definitely cases where large established sites see new opportunities and want to go into those opportunities and maybe are being motivated primarily because they see them as a good new search engine opportunity. We can get all that search traffic. So let's start doing that sort of thing. And to that end, I don't care if you're a large site or a small site. If you start doing things because you're primarily thinking that's a way for me to get a lot of search traffic, now you're starting to run up against our potentially scaled content abuse policy, which is don't do things because you just want to do a lot of stuff just to get the search traffic. If you're going to be branching out in these other areas, you should be branching out in these other areas because you actually think you have an audience. That is expecting that from you. And that this is the kind of stuff that you're going to be offering real good expertise. I don't care if you're big and I don't care if you're small. And with all that said, we can certainly understand some of these excesses that we've seen. And part of the reason that we have the policy, like we talked about scaled content abuse, because we took policies, we came up with new policy, is because we want to be able to act on these kinds of behaviors that really are more from I just want to get the search traffic type of thing. And it's not just that we then want to be able to take manual actions. We want the ranking systems themselves to evolve to the degree that they are sort of the first line of defense with these spam policies, that they already aren't naturally rewarding that stuff. So all that said, and I'm going to say those words a lot, all that said a lot, the standard things, we don't just have instant changes. We can't just, well, you do instant change, you roll out an update, but They take time for us to figure out what we're going to do and how to adjust it and how to test for it. And some of these things that we look for, how do you curb certain abuses or how do you improve their search results overall, just take time for us to keep developing. So those concerns that you've said, again, well understood, well heard. No one's going, that's, everything's perfectly fine there. And people are looking at how do we further improve in these kinds of situations.

I understand. And well, looking forward for the next improvements and to see a little bit like it potentially, hopefully like a more balanced type of approach into that because it's completely reasonable, right? Like—

It is. And I think it's not so much the balance should be big site, small site, and mix them all up on the page. The balance should really be, can we do a better job of showing the best content overall? Doesn't matter if it's from a big site, doesn't matter. Small site, and that's what we need to work on. And there have been cases in the past where small site was doing better, even though it really shouldn't have been doing better. Not because it was a small site, it's just our ranking systems weren't picking it up as content that really wasn't that great. And that's annoying to searchers. It's annoying to searchers if they get it from a big site. If we don't make the improvements so that the searchers don't feel happier with the searches that they're getting and the search results that are being presented, Then they don't want to search and then we lose. And you can say, good, I hope you lose. But if we lose, there's an ecosystem, as you know, that depends or uses or interacts with the search results. And we actually want that ecosystem to do as well. So it's, it's in everyone's benefit, we feel, if we are improving the search results for people who search overall.

This, I believe, that takes well to the next question, which is a little bit of Like this, but with a different angle is the platform type, user-generated content. We have seen so many more instances of the forum and discussion SERP feature and also Google ranking organically many more social platforms, Reddit, Quora, LinkedIn, et cetera. And you have mentioned many times too, in this conversation, questions that you have gotten in the past regarding users want to see more firsthand user experience in search results, and that is The goal and the intent of the giving visibility to these type of platforms, which is great. But then it comes also a little bit the challenge that this type of platforms also have that are huge. And MyTown's generic, right? The type of user-generated information that is shared in these platforms are not necessarily high quality or they don't have real expertise, let's say. They piggyback on the fact that have been created on this very big Authoritative but generic platforms that get this visibility a little bit more easily because of that. I think that the bigger challenge here is how a lot of the social information UGC right now is being shared in dark forests, in platforms that are closed that you can access. So we have seen how in this event where we were in, there was this session saying, oh, I have asked my teenage kid and she doesn't search in Google anymore and Google is for older people and I am all TikTok. So there's really this trend of especially younger generations going to social platforms. So I hear that for many different reasons, Google wants to give more visibility to social content and user-generated content. It really feels from a user standpoint, in my case, that there's this quality filter layer that is needed and potentially because of this gap and a lack of accessibility to these platforms' information and social signals that unfortunately Google cannot access because these are many times closed platforms, that you are not able to balance that or take that into account? How are you tackling this or how do you expect that this evolves?

We're showing content from hundreds of forums. It's not just 4 forums. It's not just big forums. I see it all the time in my own searches. I'm trying to fix something on my house or something weird has happened to my car and it ended up in the forum for, which I didn't even know existed, like for my HVAC system at home or for my car has a whole like enthusiast area where people are talking about, I had a car and I wanted to put a hitch on it, but you couldn't put the hitch on the car because it has this weird air model, air, air thing that's there. And I ended up in this forum just for people who have that car who were sharing in this. Someone shared an entire thing on how they managed to do it. And I was like, this is amazing. And it was amazing content. And I don't know if you want to say it wasn't expert content, but it certainly was experiential content that I found immensely helpful. And I don't think I'm the only one. And it wasn't from a big giant forum. So I think there's a huge value in what we've been doing is showing more of this kind of content. Because it is really helpful and it is not, we're just showing people what they want. And I, again, when I go back to nuance, because you share things and you try to explain to people and then people will take one little thing and throw it back at you like, it's what people want. And I've got that over and over. Thank you for everybody who shared it. I'm sure you can share it again. But that's not what I said in full. I said that people are trying to get more authentic information. They want to hear from other people with experiences. They value that type of thing. And we want to show it when it's relevant or when we actually think it's of good quality or that it will be helpful, just like we want to do with any kind of content that's out there. We show pages that are not UGC that are not necessarily expertise, but share experiences as well. You can pick any number of blogs that are out there where person may not be an expert on a place because how many times have they been to it? But they certainly are sharing their experiences of having been there. Maybe they're sharing it having been an expert as I am a travel blogger. So there's again, these sort of nuances that I talk about. It's really easy to go blah, blah, blah, black, you know, this side, this side, this side, this side. But it's not to me always these sort of binary kinds of notions. That said, I told you I'd say that said like a million times. Yes, we'll show UGC and it's not perfect. We'll show UGC sometimes from forums and it's embarrassing. And there's been no lack of people who will show us examples of that that's happening. And I'm sure that that will continue. But like with everything in search, you just keep working to try to improve it and adjust it and figure out how to get the better signals and how to get the better information that you're going to present. And, and, and hopefully give everybody a better experience overall. Um, you mentioned some sense of, well, we don't have all the signals from all these different places, but it doesn't matter whether it was a big place. And I don't even know that the big place has all the signals that tell it within its own little walled garden what is the most relevant thing or not. Because you can use engagement stuff, but that doesn't necessarily give you a sense of that's an expert type of thing. You could look at followers, but that doesn't necessarily tell you that something's an expert thing. So I don't know that that alone is necessarily the issue. In contrast, we have a lot of experience in trying to look at signals from across the web to all kinds of web content to understand, is it useful to show this content? Is it not useful to show this content? Does it make sense for this or not? So it's like a whole bag of things that we'll keep adjusting and working on. One of the things that comes up a lot, or had come up a lot, would be like with medical queries, right? And you could say, well, for anything that involves medical, you should never show anything at all. Never, ever, ever, ever, ever, ever. Because how possibly could you ever show anything that's expertise from a UGC forum? And yet, anybody who's ever actually suffered a medical ailment may find it's really useful sometimes to be able to connect to other people who have that same ailment. And understand what their experiences were like going through a treatment, or what their experiences were like dealing with an illness, or how they're coping with something. So you really, in my view, don't just rule out the idea that you just could never show UGC for anything. You try to figure out where you can use the signals and how you can figure out with the system to show the stuff when it seems to be relevant, when it seems to be helpful, when it seems to be part of An overall collection of search resources that you can present to people. So, that said, still going to always be work to be done. Always going to have examples. I'm sure as soon as this goes out, within 2 minutes, somebody will say, well, what about this, this, this, this? And I already acknowledge when you do that, that you'll be able to find that kind of things because search is a big thing. There's always going to be things that you'll find out there. What I can tell you is, well, we'll take those examples and we'll try to learn from them as well as we try to do with everything and just try to keep improving it. But it really is something that people are craving. And also part of the reason people are craving is because there's just a lot of terrible content out there. And there's a lot of terrible content out there that's been generated through content marketing. I'm going to create some pages for this client. This client wants to be found for all these 20 different topics. And again, I told you I've been to a lot of websites, not just, you know, in my career, but recently as well. And you look at it and it's like, well, I'm a local mechanic. So I'm going to write 25 different things and I'm an expert local mechanic to do it, except the mechanic didn't write it. The mechanic hired an SEO for who wrote it. The mechanic probably didn't even look at it at all. And then the stuff is not actually helpful to the mechanic's audience, which is the local business that they serve. It feels weird to somebody else who comes into the page to land on some of these types of things. Like, what is this business? Why are you showing me this side of content? And why are you telling me at the end? So sign up at this place to get your car tuned up or whatever. It's weird. It's one of the reasons why people are looking to have that more of it. And then at that point you go, well, it's happening because you caused all Google and we're just doing it and it's all your fault. It's like, fine. If you want to lay 100% of the blame down on Google and our ranking systems and what we've rewarded over the years, you can do that. And I understand that, really do. But then part of the reason you're seeing more of this UGC content is because people feel more comfortable with it, or they feel like it isn't got some kind of ulterior motive that goes with it. So that's part of the adjustment that we're trying to do, which by the way, will then lead to, and now they're all getting spammed and you've caused that. And it's like, yes, there's always going to be causes and effects. And then what you try to do is keep readjusting and keep trying to go somewhere because If all you did was say, this is it, we've done everything and now it's perfect and we never need to change it, it would be broken the next day. What I will say though, is rather than the generic content marketing that I talked about, which is just not what anybody really wants. I think even if you are an SEO and you were involved in content marketing and you land on someone else's content marketing, you probably go, ugh, right? So, and then you go back and do the same thing for your client. What I would love. Is if you're a local mechanic and you have a passion about fixing things, I would like to just hear your voice, your authentic voice about what car you worked on or what's going on or whatever it is. If you've got your passion that you want to share, that is what people are wanting to hear. And that's what we need more of. And if we can get more of that back from websites that are out there, Rather than the sort of generic content marketing, that should fill into with the UGC stuff as well. And then anticipating for the creators that are out there that have already been doing that. And I know I've seen your SARTs. I feel the, the heartfeltness in them. And we, I, please keep doing that because I want our system to get back to catching up with you because you're doing great stuff. You really are. Don't listen to the people who are telling you, if you know you're doing in your heart the right thing for your audience.

Yeah.

That is what you should be doing. That is, just stay at that. We are going to be trying to come to you with it. And I'm sorry, it may be taking us longer to get there, but that's what we're trying to do.

That's great. Thank you very much. And relieving to hear because this is where like the independent indie angle comes handy too. They can be like this specialist sharing about their day to day, et cetera, et cetera.

So to go back a bit more, we did the feedback form that you might've recalled after the core update form. And, and I wrote about this, by the way, I think on our Search Liaison post in LinkedIn or whatever. We had a new thing over there. But anyway, so we had all this submission come in and I went through every submission in some way that came in. And like, I think I said there were like 12,000 submissions from about 1,000 different sites that said that they weren't in the top results. And, um, some, a lot of them, like one site submitted 1,700 times in a row and then on Twitter said, Yeah. Yeah, of course we did. It's like, okay, still a lot. Maybe only 17 would've been enough, but I got the point from your submissions as well. But what was interesting is I went through all those submissions and categorized them in some way. This is from a creator who does reviews. This is from a travel blogger. This is from someone who's a recipe site. And I can see when I would go to these sites to categorize them, like What they were about. And so again, for those creators, like I understood, like, okay, we have a real creator audience out there. They don't think of themselves as SEOs. A lot of them want nothing to do with SEO, which is great because I don't want creators to have to even think about SEO. Like, not to take away from the SEO. Now the SEOs are going, he's saying SEO sucks. I didn't say that. It's just that if you're creating great content, you want to be focused on that. That's really the core of what we at Google say you should be doing. And that's what we want you to stay focused on. And we want to get better at rewarding it. But then I could definitely see beyond the creator group, which really kind of think kind of got lumped in with the SEO group and then the niche site group. And sorry, when I'm saying niche site group, I tend to refer to, I want to create a blog just to make a lot of money off of search traffic and I don't necessarily have my heart behind it and I don't really care about what's involved with it. And this will spark more debate, no doubt, but I think that creators who have their hearts in the right place, even if you think of yourself as a niche blog, understand the difference between that and someone else who's like, I want to create this thing and I'm going to flip it on a place. I'm just going to sell it off. That's all I really care about. Right? So you start seeing these nuances in these different kinds of audiences. And I'm like, wow. We've got to do a better job for these creators where their hearts are in the right place. We really need to do a better job for them. For the fly-by-night niche type of thing, sorry, probably not going to want to try to do a better job for you. Probably shouldn't. And in part because if we try to do that, we're not rewarding the creators who really, really have their hearts in it. But then you'd also have sort of the content marketing stuff that I said before. And I could see, well, while that's really your more traditional SEO type of thing, That this group of the creators over here don't care to do that, don't want anything to do with that. This group over here, the SEOs doing it, probably are disconnected from this sort of creator group. And this stuff is also an issue because when we start showing that, if we have better stuff from the actual creators, then more of them can't show. So anyway, as a long-winded type of thing to just say, nuance. There's a lot of nuance to nuance.

Yeah, I understand that. Definitely see that like subgroups of Different type of people involved in creating content from different perspectives and goals and quality and expertise, of course. Let's move to the future or what it's starting to become already, the present and the future. AI overviews. They were released to the public back in May. Then after a couple of weeks, additional filters were added and then we hardly saw them for a while. Then all of a sudden it was like a month and a half ago that they started to being showcased more also for non-signed-in users, for logged-off users. But still today, even if we can see many more instances and they have started to become tracked by third-party tools, because unfortunately Google Search Console is not showing them, which by the way is, please, please do to understand the impact engagement and positive and negative, right? Even to show that it's something that should have a good effect on websites, experiences, and traffic. But the point is that still today, many of these are very informational. I literally, 2 weeks ago or so, I started to see some shown for the gifts query permutation. So gift for Father's Day, someone looking for Christmas gift, things like that. And the snapshot was not so informational anymore, but it was including some listicles that when you click on them. They were showing product carousels. When you click on any of the product carousels, it will show the product knowledge panel with the information. So I could see that, oh, already instances of more commercially focused snapshots. Is this something that you plan to start like showing more and more? It will be great again. And what are the plans to give us tools and ways to measure the impact? Because I can understand how our stance and the stance of not necessarily us as SEOs, but from the website owners and business owners is like, Okay, this shown for informational content. Okay, this is top of the funnel traffic, not money-making traffic necessarily directly at least. But once that this starts to be shown for more commercially focused traffic or queries, sorry, it's a little bit different, right? So I would love to hear your thoughts on that, any plans that you can share, and also even tips.

Let me jump with all like the bad news part first. Can we get stats on things in Search Console from AI overviews? And no, and there's no plan to do that. And similarly, we don't do that for feature snippets. And that's another thing that people have wanted. And I don't really have a good answer for you other than overall the company views these all as part of search and not something that's worth breaking out. We have a lot of features. We have knowledge panels. We don't break out the knowledge panels. Here you were, had a link there and you had a link over there. In part, I think also there's a lot of sort of You know, the term is technical debt you take on. You launch something and then you have to maintain it for a really long time. And then if nobody's using it or things change, it becomes harder to deal with. So the AI overviews, as you just talked about, are changing a lot. So if you start building out something and then they start, you know, some reporting tool or whatever, and they start changing and they're still experimental or whatever, it could be in 6 months, this thing that you did, now you got to retool it and redo it. And they're all still. Kind of new or whatever. So that said, I diligently take back the feedback that people would like that kind of reporting. I totally understand it and I can even sympathize with it. I totally get it. Especially when people look at the IAO reviews and they go, I don't know what's going to be happening in the way that people used to look at featured snippets and go, well, I don't know what's happening there. But what I can tell you, and you probably remember, is when featured snippets first came out, there was a lot of concern about, well, will people actually be going to them and clicking on them? And again, anticipating that people will already react, that's largely gone away. People largely do not get concerned about the fact that they're a featured snippet and they don't attract visits. They actually get concerned that they're not a featured snippet, they're not getting the traffic that the featured snippets actually drive, which is why, as SEOs know, if you search for things like featured snippets, all you get are a bunch of people telling you how to become a featured snippet. But understand it. I take the concern back. I'll take it yet again. Totally get that desire to have it. So then let's talk about, well, where are the AIOs going? It's like, I don't know where they're going. Or if I do know, I can't tell you until they're like launched, right? First of all, as a reminder, there's 2 ways you're going to get AIOs. By the way, fully endorse the word AIO. I never said that publicly, but internally, the day they went out and I said, because we have our chat rooms and we're like, I'm calling these AIOs because that's just too much to type. So I was really kind of like, oh good, everybody's saying the same thing externally. Excellent. So yeah. So anyway, there are 2 ways you're going to get them, assuming you're in the place at all officially you're supposed to get them, et cetera, et cetera. One is through Search Labs. That's what everybody did in the beginning. And that takes you back to the SGE days. And if you go into the Search Labs and you do the opt-in and everything else, you'll get something like AI overviews. And it's got something else after it. I can't say it. It's like labs or whatever. And I think you get the little beaker that'll appear and that's your sign that you are getting something very, very, very, very experimental. And anything that you see in there could be something we're testing. Might not necessarily launch to everybody. And so, and I know I'll get back to what you've seen, but yeah, we're going to experiment with all types of things because there's potential with this AI technology to do both. The kind of generative AI stuff that you've seen, as well as some of the organizational type of stuff that you've seen. Like if you can say to it, as you saw with our organized search results, if you could say to it, well, someone sent a search and from that search I could understand there are like, they searched for Iceland where we're at. And from that search, I can tell people from Iceland typically who search for that may be interested in 5 different kinds of things ranging from tours. To places to stay, to Northern Lights. Could I get all the results from all those things and then organize them in a way that somebody can get satisfied without having to do more queries? Because that can be frustrating and then better hone in on the things that they're actually looking for. So anyway, so you're in the labs and you get super experimental stuff and who knows if it launches or whatever. If you're not in that, and I actually turned that all off, so I'm just seeing what people are seeing. You know, externally, then you're getting the AI overviews, which themselves are constantly evolving. And, you know, when they really kind of lock down to something, we tend to do a blog post and say, okay, everybody, this is kind of there, whatever. But at the same time, just like with search generally, we're always experimenting with things there. Even if you didn't opt in, you know, you do something on search and you go, what is this? What is this? And then you say, Barry, is that new? And he goes, not new. Or he says it's new. But, you know, you do, you do this sort of search and you get these different things in regular search. So you're going to, that's why you're seeing all these things being tested. And maybe some of them will make sense for us to do. Maybe it will make sense to have carousels and present them that way because you're using the AI technology to organize the information as opposed to providing overviews of type of information. Maybe we'll keep calling them AI overviews. Maybe we'll get to some point where people don't necessarily need them to be labeled at all. And it's just like, I don't know, it's just, you gave me search results and I'm kind of happy with it. That's the story I told today where, and you weren't there so you didn't get to hear my story, but you can hear my story now if you want to hear it. But I had somebody going back to my HVAC system, so clearly I couldn't figure it out. So I had someone come over to figure it out for me, the person who installed it. And they were looking and trying to understand something to hook this up. And they did a, um, search on Google on their phone as I watched. So that was like, whew, that's good. And they got an AI overview and they started going through and looking at that. They also looked at some of the forums that came up there. And it was fascinating to me because I'm watching this person who didn't even know I worked for Google, but they didn't go, oh, what's this AI overview thing? I, to them, it was a natural part of the search results page. And they looked at it as they looked at everything kind of else.

Thank you very much for that in-depth answer. No, no, I love it. I love it. No, thank you very much. It gives a lot of context. On, on the thinking and, and the criteria that you, you use around things. But there are 2 things that really stood out to me from what you mentioned, right? Like all the testing that you're doing, which makes complete sense, but also that you don't want to provide yet access to granular information about the, the impact of the AIOs because you see it a little bit of like a, something similar of features and impacts. Right. And you don't give those for featured snippets. I have to argue though, that the way that the users interact with AI overviews are, is much more engaged and the potential impact because of how many pages are referred from it. So it's less straightforward for a website owner or a business to assess the impact. And also, especially given that you are trying so much and testing so many formats and things right now. Maybe they want to opt out, which brings the other thing, not only to bring a way to monitor them in particular, but also a way to opt out if you want without having to opt out all of the snippets, but only from that. Any plans to do that eventually? It's giving control to website owners and business owners. We are at the end of that bridge here.

I totally understand that. I totally, totally do. And it's the kind of thing you saw with, say, like feature snippets as well, where people say, I would like to have this sort of control. And it's potentially something that you have for all the kinds of search results. But in general, we view search results as all search results. And, you know, you're part of search and it gets complicated when you start saying here, but not here, not here, not here. So to date, these kinds of things, when they've come up, have been dealt with in terms of dealing with the snippet controls, right? That, that is your control. So you don't want to be part of featured snippets. Your effective opt-out from featured snippets is to not have snippets and or be more selective in shortening the amount of snippets that you display, right? Because you can set max snippets. That doesn't mean you don't get snippets at all in search. It just means you get a shorter amount. Or you can, if you have information you just really don't feel like should be snippeted at all, you can data snippet just the text on the page, right? That doesn't mean you don't get snippets. You have all the snippet length you want. It just means I don't want this particular content to appear in snippets for whatever your reason is. You have that kind of control with it. So today that's been extended with how we are dealing with it with AIOs for people who ask the same kind of question. I don't really want to be in AIOs. Okay. You can make use of snippet controls, but they already do that with featured snippets as well. But you're not just purely binary. You don't have to turn on all of your snippets. You could. Use data no snippet. So say you have paywall content and you're concerned that, well, I don't want my paywall content perhaps showing up in an AI overview because the AI overview system tries to do an overview and it's predictive, but then somehow replicates what might've come out of it. Data no snippet, that's taking care of it with it from there. The other thing you could try experimenting with is a shorter max snippet, and potentially that might drop you out of. the AIOs and the way that it works from the featured snippets as well. And if you're already doing that for featured snippets, you may be able to tell if it's making that kind of a change with it from there. Ultimately, the only control you really have, you're just like, I don't want to be there, would be to say no snippet. And I totally get that is a really, really, that, that really gets you into that binary type of thing. But you do have those other things you can try first. Maybe at some point. We'll say, yeah, we should let people like toggle here, toggle there, and kind of go with it from there. But it's, it's just not an option that we have.

Your plans in Europe for AI over—

Oh, I don't have any. I literally have no idea.

To start wrapping up, because it's taking a little bit longer. Future of search and SEO. When someone asks you like, please finish the sentence, what is the future of search? What will you say?

No idea. Other than it's always going to be changing. I mean, it is always changing. Like, what have I— it's like I'm coming up on 30 years of doing search now, right? And every time you think that's how search works, it changes. Big changes that I can recall were things like when we moved from desktop to mobile. We went from an environment— there's a great picture we have where I used to do a presentation on how search works. We all, various people at Google have done this presentation on how search works. But one of my favorite things is this opening and it's a computer, like an old computer, CRT, big box next to it. And that's how we used to search. You went home or at work, the modem went and you waited and you did your searches there. And then the next photo is like a father on a bicycle. His son's on the back holding a phone. And that's like how we search now. Like we. This shift to always able to search for anything you want whenever you go. So we had that shift. We had this big shift towards voice where you could just ask and get responses back, which I thought was amazing. And I think we're going through one of these shifts now where we're looking at this power of generative AI. And I think a lot of people recognize there's something there. And we don't know what exactly and how exactly it's going to go. But my hope is that we're going to keep looking at that and how do we harness that in a way that helps people better access information, including accessing websites that have that information, because none of this works if the websites and the ecosystems that's out there itself isn't supported. So.

And what about SEO?

SEO is dead.

Again?

Yes.

I thought it was already.

Every year it dies. Um, and yeah, that was the other story I told today, like being old, you know, I started writing about all this stuff and it was either '95 or '96. I can't remember. But like, I remember it was either '97 or '98. There was a forum and it was like SEO, I think calling it, I think SEO is dead in '98. And every year another one came along and, and I've written about it in the past so many times, like SEO is not dead. It's not dead. It's not. Because when I used to talk, when I was outside of Google, I used to say, Your job as an SEO is not, I'm going to make content show up from a webpage on a particular search engine. Your job as an SEO, if that's your profession, is that, in my view, you understand how people seek information, and that's very diverse. And then you understand what are the ways to help them access the information that you have. And hopefully, and speaking especially with my Google voice here, that would be helpful to them. Those are fundamentals that don't change. Those are fundamentals that happened before search engines and you just had Yellow Pages or phone books, right? Or however they go back to, you know, talking to people or word of mouth. So I don't think SEO goes away at all. I think if you're a smart SEO who's always thinking ahead about how are people accessing information and how did that information get gathered and what are the things that are important to appear there, then you're fine.

Connecting the user with the product, as independently of the platform.

Yeah. So that's it. So just another thing too, I just want to end on like, there's a lot of feedback that's out there and that is incredibly welcomed. And I am never upset to get the feedback. It's nice if it's not mean-spirited, but okay, if that's how you want to present it, go with it. I mean, I'm always looking to find what's the productive thing I can take away. And if I've got real examples and I've got real things that are out there, I will take that back into the team and it can be compelling and people look at it. I was on a recent session we did. I loved it. We had 3 examples. It was like a query debugging thing. And I got to watch the engineers going through like, why is that happening? Why is it? Maybe we should be doing this. It was fascinating. So that's all, all really welcome, but I also want, you know, people to feel like You, you are heard. You are, you're definitely being heard. And there's nobody suggesting that everything out there is perfect. I think it's fair to say everything is not imperfect either. And that there's a lot of searches that happen that are very successful for a lot of people. And there are a lot of sites that are very successful and continue to be successful with it. But, you know, for people who've been impacted, You know, especially the, the sites, the HCU sites or whatever that are the ones with the hearts in place, right places that I've talked about. You know, that really came through on off the recent feedback. And I really do feel that we're going to keep, and I hope that we're going to keep trying to do better for that special group of things, because I think that these authentic voices that people want to hear, that they're like, well, can you give it to me more on a UGC or that it is on the web and it is out there. And I want to see us rewarded even more for the people who are doing that and sharing those voices that they're. That they are getting that. So, you know, maybe it's not worth much at this point, whatever. You can continue to share. What I will say is sometimes I'm just being shared the same things and I'm being asked the same things over and over. And that's not really, there's nothing productive further I can add. Like I've, someplace I have, like I've said all I could say, I've acknowledged all I could say this. So it's not that you're not heard and it's not that those things aren't there. I've taken that back and you can keep sharing with it from there. I just want people to understand that they really are. It is really being taken to heart. It really is being there. And more than anything else, I think I would say to those who are trying to figure out, what do I do? As I said earlier, what do I do? I feel like I've been doing everything right. I'm getting all this advice that's coming at me. A lot of it's unsolicited and all these people who tell me I got the secret way that's going to figure it all out. Just keep listening to your heart and doing what it is that you think is the right thing to be doing. Because I did this talk last year where I talked about Our ranking systems are trying to reward great content that's made for people. And if you feel like you're doing that, then we're going to try to catch up to you. When you flip it around and say, I'm going to try to anticipate what their ranking system is doing, then you're behind what we're actually getting to. And so that's, that's where to go. And it is not, oh, I put on my page that I'm an expert. And it's not that I, you know, changed the date. Now it's a new date. It's not any specific type of thing. It's just that you're doing something, whatever you're doing, because you think it's the best thing for your people that are coming to you. Whenever you're in doubt about anything and you're like, I don't know if Google will like it, just ask yourself, do I think that people coming to my site will like this? Do I think if my URL is this way, that's what people will like? Do I think if I put this here on this part of the page, people will like it? Do I think if I lay it out this way, people coming to me will like it? That's what we want to be liking too. So that's what you should be doing, and that should be your grounding, your North Star, whatever it is you go with.

Thank you very much, because I, I really think it's a nice way to wrap it up. And I have to say something, right? From an SEO standpoint, this is something that really resonates, but because it's how I advise clients usually and how I think if you want to be ahead of the curve, you should do. If you're only doing whatever the best players already are doing, you're always going to be behind. You do something because you know that it's going to be ahead of the curve and it's going to be like best class for your users. Even better if others are not doing it yet because you're ahead. You should be rewarded for that, right? So ideally, hopefully. Thank you very much, Dani.

For all your help and for having me and for all the things that you do to try to help people beyond just yourself.

Oh, thank you. Really appreciate it. And hopefully you have liked today's Q&A and remember to leave your comments and looking forward to the next edition of Probably Mondays. Thank you very much. Thank you.