Video: AEO for the CMO: An Actionable Maturity Model | Duration: 2816s | Summary: AEO for the CMO: An Actionable Maturity Model | Chapters: Welcome and Introduction (4.96s), Introducing Galyleef's Background (64.93s), AI Reshaping Brands (124.08s), AI's Rapid Adoption (201.37s), AI Narrative Risk (267.26s), SEO Traffic Decline (357.575s), AI Content Saturation (454.85s), Introducing AEO Model (539.04s), AEO Maturity Model (627.86s), Content Strategy Evolution (722.615s), Website Technical Optimization (1096.025s), LLM Response Analysis (1340.8999s), Building Brand Authority (1370.4601s), Improving EAT Recognition (1537.2001s), Advanced Measurement Capabilities (1623.085s), Measurement Tools Overview (1722.8099s), Evaluating AEO Strategy (1837.66s), AEO Maturity Model (1932.26s), Webflow Enterprise Features (2088.135s), AEO Maturity Model (2200.76s), Conclusion and Recap (2276.7249s)
Transcript for "AEO for the CMO: An Actionable Maturity Model": Okay. And we are live. Welcome, everybody, to today's webinar. Today, I am joined by the wonderful Guy Yalif, who is a chief evangelist at Webflow and will be speaking to us about AEO for the CMO and presenting an actionable maturity model today. So without further ado, Guy, I'm gonna hand it over to you and hope everybody enjoys the session. Thank you so very much. Good morning, everybody, and welcome to the AEO for CMO. In this session, we're gonna share the latest thinking and a brand new Webflow framework presented here for the first time on one of the biggest challenges we're all facing as CMOs today, answer engine optimization. I'm excited for this conversation as we all triangulate on how to navigate this rapidly evolving space. Please ask questions in the q and a anytime. We might get to them during. We might get to them at the end. We have time for, set aside for them. My name is Galyleef. I'm a former CMO and most recently the cofounder and CEO of Intelemize, which was a website personalization company built on AI long before GenAI that Webflow acquired a year ago. I've had the privilege of being a marketing leader, helping fellow marketers optimize digital experiences for more than twenty years while at Twitter, Yahoo, BrightRoll, Intelemize, and Tradeweave. Sixty seconds of context on Webflow. Webflow is an enterprise platform that makes it easy for marketers, designers, and developers to work together to design, manage, and optimize visually stunning enterprise grade websites that build your brand and also drive measurable business results. Coding entirely optional. Brands like Sephora, P and G, NCR, Jasper, BBDO, Comscore, Duracell, New York Times, and the UFC trust Webflow today. Alright. Let's dive in. If I were to ask ChatGPT, what is your brand? What would it say? We've each spent years crafting our brands and trying to control our brand narrative. Our teams have dutifully crafted every word on our website, every campaign message, every piece of content. Now imagine waking up and discovering that somebody else had been telling your brand story to customers. Not just any someone, but a machine that is confident, articulate, and sometimes completely wrong about who you are. For many of us, this is happening right now. Today, searchers are letting AI write the first chapter of their relationship with your brand, and sometimes it gets our stories wrong. LLMs are the new gatekeepers to our brands and our traffic. And today, we're gonna talk about the journey to take back some of that control for our brands. First, let's get a sense of the magnitude of the challenge in front of us. AI is rewriting the rules of search because it is genuinely giving buyers more of what they want. Headlines are everywhere. This isn't a fad, and LLMs are changing how brands are discovered and engaged with. Couple of data points. ChatGPT hit a 100,000,000 users in two months, faster than any other app in history because it's useful. And now it's above 800,000,000 monthly. A prominent SEO firm predicts that AI search will surpass traditional search in three years. And Forrester surveyed a bunch of b two b buyers and found that 95 of them plan on using GenAI in their buying process this year. At Webflow, we're seeing LLM bot traffic is now the number two source of bot traffic behind traditional SEO crawlers. I invite you to think that AEO is simultaneously, it is a threat and a real opportunity for both parts of the CMO job, building brand and driving revenue. And this is a wake up call for all of us. And if my conversations with hundreds of CMOs is any reflection, we're listening, and we're trying to navigate our way through AEO. And that's good because we're in the perfect storm for us as CMOs. That perfect storm begins with a loss of control over the words prospects see about our brands. Increasingly, people meet our brands through LLM answers, not through our carefully crafted copy. And as you know, in the AI answer world, we don't fully control those words, the AI does. An AI can give wrong or bland descriptions of our brands, admitting key points, diluting positioning that we spent years building, eroding customer trust, or inadvertently spreading inaccuracies before we can react. Worse, if we have some out of date content on our site, I know that is nobody listening to this, other people, if they have out of date content, or if a third party has out of date content about us on their site, AI is more likely to deliver an answer that misrepresents our brands entirely. Columbia University did a study of top eight models, including ChatGPT and Gemini. They found that it gave wrong answers to more than 60% of queries. Now while a certain Columbia study is sound, anecdotally, that feels quite high to me. I think, subjectively, 10 to 20% is wrong unless you go asking really detailed questions, in which case that number could be higher. Regardless of where that number is, net net, there's a real narrative risk for our brands. And if we don't proactively manage how our brands appear in LLM outputs, we risk letting the machine tell our story, and they may tell it poorly. Point number two in the perfect storm. AEO is delivering less traffic than SEO typically, making it harder for us to deliver the pipeline and revenue we are accountable for. Organic search is declining, including for sites that depended on it heavily. Google AI overviews. They're now in 47% of searches, and a well known SEO firm found that click rates with searches with AI overviews are down 34%. Then when we do get a mention, ChatGPT search links to the wrong source article nearly 40% of the time and just doesn't provide a source 21% of the time. Virtually all the CMOs I talk to say they're seeing less SEO traffic, sometimes a bit less, sometimes a lot less. Bain estimates that SEO traffic is down 15 to 25%. There is a silver lining. Traffic from LLMs is more qualified and converts somewhere between 4.4 x to 23 x more often, depending on who did the study, than normal SEO traffic. And LLM visitors go to 50% more pages than search visitors. So they did more of that qualification in the LLM and then came to us. Who bears the responsibility of driving that traffic? We do. This drop in traffic, when it gets this big, stops being an SEO problem and starts becoming a company problem. And we, as CMOs, are already being squeezed with higher goals, flat budgets, and more. We don't need this too. Final piece of the perfect storm is being driven by the fact that GenAI makes infinite content creation super easy. The predictable result when everybody's drawn for, inspiration from all the same LLMs, A sea of sameness. We become indistinguishable voices in a crowded room, all claiming to be the best first leader. How can AI tell our unique story when we haven't been telling it ourselves? The data supports the emotion. LinkedIn revealed that viewers, buyers, when they were shown b to b ads, attributed the ad to the right brand 19 of the time. 81% of the time they got it wrong. And we watch in the public markets brands playing along. If you look at the S and P five hundred, three years ago, less than 10% of them talked about AI. Now, that's more than half. This sameness causes prospects to tune out, especially in crowded markets, and that makes it harder to stand out in LLMs. All that having been said, as with any new type of media, those who act early find opportunity and bargains, just like the early days of search and mobile and native ads, and it is early days for AEO. Despite its meteoric growth, ChatGPT and all other AI tools are currently less than 2% of search share. So how do we navigate these challenges? Well, today, we are thrilled to introduce for the first time Webflow's AEO maturity model. Not what you're seeing on on screen. We'll cover it more. What What we're gonna do is talk through a very well researched, very educated guess because no one actually knows for sure. Anyone who says they know with certainty is lying to you. We're sharing the best thinking we see in market, and we'll keep evolving this actively as we learn more. And to illustrate it, to walk through it, we're gonna pretend we're this fictitious company, CloudSecure. We're gonna go look at what we, as CloudSecure CMO, think our brand is on the left and what LLMs tell our prospects our brand is on the right. Let's see where we might start. On the left, we've got our brand on the right, we've got the LLM. On the right, you will see a lot of red, showing that the LLM kinda got it wrong. And on the left, you also see plenty of red showing that the LLM is missing lots of the messages we really want pulled through. And by the way, in search, they were pulled through more often because it was pulling snippets from our site, not reformulating them. Okay. So how do we change this? What's our roadmap for telling our brand story and delivering the pipeline we're accountable for? That's the map we're gonna look at now. And it begins with the number one question I get asked by CMOs, which is, how do AEO and SEO differ? I think there are four things that are worth your attention as CMOs to guide your teams. One, content, where, and we're gonna talk through this in more detail, we'll go from keywords to answers to personalization. Second category, technical, where we're going to go from on page SEO to automated site structure that makes it easy for LLMs to understand and consume our sites on superfast sites globally. Three, for authority, we're gonna go from backlinks to widespread positive mentions in plain text on other sites. And our measurement will shift from keyword ranking, which doesn't exist in LLMs, to mentions and share of voice. Does this resonate? Another lens on this is to say that AEO is close to what good SEO always should have been, original, valuable, relevant content. In fact, if you look at Google AI overviews, 52% of the top of the entries in AIO, 52 of the things cited there are also in the top 10 search results, meaning that ranking higher in SEO is better, but not a requirement for AEO. So let's map out our journey through this silly, fictitious sea of sameness. Today's discussion is about your map to take back brand authority and drive new demand, and Webflow's AEO maturity model is that map. The model itself looks different, but we'll use this this pirate sea treasure map along the way. The seas are rough and rapidly changing. Like the early days of search, the ground truth is changing often, algorithms are changing monthly or more frequently. There's a reason that when Google changes its algorithm now, it makes the news because it happens far less often, and it took years to get here. So let's begin our journey. And our first stop, Content Island. Your content strategy evolution determines whether you capture existing demand or create new demand. At level one, your team is likely focused on keyword driven pages targeting immediate buyers. This is an effective approach, but it tends to limit your engagement mostly to prospects that are already familiar with your brand, so you're capturing existing demand. As your team matures, your content strategy shifts from repeating keywords to answering the real questions your prospects are asking throughout their buying journey. This widens your reach into non branded terms and earlier buying stages. Like what SEO should have always been, AEO rewards, as we were talking a minute ago, genuinely useful using, pardon me, original content, especially original data. And in some sense, LLMs kinda have a better view of what's good content because they get a full conversation back and forth rather than the 10 blue links. Advanced content strategies involve comprehensive topic ownership, where your team systematically covers an entire topic that your prospects care about with artifacts that are regularly updated. Real world example on the regular updating. We use AirOps to refresh five x more content than we did before. That helped us go from refreshing less than 50 articles per year to automatically optimizing dozens every month. And our content team saw literally within days 40% uplift in total organic traffic for the updated content. Okay. At this stage, content's part of a broader strategy. One off content is rare. And at the highest level of maturity, your content becomes a flagship business asset. Anecdotally, you start hearing people say, I've been reading your reports for months, so I already know you're good. You've established your brand as the go to authority, resulting in shortened sales cycles and even creating partnership opportunities. At this highest level, your content becomes highly personalized, speaking to distinct segments, accounts, personas, even individuals, all while staying on brand voice. McKinsey found that 71% of consumers expect companies to deliver personalized interactions, and three quarters, 76%, kinda got frustrated when that didn't happen. And a real world example of the positive impact this approach can have, NerdWallet recently reported 35% revenue growth despite 20% decrease in site traffic because they focused on providing expert answers across multiple platforms. One other note on content. I firmly believe we should be creating content throughout the funnel. Nonetheless, I've still heard anecdotally, LLMs tend to look to third party sites like top 10 lists upper funnel and to our websites lower funnel. One other notable, LLM queries tend to be longer than search queries. Average LLM query, 23 words. Average SEO query, four words. So LLM's appetite for answering longer tail questions on your website is likely really high. I still think you need to create head, torso, and tail content, but the appetite for tail content is probably higher than it was in search. Now I'd like to check with you on what you see where you are in your business and this journey. So I invite you to scan this QR code or go to Slido and type in that number. We'll pause for a minute or so while people do that. And I wanna ask, hey. Is your team targeting the questions prospects are asking? Is that a primary way that you choose which content to create? And once you go there, please keep that up. Thank you for whoever voted. Please keep that up, as a web page because we're gonna use this three more times. So I'm gonna wait a minute while people do that. And if you've got questions, please put them in q and a, and I'm gonna scroll through those now. We've got a good balance here. 12 people, 13. Awesome. I think we still have more people. I'll leave another minute there. Anecdotally in the conversations I'm having, the core driver tends to be keywords. Answers may be the inspiration, and I think LLMs will cause us to shift that a bunch to have answers be the primary pivot point. Alright. We have a lot of people say they use, answers as a primary. Awesome. So roughly 70, seventy thirty. Thank you for continuing to vote. I'm gonna keep going. Please keep voting here and keep that up because we'll use it again. So in this hypothetical cloud secure example, what impact might this focus on content have on our presence in an l LLM? Well, here's what we had before. Here is what we might have now. Again, I made this up just to illustrate, but now the LLM remembers what our product does. But there are still some things wrong, and there are lots of things that are missing for what we want on the left. So let's move on to the next stop on our map, which is Technical Island. In the technical category, your website infrastructure becomes either an advantage or liability with LLMs. The biggest shift is the increased importance of explicitly adding structured metadata to your site to make it easier for LLMs to understand the meaning of your site. Let's weave that in through the five levels. At level one, your team's focused on basic SEO hygiene, such as ensuring your website's crawlable, mobile friendly, and stays visible where your prospects already search for you. Even at level one, we collectively have a bunch of work to do. 25% of the top ranking pages are missing meta descriptions. Now as your team progresses, your technical foundation supports systematic content creation at scale while ensuring super fast, 100% uptime, reliable user experiences that signal quality both to prospects and LLMs. AEOs also reward your content at this stage because it's well organized. You're using phrases like in summary. It's easy to parse. You might be using bullets. And dense with net new meaning. Mature technical capabilities mean your infrastructure is actively working for you, automatically generating the structured data that helps LLMs understand and cite your content. Also, you effectively maintain multiple interfaces, one for humans and one for machines. I wanna come back to the structure. The importance of it cannot be overemphasized. There's one in particular. Google, Bing, Yahoo, Yandex, and others created this thing called schema. It's at schema.org. It's worth remembering as you guide your teams. There's potential competitive advantage waiting for you in it because 88% of sites haven't implemented schema.org markup. And you can think about it like we all did, some of the tagging for, Facebook and help them understand our sites. This is another generation of that. And those 88% of sites that haven't done that, receiving visibility and traffic to arrivals who have done that optimization. At the same time, today, in Google, 73% of first page results have used schema. And a real world example that Google shared, they said, hey, Rotten Tomatoes, they added schema actually, structured data. They didn't say schema specifically, to a 100,000 pages, and they saw 25% higher click through rates for pages with the structured data than other Rotten Tomato pages without structured data. Again, schema.org markup, worth remembering here because it's so crucial to visibility. So now at the highest level, you are staying up to date with or ahead of rapidly evolving standards. This technical excellence directly supports your content and authority strategies while reducing the manual overhead that slows down content production and optimization efforts. So where are all of us on this journey? Let's go back to the poll. Please go back to your Slido and answer the question, does your website use schema.org structure today? You may have other metadata on there giving structure. I'm curious, schema.org specifically. I hope you can see the results coming in live. That's great, the yeses. I figured most of us would be in don't know or no. It is great to see yeses. Gonna wait for a few more here. Jeff, thank you in the q and a. Okay. Actually, the numbers are not radically dissimilar to the ones we heard before. By coincidence, it was 88% hadn't, and we have, 87%. Either don't know or no. Okay. Let's play this forward. Now let's say we do all this. More yeses. That's impressive. What might our hypothetical cloud secure LLM response look like? More progress. The LLM parsed our enterprise content. Now it says we're enterprise. Understands we're enterprise grade. We're still missing that we're a platform. We're still missing that we're industry leading and that we're next gen, so let's move on to the next stop in our road map. That's authority island. And in authority, we are building. We're maturing from traditional link acquisition to establishing our brand as a definitive industry voice that AI systems consistently reference. At level one, your team's authority efforts focus on basic credibility signals through backlinks and basic SEO driven reputation management. Progressive authority development involves you and your other thought leaders actively engaging in industry conversations. You might call on podcasts. You might say yes to speaking engagements. You might do strategic content partnerships while your teams are probably building a presence in systems that AI prioritizes more than searched it. That might be Reddit, where you've gotta be super careful about how you show up. Quora, YouTube. Take your podcast. Put it on YouTube. Your team is probably also focused on Google's EAT. It's a made up construct by Google, e e a t. When you demonstrate experience, expertise, authoritativeness, and trustworthiness, you're basically demonstrating that you've got the goods to talk about the thing you're talking about. Like, as an example, you might put author bios in your content to show, yep. Yep. We've got the experience. Google values this in SEO. Can't imagine it won't be valuable in, AEO as well. With more advanced authority, your brand consistently cited as an authoritative source across industry publications. Your research influences industry conversation, and your expertise shorten sales cycles because prospects already trust your knowledge. Here now, you've started also to tell visually engaging, emotionally evocative stories on your site, maximizing the impact of the traffic you do receive, demonstrating to people your authority, telling your story your way. This matters not only to create an emotional connection with prospects, but because when surveyed, 94% of first impressions, they were design related. They had to do with all the work you put into that storytelling. Now at the highest level. Your brand defines industry standards in conversations. You've got widespread positive mentions. Those translate into faster sales cycles, easier partnership discussions, media opportunities, and premium positioning in your market. At this level, your team is not only proactively creating context, they're also proactively going out and correcting inaccuracies in other places because we know that helps shape perception. So let's check-in where each one of us is. How many of you explicitly aim for Google's EAT as you are creating content, knowing that it might impact our authority in SEO and in LLMs? Seeing fifty fifty. Give another minute for folks to chime in. Thanks all for those that are doing so. It makes this a lot of fun. Yeah. I thought what is eat would come up more often. Alright. So some of us are. Google has you know, they change their algorithm regularly, but the notion that, expertise experience, authoritativeness, and trustworthiness would be, meaningful in AEO, that seems assured. Thanks for chiming in on this one. So in our hypothetical cloud secure, what might the world look like? Well, looking a bit better. The LLM recognizes us as an industry leader. Bravo. Still a bit more to go. Getting next generation in there. Let's move on to the final stop in our map, which is Measurement Island. Your measurement capabilities help focus your team efforts on a rapidly evolving search landscape like the early days of search. And at level one, you have traditional keyword rankings that show you where you already appear. In level two and more, you start proactively measuring your presence in AI generated answers because the concept of ranking doesn't matter. Right? It's either you appear or you didn't. And you're tracking mentions. You'll also, over time, begin measuring sentiment. You want, to know, are you being measured mentioned positively, negatively? You wanna know the effectiveness of your content strategy at driving qualified traffic and engagement. Advanced measurement, that provides systematic share of voice tracking against competitors for a basket of questions you care about. It helps measure systematically content accuracy and content freshness. That guides your team where to invest next for maximum impact on how your thought leadership efforts translate into market perception. The most sophisticated measurement capabilities enable predictive decision making through real time predictive and strategic analysis. Your team anticipates and responds proactively to shifts in AI driven search behavior, which is changing monthly, And they allocate content and authority building resources for maximum competitive advantage rather than reacting to changes after competitors have already captured the opportunity. Okay. With this as context on measurement, final poll. Does your team today mention excuse me. Measure mentions in LLMs? Mentions, sentiment, share voice, any of those. Eighty twenty feels about right from what I've heard in other conversations. I I'm betting we all believe a year from now that's gonna be eighty twenty the other way. Alright. Two thirds, one third. Wow. This is great. Well, it turns out you may be using one of many of the options that are out there. Here are several. There's nothing evaluative or endorsing in this list. This is just a list that Andreessen, happens to put out that overlaps with what I've heard as well. You see some of these are like stalwarts, established brands in SEO measurement like Semrush and Ahrefs. Some are brand new, like Profound and Scrunch. Okay. So now we've done all four of these things. We've worked on the content category. We've worked on technical, on authority, and on measurement. The ideal is that the goal at the end of the roadmap is us controlling our brand narrative and maximizing traffic to our sites as much as is possible. And so what might that look like in the LLM? Voila. We have brand control as much as we could possibly have. Hopefully, this also means maximizing traffic in a new search world because prospects like this message. In fact, academic research found that showing up in LLM mentions, actually, it it it was, if you optimize for LLMs, you can show up 40% more often. I wanna quote it the right way. And so this work directly can have an impact. And standing out in the sea of sameness, is really, really valuable. And, it in summary is these four categories. And so this is one I would suggest screenshotting. These are the areas to ask questions about. These are the areas to probe for knowledge and progress on among your team. In summary, content, where you go from keywords to answers and then personalization. Technical, where you go from on page SEO to automated site structure to make it easy for LOMs to understand your site to super fast sites globally authority, where you go from backlinks to widespread positive mentions with visually stunning sites and experiences, and measurement, where you go from keyword ranking to mentions to share a voice. There is a lot of opportunity here, and it is super early. Candidly, I've spent some hours looking far and wide for real world examples of this working. I had a tough time finding any. Why? AEO is still new. Most companies are just beginning this transition. I actually bet many folks are doing this, but they haven't gotten far enough or aren't yet ready to publicize it as a formal strategy given the magnitude of the opportunity. Thank you for those sharing in the chat. So how then do you evaluate your team? How do we wrap all this all those levels we talked about earlier together? We've come up with a five level scale to measure yourself against. Level one, keywords. Your search strategy relies primarily on brand or main category terms where you're capturing demand from prospects that know you. Level two, answers. Content begins focusing on answers that typical prospects ask. You widen reach into earlier buying stages. You're going after non branded terms. Three, structure next level of maturity. You're systematically creating answers and structuring your site. Four pillar, your brand is now an acknowledged pillar of thought leadership, attracting high value links and AI citations. And five, authority, you are bobbing and weaving with this rapidly changing space. So what does it look like when we put all of this together? Here is Webflow's AEO Maturity Model. If you take one picture from today, this is it. Here are the guidelines for each of the four categories content, technical authority, and measurement, laid out for each of the five levels. This is the key actionable framework for today, sharing one screen on a summary the steps to AEO maturity. And there's a lot of depth behind these for your SEO teams, for your content teams that we'll lay out in future blog posts and make available. But for today, right now, I invite you to, use this framework and to use it back when you go back with your teams. And I invite you, should you want to know where you are today after you've taken shots of that, go here. Take a go to this QR code and take this maturity model for a spin. When you go there, you're gonna enter your domain name, and we use an LLM, of course, to look at what you and others are saying on about your brand. We look on your site. We look off your site, including LinkedIn, YouTube, Reddit, top industry sites, and more. And then we run that and more against the maturity model we were just looking at. You're gonna get a summary in your email inbox a few minutes later, and you can share the results with your team with, hey. Here's where we are in each of these four categories. Here's what we see. Here's what we can do next as some ideas. There are many paths forward. These are some very thoughtful suggestions. This is what that page is gonna look like on screen where you enter your info, and we hope this helps make this actionable today. Now actioning this with your team, part talent, part process, part tooling, like anything else, on tooling, I'd be remiss not to spend one minute on the fact that Webflow can help. We are a couple $100,000,000,000 plus revenue enterprise platform. We have 300,000 customers, 1,200 agency partners. We host millions of sites, publish 15,000 new ones every hour, serve eight terabytes of assets every hour. We're ready for enterprise scale. We are at enterprise scale already. Briefly, what Webflow does in each of the four categories, in content, make it really easy for you to create content that's on brand voice for human review, make it easy to personalize for everyone by segment, by account, or do one to one personalization at scale. We have a powerful, flexible, composable CMS for dynamic content creation and LLM readiness. We'll make it really easy to localize your site, including AI translation. Technically, super fast global sites with two fifty data centers around the world, mobile optimized by default, excellent Lighthouse scores with structure, automated site map creation, robots. Txt and schema.org support, and clean semantic code that both search engines and AI bots like. And on security, everything's auto updated, built in, anti form spam, along with fine grain SEO control. On authority, when you get that traffic, we make it easy for you to create on brand, visually stunning, engaging experiences, and to scale that across your marketing team so they can't break the brand, they can't break the site. They can just go create compelling experiences. And finally, in measurement, you have real time content analytics, helping you understand where people are going and the content they're consuming on your site along with real time experiment analytics. That's a bit on Webflow. So here, again, the one picture for the day, this is our AEO maturity model. LLMs and answer engines, this is a once in a generation shift in how buyers find information. And and I say this as somebody who lived through .com one, mobile, social. As CMOs, we've gotta adapt our brands or we are gonna be sidelined. It'll impact our ability to build our brands, deliver the pipeline and revenue we're accountable for. This AEO maturity model, we think it can be your map through that sea of sameness. So we humbly suggest you use it to assess honestly your organization, where it stands, and plot the course forward. If you wanna give your team just two pieces of homework to begin this journey after running them through that assessment, shift from counting keywords to answering questions and add the schema.org structure. I'm now gonna go look at the questions we have. We hope this was genuinely useful for you. We love talking about this stuff. It is genuinely interesting. Happy to go deeper, on LinkedIn, or there is the, assessment that we were talking about earlier. And now I'm gonna go look over questions to see if what we have. Feel free to add more. How do we measure mentions in LLMs? So, there you basically end up taking a set of questions you think are relevant to your brand. You pick a group of them. And a good place I would suggest to do that, go listen to your sales call recordings, especially mid and late funnel. Go look maybe at support emails if you've got expansion, and then go pick groups of questions around those. And then ask one of those bunch of tools we shared before, to go look at how often you show up in those. Is there evidence that LLMs recognize and treat sponsored content differently from more organic mentions from your brand or company and third party industry pods and content? I don't know yet. I don't know that people know yet. If we do know that they look at Forbes and we know Forbes has a lot of sponsored content, and that they are consuming that. I don't know yet. If someone does have a lens on that, I'm all ears. What are your thoughts on bifurcating your website? Serving experience specifically for humans, and then one for LLMs, an AI that no human will ever see. Actually, we think this is sort of exactly the answer in that you shouldn't, one could interpret that, Josh, and I know you didn't mean it this way, as black hat, where like in SEO land, you'd present everybody with one site, and then when the Google bot shows up, you show them something else. And that's considered bad, and they will penalize you for that. And I know that's not the direction you were going. We do think that you do need an interface for humans and an interface for LLMs. They should be consistent. They should be showing similar things. But to our discussion here, the LLMs, they really appreciate laid out explicit structure content. That schema.org, for example, it's got like, here's the structure of my blog page. Here's a product page. Here's an event page. Do humans need to see something that says, well, in this box, I've got the date of the event? No. They don't. Right? They just wanna see the content. So they are consistent, but they are two different interfaces. We make that super easy to do, and LLMs, I could imagine over time, will continue to give us more guidance just like Google did. A great one. How important are websites going forward? Websites aren't going away. Neither is SEO. Why? Well, SEO because I think there will continue to be demand for people to just be like, I want this. Websites, because we want and our prospects want for us to tell our story the way we wanna tell our story, to tell the whole thing not to have it synthesized, to hear it the way we wanna project it. That is something prospects want at various stages of the funnel, and so I don't think the need will go away at all. In fact, I think the need goes up to have these visually engaging, emotionally evocative websites. By the way, you might have noticed that a couple of the LLMs have created their own browsers. Only they know why. But I could guess part of it is to not pay a gazillion dollars to be the default search engine like Google paid Apple. Okay. The other part that's relevant for this discussion is they now can see not only the conversation you had with them on the LLM, but when you went to a website that was cited, what'd you do? How long did you stay? Was the content useful? They will have a better read, which makes it even more valuable to have the human interface of your website be excellent and engaging. Let me see if there are others. Do LLM answers differ for the same questions from different accounts? So I don't have the direct answer to that. I do know for sure that LLMs differ among themselves. They differ among the models within themselves, and they differ run to run. Like, if you ask the same question multiple times, you will get different answers. Chances are they won't be radically different, but it won't be identical. Whereas in search, it tended to be identical. Why? The LLMs are probabilistic. There are, statistics and some randomness involved. It's part of what makes them so powerful, but it's also part of why, like, an LLM is not what you want calculating your paycheck. So you want your paycheck to be deterministic. You want, like, I'm supposed to be paid a $100, I should get exactly a 100. Not probably around a 100. I'm being overly harsh on LLMs, but, like, that's the explanation for why it won't always be the same. What is your advice for using this for small businesses that don't have teams but wanna invest in their online engagements and freelancers or small agencies? I think there's great opportunity here, especially for small businesses because the preciseness with which people are going to ask questions, we talked before 23 word long average LLM query versus four word long average SEO query, creates great opportunity for SMBs. Right? I'm not just gonna say who's got, well, I'm drinking Coke. That's not a who's got, you know, the nearest Coke to me, or what's the nearest, convenience store to pick things up on. I'm gonna look for something very local or very precise. I need a plumber specifically for this in my area. And so I think there are real opportunities to create some content that speaks to those, right, that may be what's on your blog, to put a little bit of structure on your site. We can help you do that. A freelancer can help you do that. Or to offer those as services if you're a freelancer. Let me take a look at some of the other questions we have here. If we have always been focused on creating content that solves the problems customers are looking for, am I on the right path? A 100% you are. Right? The AEO is not radically different than what great SEO should have always been. And many people were doing exactly that. At the same time, you see, you know, we collectively were rewarded for creating things that drove engagement. HubSpot's famous drop in traffic. Folks, I think, exaggerated it some, saying, it went down a ton. The data I saw said it went down 36% from November to December, and they pointed at AEO at LLMs. Everything I see who actually knows? But everything I see and the thing that the SEO expert at HubSpot said was it relates to, this, creating great content, in that they've been rewarded forever for creating content that people engage with. And that content went far and wide. Google, probably, among other things that HubSpot was doing, one of the factors, changed their algorithm to say, hey, I'm gonna give you credit for stuff that's a little closer to home. So now I'm gonna make up an example. HubSpot may have said, you know, here's how to get the next step in your career. And Google's like, no. No. No. You're a CRM. Talk about CRM. That kind of thing. So focusing on great content that relates to what you have authority on, great, great, great use of time. If I'd like to simulate questions in a certain LLM, say, chat g p t through my pro account, will it then give me biased answers based on my previous conversations about my bread? That is an excellent question. Maybe ChatGPT, as we know, by default now is turning on memory, where it remembers some of the previous things about you. You know, when I use it, I remember I asked it something, and it said, well, for CMOs, this will be blah blah blah. I was like, how'd you know I was concerned about CMOs? And it said because of all the stuff you've said previously. So you might want to go, incognito to chat g p t, there. You may wanna ask it to not remember the conversations. You may say, hey. Don't use memory in this one, but there is something, Christopher, to going incognito with the LLMs that's not there by default. And, yes, Jeff, stay in your authority lane, and you can explore the boundaries of that authority lane if you've got the resources to go exploring there. But, eat, really. The authority, is is where you will get most reported, it seems. Okay. Other questions? I'll let the awkward pause for one minute and, then, see if Martha wants to come back and wrap us up. But I appreciate these questions. These are excellent. We, again, love talking about this. We would love your feedback on the maturity assessment we shared there. We're gonna post actually, we probably did at the beginning of this, a blog post about the maturity model to share more about it, and would love to dive in more deeply together. And with that, I don't see more questions, so I will say a huge thank you for spending the time with us. And, Martha, back over to you. Thank you so much, Guy. What a great session. And if, you click on this ticker at the bottom of the screen, everyone, you can access that maturity assessment directly too. So, yeah, just huge thank you, to Guy and to Webflow for this session. Yeah. I look forward to the next one.