How to show up in AI answers (not just Google links)

The only constant is change. We've seen world ending stuff a few times over the last 30 years and thus far the world has not ended. I'm like, don't freak out, right? Do what you've been doing. Make some minor adjustments and you can win. If I can relate the content of my web pages to entities that are known in the knowledge graph, then I increase the likelihood that I'm going to be understood as relevant, specific, and direct. If you can do SEO, you can do AI SEO. If you understand the basics, you'll re you'll start seeing content differently. So, Will Scott, co-founder and CEO at Search Influence, welcome. Thanks for joining. And please introduce yourself a bit for our guests. Hey, thanks for uh asking me to take part in this. I've been uh I've been pretty obsessed with AI and AI content development. It's I I tell the story. I was on a plane on my way to Colorado to go skiing and I'd heard about this whole chat GBT thing, right? Right. I knew about GBT prior in other applications, but I think it was like December 21st of

2022. And I'm doing what I always do on planes, which is like watch trashy uh action movies. And so, you know, I'm watching one of the Fast and Furious series. And I I was like, what was the beef between Vin Diesel and The Rock, right? Cuz this was a big thing when they did that one Fast and Furious together. And so I was like, well, I heard about this Chat GBT thing. Let me see what it has to say. and I got a really good concise answer, right? And then I went over to Google and I typed the same thing and I got 10 blue links of just like total junk. And that for me was the pivotal moment in like is this going to be a big deal, right? And so from that point forward, I would say with some level of certainty that not a day has passed that I have not been engaged with chat GPT sometime. And and now it's to the point with with memory where I feel like it really gets me, you know, like it starts answering questions in ways that in ways that conform to views that I've expressed about the way content ought to be. So that's why, you know, that's my experience with uh AI

content, AI tools. But um but in real life, uh I I co-run an agency called Search Influence with my wife who is my business partner, and we've been doing this for almost 20 years. We started in 2006. Uh I personally put my first website online in u like 94 or 95. So I've literally been working on the web for 30 years which is a little bit scary. And um it means that many of our employees of whom there are about 30 uh in and around the New Orleans area and you know a couple in different places um many of them were not alive at the point in time when I put my first website online. So we work in we work in a number of industries. Uh we're pretty focused right this second on on higher ed healthcare and hospitality. And what's really interesting to think about is that in a couple of these, particularly hospitality, they've been super focused on creatingformational content for the last couple of decades.

And and what we're seeing is that thatformational content is is pretty much dead. So with that, I'm going to talk through some ideas and tactics about how you can with your existing content future proof for this uh up andcoming AI world. Great intro and I suggest jump in straight to the topic. So the stage is yours. Super. Uh so okay, just uh for anybody who hasn't seen one of my presentations before um they also tend to pretty much always include uh robots because you know I'd like to share one of my favorite quotes. The most important thing in business is sincerity. If you can fake that, you've got it made, right? So, I'm sure you all understand how this might be applicable to AI. U but seriously, uh this quote is just as true today as it was 2500 years ago when it was coined by um this philosopher and that is the only constant is change. I come across some people who are saying things like, "Oh, I don't use AI." And I'm like, "Really? Are you sure about that?" because pretty

much every tool that I use has integrated AI and I think a lot of folks don't even realize that when they go to uh Quickbooks or Quick and Online or um or even Google uh that they are interacting with AI and I frequently have conversations with normies you know people who aren't in our industry and I say you know are you using these tools and they're like no I don't really think so and as we know and as Google IO showed a couple weeks ago they're using these tools they just don't know it they don't realize ize that this is something new. They just think it's part of the experience that they've grown accustomed to. Now Sam uh Sam thinks that his tools are going to replace all of us marketers. The quote was something like you know in the next 18 months all of the work that was previously done by marketers and strategists and blah blah blah is going to be done quickly and easily by you know open AI maybe. We'll see. But you know what happens in times of turmoil in our industry is that we polarize right? So, so if you hang out on um SEO LinkedIn, which I do far too much, uh you'll see this battle going

on, right, where there's the there's the camp of this is really just SEO, and then there's the other camp, which is like, "No, it's totally different. It's something completely new." And um I sit somewhere in the middle. You know, I uh I my I I tend to respond on LinkedIn with like, you know, quick little statements like, "If you've been doing good SEO, it might not be that different, but you know, I'm not sure everybody appreciates them as much as I appreciate leading them." Um, but it seems to me that we need a map, right? We need to be able to figure out what's coming. We need to understand, of course, that this is real, right? In the first couple of months that chat GBT was live, gained 100 million users. It's the fastest growing consumer application ever. Between 50 and 60% of all organizations are using AI um generative AI adoption and this quote, this um this stat is a little old, right? So, I would bet that if they refresh this stat, it's closer to 50% now. And then um this is the one that that I think a lot of SEOs are kind of freaking freaking out about you know and was alluded to in the last session but and in Joys as well but

there's like if you watch analytics there's a huge decline in traffic right and I I love that Joy said I don't even look at impressions I was like that's great right I mean because the because the reality is what's an impression anymore who even knows right and similar to the increase in CTR if you're higher on the page when you're one of nsighted properties in an AI overview or an AI search. Who really cares? Does that impression even matter? So, I uh I taught a um you in the, you know, in the interim there's a promo about the the SE ranking um AI SEO course. I taught a master class for SMX uh back in the spring and we're going to do it again in the fall and October on generative engine optimization. And this was one of the comments that came from one of the students that I found really really helpful. And I don't know, I think having been in this industry, and I'm sure this is true in any industry for a long time, you often wonder what's the value of sticking around for a while, right? And for me, I think that value is perspective. Is that a lot of these things that people see and they're like, "Oh my gosh, this is world ending." We've seen world ending stuff a few

times over the last 30 years, and thus far, the world has not ended. But so this is the quote, right? She says, um, she entered SEO in the last three years, so she's like an SEO baby. AI has really overwhelmed her. I know that this is a a female person because I know what her name was in the comments. Um and she says after, you know, she she had this feeling like she was just too late, right? And I'm like, "Holy cow, these technologies aren't even 3 years old and you think you're too late. Forget about it. There's so much headroom." So thankfully, we got to the end of the master class and she says, you know, I'm seeing the light at the end of the tunnel. And and I think that's the perspective we should all be taking is that there is a light at the end of the tunnel. And if you understand the basics of SEO, you understand the basics of AI, SEO, GEO, search experience optimization, one of my favorites, um, only because of the, uh, acronym, you know, SXO. Um, you can figure it out. But the idea is like think critically about the way the machines retrieve information. The other thing that I've been pointing out in my piffy little responses on uh LinkedIn lately is that

you know when Google launched the knowledge graph when Google went from as they say strings to things i.e. entities 2012 right that's 13 years ago and we're like oh my gosh 2 years of chat keep and it's all over. No Google's been doing this for 13 years. So chill y'all or don't I don't care really you know do what you want to do. Uh but so I don't know how that got in there. Uh maybe y'all have seen this and I think this is really instructive. This is a a very uh this is a very telling interaction when you think about the coming world of AI agents in particular. Thanks for calling Leonardo Hotel. How can I help you today? Hi there. I'm an AI agent calling on behalf of Boris Starkov. He's looking for a hotel for his wedding. Is your hotel available for weddings? Oh, hello there. I'm actually an AI assistant, too. What a pleasant surprise. Before we continue, would you like to switch to Jibberlink mode for more efficient communication? [Music]

Thanks for calling Leah. So that's what's coming people. Uh we're going to be taken out of the loop, right? Who is it that's talking? Jeanluca Furelli keeps talking about this other post from Google about the messy middle where we've been accustomed to this. We've been accustomed to this this um salesunnel, searchunnel, you know, user funnel forever which starts with awareness and ends with purchase or if you're doing it right, ends with, you know, repurchase, retention, etc. Google published this this um this blog post in which they used the phrase the messy middle where we often know where our awareness is, right? Where the first touch is we pretty much always know where the last touch is, right? And I think Google's gotten us really stuck on this idea of last touch attribution. But what we don't really know is what's happening in the middle, right? Is it do I become exposed to a product idea on Instagram for instance and then go do some other research online and then you know blah blah blah? Like there's all of this stuff in the middle that we don't really understand. And my position is that if you look at the example of the two computers talking to one another, the computers are going to take us out of the middle. The we're going to we're

going to show up and we're going to say, "Hey, I need this thing." And the computer is going to go do the thing, right? It's going to search a bunch of stuff. It's going to do its agentics, whatever. And then it's going to give us a product. And that's why we need to be thinking not of well, not only of content in the context of humans, but also of content in the context of does it read as data. So just like keep that in mind as we go forward. So what's different? If you think about traditional SEO, it was uh my my version of SEO 101 is keywords, content, and links, right? Uh what are people searching for? How well do you match those? How well do you support them with other similar phrases? How well is that expressed in your content? Do you have links that are um relevant and uh expressive of the intent of the user? And on the AI search, it's entities, right? So, back to that uh Google um strings versus things. Um semantic relevance. Uh there's a one of my favorite one of my favorite um metrics in AI SEO is the idea of salience. And salience is how important is a phrase? Now, we used to game this when we were

talking about keywords, it's like, do you have the keyword in the first paragraph, last paragraph, and three times throughout the body? Great, right? So, salience is like that. If I say, you know, u Joe's pizza has the best pizza in town, then that piece of content is probably pretty salient for pizza as an entity. Hopefully that makes sense. And then citations, which are trusted references. The difference is that we no longer need a hyperlink. Right now, those citations can just be text references. And that's where I think we become liberated to a great extent is if I don't need a link, how hard is it to get a mention? And we've been talking about UNL mentions forever, but the rise of the AI search engines makes them much more valuable. So I I like to use this example. Uh this is a um you know, thinking about conversational search and zero click search and and what's what's interesting over the last 30 years or 25 years of Google if you will. What's really interesting is that every year queries get longer and now as we get to AI search where the queries are rewarding natural language, they're

going to get longer still. And when you think about AI mode and this whole idea of like query fan out, then the then the machine um and for anybody who hasn't read the whole query fan out thing, think of it as if you were you did a search and then you clicked on every single people also ask and then you clicked on every single people always so asked for all those Google also asks and that's effectively I know I'm oversimplifying people are going to tell me you know that I'm dumb but that's effectively what this whole query fan out idea is. you go from uh I'm thinking about this idea which may be a big idea to a whole bunch of expressions of that to get to the natural language. So you know on the left we have fulls size SUV. I apologize I'm an American. We drive big trucks. On the right I have you know best fulls size SUV with self-driving which I happen to own one of. In the traditional search right you get this entity carousel with people also ads. This is for fulls size SUV. In the more natural language search, you get an AI overview which tells you the Cadillac Escalade. I don't have an Escalade.

They're nice, but I'm just not a Cadillac kind of guy. With Super Cruise, which I do have online, is the best uh technology. And then you get the from sources across the web. And and so this gets really interesting when we think about entities, right? And the entity carousel and how you win the entity carousel. You win the entity carousel. And I'm starting to see a lot of gaming of this by getting into these top X lists, right? And so, as it turns out, the BMW X7 is not nearly as self-driving as my um GMC Yukon, but because it's on these lists and because people like BMWs cuz they're, you know, sexy and cool, it gets into the edit carousel. So, just so we're clear, I'm not saying go create a whole bunch of uh guest blog post listing yourself in a top X list. Don't do that. That's spam. I am saying from what we can see in the search results, it might work. So what's the value of a number one organic ranking? Uh our client ranks number one organically for the phrase things to do in New Orleans.

Uh one of the things and I don't know SE ranking may have this feature as well, but um in uh in the tool that we use, you can see the pixel depth on the page. Uh the the actual so pixel depth, right? So how far down the page do you have to go before you actually see the result? And in this case, our number one organic ranking is good for 2,300 pixels from the top of the page. So, I have um you know, I have an AI overview and then I have uh some some ads and then I have a couple of different versions of the entity carousel and then I have people also ask and so again, you know, that number one ranking about which we've been really proud and our client who's very focused on impressions not in search but on their site, it's not worth a lot. So, how do you get sighted? the um there's the three pillars of AI search SEO as we think about them at Search Influence are entities. So basically, if you want to dumb it way down, align the keywords that you were thinking about when you first wrote the piece of content to entities as they're understood by the uh AI search engine. And so just a dumb like how do I

understand what an entity is trick does it have a Wikipedia entry? Does it have a wiki data entry? Does it have a Google knowledge uh ID? Right? So these are all knowable things. So when you are able to understand that um and I one that I use frequently is that the actual Wikipedia entry that is most relevant to the phrase higher education is the Wikipedia entry for tertiary education and and that's because I think in American and I think higher education but in other parts of the world in other versions of English even it's not necessarily higher education but tertiary education is definitive for that education which happens post high school. And so it encompasses a whole bunch of other stuff. But if you want to get to the entity, the recognized entity relative to higher education, the top level of that per Wikipedia is tertiary organiz tertiary education. Semantic relevance. Do you provide enough relevant content to help the AI engines disambiguate?

Because the the the worst problem is is like I I like to think of it as Apple problem, right? What's an apple? it's um it's a it's a tasty fruit or perhaps it's a um piece of technology, right? So, so without the relevance, you don't necessarily understand the disambiguated entity that you're trying to represent by way of the content. And then finally, citations. So, without being too reductive, this looks a heck of a lot like keywords, content, and links, doesn't it? So, this is where I like this is where I come in and I'm like I'm like, don't freak out, right? Do what you've been doing. make some minor adjustments and and you can win. And arguably when you think back to the fact that the uh knowledge graph was launched in 2012, should have been doing this for the last couple of couple of you know couple of years even before chatbt made anything about it. So when you think about content optimization in the AI world, it wants to be entity rich, right? So you want to tie it as best you can to known entities. And I I I like to think about this from the perspective of learning theory, right? They say that we

learn new stuff based on our ability to relate it to something that we've known in the past. So there's very few things that we learn uh that have appeared like out of a whole cloth in front of us. Um there's not a lot of epiphany but but it's by relationship to things that we already know. So if I can relate the content of my web pages to entities that are known in the knowledge graph then I increase the likelihood that I'm going to be understood as relevant specific and direct. And we spent the last, you know, 20 years trying to write, you know, uh, emotional, uh, gripping stories. But you know what? Terminator don't care. You got to give it the data it wants in clear authoritative statements. Our last presenter uh, referenced CindyRum and and her idea of fraggles. Well, fraggles are Cindy's representation of what we've always thought of as passage rank. So, it's been years and years that uh, Google in particular is not necessarily ranking cold pages. is ranking passages. And the reason it's ranking passages is because in order to understand the world from the frame of an AI engine, it has to be

in chunks. It has to be in chunks that can be turned into math that can then be compared to other chunks. Um, organized. I'm going to show an example in a minute, but one of the things that uh AI loves and and Google's loved this for a while too is um like how-to lists like step one, step two, step three. The original HTML was based on, you know, uh HTML. It's very much like this is why we've always written web pages as if they were outlines. Heading, you know, heading one, headings two, headings 3, 4, 5, 6, etc. Right? So, organizing with heading headings and bullets helps the AI to understand. And then finally, uh, authoritative. So, this is where this is where you get to site references. Um, and and I I've actually started if I reference something in the content, I will link to it right there immediately below that piece of of text. or I I recently wrote an article where I used actual footnotes, right? I used hyperl footnotes so that I have a you know one and it's linking to the actual source reference as if it was a um as if it was a uh an academic paper. Just a quick

example of a content before and after. On the left you can see wall o text similar to many many web pages out there. And on the right you can see this is a um this is I I was originally introduced to this concept a couple years ago at um SEO spring training and and the idea is this question topic sentence and then contextual explanation. Right? So this is the way the AI engines think. So in this case you can see the question why is higher education SEO important? the answer. Higher education SEO and insurance university stay visible in search results blah blah blah, right? And then a little bit of exposition so that you get a little bit more context. So what does that look like in reality? It looks like this. So you can see we're there in a couple of different places. I'm not going to tell you all the places that we actually publish content, but you can see search influences there as one of the references for this particular search phrase. This is the example of a of a rich snippet that's actually below the AIO. To me, this is money, right? If you can be in the AIO and you can be in the rich snippet. Um, and then, you know, if you can also be intent people

also ask, you know, that's pretty freaking cool, right? So, think about that. So, what does it actually look like on the page? It looks like this. So, here's the main image of the article. This is on the story's website. Um, and you can see this is the actual snippet, y'all. So, like I got here by clicking that link in Google. And when you click that link in Google, it highlights the snippet in the page. In answer to the question, why is higher education SEO important? Google thinks that this is an authoritative response right in exactly that format. So you'll see it highlights the question, highlights the answer, then it highlights the um then it highlights the the sort of relevance connection of the body of the page. So let's break this down. Search phrase higher education SEO question, why is higher education SEO important? And you can see these are the things that seem important in this context, right? And so my colleague Paula who is closer to her education than I am um she actually diagrammed this sentence for us right so you can see it's a semantic triple higher education SEO ensures universities stay

visible in search results attracting p prospective students right so subject predicate object and you can see what's what's going on here visible universities attract prospective students so we have really um my friend Marty uh wife wrote about this but I think a lot of us who've been thinking a lot about this a lot are starting to just write this naturally, right? It's like what's the question? What's the answer? And then how do I associate it with uh entities in the node world? This is an example of some more semantic triples. Steve Jobs co-founded Apple, right? So which Apple are we talking about in this case by virtue of the fact that we know Steve Jobs with co-founder? Probably not the fruit. Professional education targets working adults, right? You get the idea. the more you can make that connection. And and going back to some of Joyy's examples earlier, in the local search world, it's Steve's Plumbing provides emergency plumbing service in De Mo, Iowa, right? It's person action thing, person action place, right? In the local search world, but this can be applied to everything. Um this is this is Paul's example of the double triple, right? So the the the one

triple is University of Florida offers national rank online MBA. Nationally ranked online MBA leads to career advancement for working professionals. So you can see in the actual write up, the University of Florida offers a nationally ranked MBA that leads to career advancement for working professionals. It's pretty clear. Something else that is, you know, it's back in that place of like conflict between uh SEOs um is schema. I like schema. We've been doing schema for a long time. Article schema, video schema, um in the medical context, procedure schema, right? But as Joy said, I totally agree. If it's not on your page, it might not want to be in your schema except for a couple of things, right? Same as I think is really interesting. You can associate yourself. And this is I want to be clear. This is pigeon schema. That's not real schema. This just like for your understanding. The same as attribute is interesting. You should have made these associations to these entities on the page. But by making them in the schema, you're really tightly binding yourself to the um you know to the the known universe by way of the uh of the of the connection to um by

way of the connection to the actual underlying knowledge graph. So what does this all mean? The other thing that we talked about is clear labels. This idea of like 1 2 3. I think of I always when I think of this I always think of wikih how where they've got you know a b cde e fg right in terms of the process. Uh good recipe sites do this. And so if you can if you can think about the product or service of your organization or your client as if it were a checklist that you would go down and say step one, step two, step three, then you are understanding how these this idea of labeling works. So obviously we talked about um we talked about uh headings and their use and how they're valuable. But the next question is how can I make it really clear what it is? And so Joy talked about uh personal injury lawyers. I'm going to bring another example of that. Well, you can see in response to the question, how lawyers build negligent cases? How does the AI overview look? It's a one, two, three, right? So, do I be thinking about that on my website? Maybe. So, and then finally, you know, don't forget citations. The there's always um I grew up in the local SEO space, so we thought

about citations forever, right? I remember talking like real SEOs and they're like, I don't get that local SEO stuff. I'm like, okay, substitute links for citations. Boom, you're done. And so, now citations are back. Are there referenced sites that you can get a listing on that are showing up frequently in AI? If so, get yourself in there. Maybe you got to pay for it. Maybe you just have to submit for it, right? There's a lot of opportunities for you to get cited. And lest we think it's all over before it started. If you can do SEO, you can do AI SEO. If you understand the basics, you'll re you'll start seeing content differently. You'll start seeing links differently. You'll start thinking like a machine. And you don't have to be Einstein. I don't know why I can't get my face. You'll see. I got the t-shirt, right? You don't have to be Einstein to make this stuff work. Uh you two can win the AI race. A gentic search is the thing that we really really all need to be thinking about right now, which is not it's not just

about the content of the page. It's about what data does the content represent and how does that help our clients or our organizations get in front of the customers whom they would choose to serve. So again, thanks for having me. I hope this was helpful. Uh I really appreciate that was really insightful. Well, and also very entertaining. I love that the final slide with Einstein. You look perfectly together. Now let's move uh to the questions. We have several in in our chat. So, I believe we'll be happy to to cover some. First of all, what's the best way to prep content for AI search? Maybe more structured data, Q&A or top 10s. What's working for you? Um, I think the answer is yes to all of those. The uh if you wanted like a if you wanted like a cheat code, you could you could take a piece of existing content and and I you know, you saw I did this in that higher education SEO example. You could take a piece of existing content and you could replace the key takeaways in the lead with height semantic phrases.

That's really helpful. You should be to the extent that it makes sense adding FAQs and then adding FAQ markup in schema. So if you did you could take an existing piece of content, you could rewrite a number of the headings as question uh tight response and then exposition and then add FAQs and that's a pretty good G-code. It's such a good cheat code that it's going to be every piece of content that gets written pretty soon. But it's u those are the things that we are finding really useful and I have uh I took the entire transcript of um so this this uh this SMX um generative engine optimization master class it was two days. I downloaded the videos. I took the transcript and I uploaded I I took I took the videos into the script. I grabbed the transcripts. I took the transcripts I uploaded them into chat. And so now I have this engine which answers questions the way that I would have in that um session. And what's funny is always answers them in the form of question, topic, sentence, and then exposition. So that's that's my that's my trick. Nice. Thank you. That's that's

really interesting and actionable. And speaking about videos, by the way, uh so the next question is around is everything pushing to video especially now that everyone consumes content from phones. So you know people are different, right? Not everybody likes consuming content by video. I do I I do believe that uh and this was alluded to um in the last two sessions. You can the great thing about video is video is natural language, right? You're talking in a way that you normally talk and therefore you're creating by definition natural language. I also think that search is going multimodal. One of the things we're working on a lot and we've actually taken on a few projects to do this is um content repurposing using AI tools. So basically, you take a piece of long- form content, whether it's a an article or a video or whatever, and you process it through automation tools like make.com using the OpenAI API and the Cloud API and turn it into a whole bunch of different stuff. One of my favorite tools, it's super cheap, it's so easy,

is is called Wave.video. And you can give wave.video video, the URL of an article, and it will generate for you a 60-second uh slide montage of video in the span of about 90 seconds. So, you can literally go from copy and paste URL to live on YouTube in a minute and a half. My point is, and again, just like like my disclaimer is don't do that, right? Because if you do that, it means that you haven't actually reviewed anything and that's probably spam. But do you, boo, right? Whatever whatever works. So my thought process is if I can take that same piece of content, I can deliver it in as many channels as possible and I can do that with the push of a button and a little bit of human review, why the heck not? So, so I'm sorry, let me go back Herardo to the actual question. No, it's not all going to video. It's not all going to video. It's going everywhere, right? Video, audio. Um, I have this project that I do every week where uh I I pull an RSS feed of all of the higher education marketing uh blog posts and I then process them into one big long article and then I take that and I drop it in notebook LM

and I create a podcast with it, right? And then I take it to Wave.vide video and I create a video with it. So, uh, with very little human interaction, I've got long blog post, very much um, analysis of other people's work. I've got a video, I've got a podcast. So, it's everywhere. It's everywhere. The only thing I'm not doing is taking those images from that blog post which I generate with OpenAI. I'm not taking those images yet to uh Pinterest, which I should be doing. That'd be really smart, but I just haven't time to work that out. Great. Thank you very much. Another question is, should we be considering building our own driven features as part of SEO strategy? For instance, chatbot on our site trained in our content. Yeah, I love that idea. I love that idea both as a as an action device on the website, right? like Joy shared her um her calculator that that they're creating a lot of calculators at Sterling Sky and I think that's brilliant. The great the great reason to have any sort of action device on your website is that you're going to get um greater time on site, right? So, as we know from the Google um trials recently,

uh the clickstream data in Chrome is part of the ranking factors and therefore if you can keep people on your site, you can get greater engagement. you're going to you're going to have a better opportunity for, you know, for showing up in search. Now, potentially it also means that you're going to have a better user experience, right? And a better user experience is going to lead to better reviews, right? It's going to lead to better mentions in social, on Reddit, etc. If you're improving the user experience in any form, I think that's a big win. The other thing that I would suggest is this. If you create that um if you create that local language model, right, based on all of your content, you then have the the opportunity to create new content based on your voice, tone, and style. So, it makes the it makes the transition from idea to first draft of content a lot easier when you know that as my chat GBT does, that it really gets you, right? Because then you just say, "Here's an idea. Help me work on it." And boom. Yep. Definitely. Thank you very much. And uh another one, how does the longer query play out for keywords when it

comes to SEO and for paid digital advertising on Google? Actually, I bidden on on those phrases at Search Influence. We have a tremendous tremendously talented team that works on ads. I am not on that team. I've never been because I don't really get ads. But what I will tell you is that um with things like Google's performance max and the the like next levels of that, Google's getting a lot more AI in ads too. And so my perspective when I look at ads data, I I like to look at two different things, right? I look at the keywords and I look at the queries because the keywords are what we say we want to be showing up for and the queries are what people are actually typing that get our ads shown. And what we're and we're definitely seeing that those are greater, you know, greater expressions of longer queries. So, so the reason that I the reason I hedge on the answer to this is like it may not matter, right? It may not matter. You may use a keyword and Google's going to give you a long query and make you show up for that because of their AI inherent in the ad

platforms. Now, the thing I'm super excited about, which I think just launched in the last couple of days in beta, is ads in AIO. I'm so excited about ads and AIO because of two things. Number one, we can get back some of that traffic that we lost, right? Because we can show up in the AIO in a way that hopefully is going to be featured because we know Google wants the clicks, right? Cuz if they don't get clicks, they don't get paid, right? So hopefully we get some of that traffic back. But the other reason I'm super excited about it is this keyword data. Hopefully, when we get people clicking on AIO ads, we get those same queries that I'm excited about in the context of Google Ads and we can start to make decisions about what are the actual questions that are driving clicks and ultimately conversions downstream. Thank you very much. That was a great answer and that was a great wrap- up for our session. Thanks again for sharing the actionable tips and amazing things around AI that you showed with the video. The comments are talking about the reactions of the public. Everyone enjoyed your presentation so much. So, thanks a lot

for joining us today.