I'm going to mute you, Jaakko, as well. 친구 All right, good morning everyone. I think it's time to start soon as people will be joining the webinar. On my behalf, welcome everyone to this future of B2B visibility webinar that I'm hosting together with Louise. I'll do a quick round of introductions in just a minute, but just give people a couple of seconds to join the call. Usually it takes a minute or two so that people get on board. But as we do that while we wait for people, let's start by doing a round of introductions. So first of all, my name is Jaaf Kosoudunsaari. I'm the co-host of the webinar partner at Obsidian Helsinki. And I'll be discussing more of the SEO and GEO side of the webinar. Together with me, I am our lovely Louise from Copenhagen. Louise, maybe you can introduce yourself. Yes, definitely. Good morning, everyone. Thank you for joining. Thank you for the introduction, Jakob. My name is Louise. And as just mentioned, I sit in our Copenhagen office. And I work with Obsidian's biggest B2B clients, often clients who are also very much interested in this topic in terms of getting more LLM visibility. And one way to do this is also by leveraging your LinkedIn profile, both on the company side and also the private part. So that's what I will present today. Yeah, so we're going to be covering basically the two, I would say top priorities of LLM optimization or visibility optimization, which are your own website, which I'll be talking more about, and then LinkedIn, which Louise will be talking about. But just basically to kick things off and to ensure that you guys are up and fired, let's start with a quick poll. So I'm going to start a poll that's going to be answering like, or asking you guys, I mean, how much time have you spent LLM optimization so far this year. So just give me like a average or a guesstimate of probably you haven't tracked the number of hours, but just to understand like, what's your initial level of people on the call. So you'll see the poll on the should be on the right side of your screen and answer at your own leisure and we'll have a look at the results soon. But while you're answering that poll, a quick roundup of what's going to happen today. So basically, we're going to give you an understanding of what you can do on your own website, where you can do your LinkedIn, and basically build on the foundation of what basically LLM optimization is all about, and why is it relevant. You're going to go away with the slides that we are sharing right now. So you can take it easy with the notes. Everything, the slides are pretty self-contained. So you also have a good checklist to go away with. And hopefully that you can get a lot of also concrete stuff that you can start executing on basically from tomorrow onwards after the webinar. Any questions that you have, feel free to answer, shoot them in the chat. And we'll try to pick them up as we go. But of course, we'll, we should have some time for questions at the end as well. But with that, let's get started on today's topic. So first, quickly, basically, the foundational thing, question that you can, you can basically start thinking about the topic is that if you open, open chat GPT and ask like, who's the biggest supplier, who's the best supplier in my field in Finland today? And, or you could, for example, who's the best digital marketing supplier in Finland? Who's the best audio equipment supplier in Finland? And whatever that is. And asking that question in chat GPT should give you an idea of whether you are currently in the consideration for this, these kinds of searches or not. So it's basically just a quick way of checking, are you basically, have you been doing things correctly? Or is there how much room for improvement that you have? And this is sort of like a quick introductory exercise that you can do after the webinar, you can do it while during the webinar as well, would be lovely to hear your answers at the end or in the chat, for example, like, what do you, what do you get when you, when you type in, type in this question to chat GPT or Claude or whatever your tool of choice is. So have a go at that and see what kind of results you get. Are you being mentioned currently? Or is it mentioning mostly your competitors? Yeah, let's have a look at the agenda for today. So I'm going to be first of all covering a little bit about the importance of the topic right now. So basically, why are we in this webinar? What's changed in the customer journey in order to make it more AI driven? I'm going to try to prove it with a couple of numbers that are going to go behind the shift that's happened towards AI search. Then I'm going to hand things over to Louise, who's going to be talking about LinkedIn as the, as the basically strongest source for, for AI answers. And then I'm going to pick the ball back and walk you through some foundational things in both classic SEO side, and then some new GEO actions that you need to do in order to be cited in those AI answers. And finally, we're going to discuss how to basically measure and how to, how to follow up with the actions that you've done in order to understand like, how well am I currently progressing? So kind of have a look at a quick look at some of the tools that you can use in order to, to see how you how well you're doing in terms of in the new AI search reality. I'm going to leave you then with eventually with a with a quick checklist of things that you can you can do and hopefully have time for time to answer some questions that you have at the end. But on that note, let's get started with discussing the shift in basically in the AI towards the AI, AI driven customer journey. So we see from statistics already that almost every B2B buyer is also using already using AI in their buying process. And it's, it was already 89% last year, in Forrester it's almost it's 94% this year. So basically, it seems to be a default for B2B buyers that whenever they are looking for a new supplier, looking for a new partner, I think that they resort to first is, is basically AI search. And so it's become basically the number one, number one tool that B2B buyers are using. So luckily, there's still, of course, this doesn't mean that the whole customer journey just takes place within the within AI search sphere. So people are still validating recommendations with a sales version. So they're just looking at the shortlist that they get from AI, then finally discussing with a with a human salesperson at the end. But there's a lot of the sales process that happens already before a salesperson has any influence on it. And another thing that's changed is now that previously, it was probably it was a good goal to be on the on the front page of of Google search results. So maybe in the top 10. And of course, it was still weighted that the first three results on the page were getting a majority of the attention and the clicks. But right now it's with AI search, you usually only get three to five brands that are mentioned at all. So if you're not in that shortlist, you're basically not, it's difficult to be a part of the conversation or the whole prior journey at all, because the LLMs are sort of restricting a number of search results they're basically showing. So it's become more and more important to be among those top three brands in your category, if you want to want to do business. And yeah, this is summarized, as I said, the most important part of the buyer's journey happens already before you see a lead. So previously, when an individual or a B2B buyer had a problem or they had a need arise, they were tasked with finding the solution, they would resort to Google and do search, research, do a search, study the web pages of companies. But right now what happens more and more is that the research takes place in AI tools, as a replacing or not replacing, but complementing the old Google search results. And more and more research is taking place, not just to define the shortlist, but also to understand the differences between suppliers is also happening through AI research, not just like the previously, where you'd go through a website and find out the pros and cons of each supplier. Now that a lot of the compression work is also done in AI tools instead of classic web. Yeah, of course, then finally, ultimately, usually with a shortlist, you check in Google, find out what kind of case studies the company has, review the website and so on. Eventually go to talk to a salesperson. So we have this sort of a new, new, newish step in the in the customer journey, which is the AI research. And the numbers are pretty staggering. When you see that the chat GPT already had like before the summer, they had almost a billion weekly active users. So the numbers say that this shift has pretty much dominated the way that we are looking for new solutions. And And not only has ChatGPT grown, but on the side, on the side or the same time, Google has also evolved. So it doesn't necessarily mean that Googling has reduced dramatically. Googling is just complemented by AI search as well. And Google has also, of course, since last year, implemented AI overviews and AI mode as well to their search, search, which is using the same kind of methodology that LLM search, like just GPT, Claude, Perplexity, Gemini use as well. Yeah, so basically, dramatic shift has happened in a year. If you've been monitoring your web analytics, for example, you've probably seen it in that clicks, most websites have reduced, at least that's what we've seen in a lot of our customers' web pages that over the last year, clicks have reduced because people are first of all looking at that AI overviews to understand an answer. So whenever they're Googling, they're seeing the AI overview and that satisfies a lot of the information needs that people have. And search results have become, instead of a list of 10 blue links, it's now an AI overview. And then it's an ad. And then it's the blue links finally at the end of the page. So climbing to the being cited in the AI overview is really important today, as the number of Google searches showing AI overviews is also growing the whole time. Up to almost 50% of searches are already showing an AI overview. Whereas the more niche topic it is, the less likely it is to show, but the more mainstream search, the more mainstream keyword or more mainstream prompt, the more likely it is that Google will show an AI overview, not show a link to your, not show just a link to your website, but show its own analysis of what it sees as the perfect answer to whatever the person was searching. So to an extent clicks have been replaced by answers and answers then need sources. So you need to have, be cited. Your goal is now to be, not just to get a click from Google, but also to be, but to be a source in the answer that Google is providing. So, as I said, number of clicks has reduced, but there's a number of reduction in clicks is not equal to, to basically everyone. Because we've, studies show that brands that are cited in AI overviews still get a lot more, still get a lot of organic clicks, and in fact get a lot more organic clicks that brands are not shown in the AI overviews. So when Google, ever Google generates an AI overview, you, it cites sources there and those sources are still getting linked to. So basically creating content that will Google will adopt in your AI overview and then getting the link is basically what, what the goal of SEO and GEO has evolved to being to. So there's a dramatic reduction in the number of searches that end without a click. So up to almost over, over 60% of Google searches don't end without a click. And if there's an AI overview, it's over 80% that don't end with a click. But you are in that AI overview, you still have a higher percentage chance of getting a click from Google, getting that B2B buyer to, to land on your website. Good. Then that's sort of a, we're on the statistics side, a couple of more thoughts to leave you with. So how is then this, this change in the customer journey and is changing in the AI overviews? How is that affecting SEO and GEO? And of course, we'll be covering like practicalities slightly later on in the presentation, but just as a couple of slides as a, as a background. So first of all, AI appreciates sources that are authoritative. Sounds familiar from classic SEO. As SEO is also based on a lot, lot too, on how an authoritative, authoritative source you are. It's also favoring search, search, or favoring sources that are easy to read and easy to understand. So instead of writing super marketing in congested language, AI favors language that's quite easy to understand, easily written, clearly written, and answers that are clearly provided. And it does though, through two different, or an overall higher level on two different ways. So first of all, there's a training data that the model has been developing over the long term. And in order to use that training data, it is going through your website, hopefully, if you're allowing it to go through your website. But AI is also, or the LLMs are also looking at other sources like Wikipedia, like media, like industry media, comparison portals, whatever the material you can find in order to understand what your entity is about. And whether it's possibly relevant to answering any prompt or question that the user provides. And this training data is sort of evolving over a long term period. And it's supplemented by basically live search. So when an AI is, when the LLM is not capable of providing the best possible answer, just by using its own training data, it is then doing a live search. So in the same way as the Google bot basically scours the internet, chat GPT search, cloud search, copilot, perplexity, and also going through the website and picking the sources that they see relevant fit. So this is basically same question, same prompt given in Google. And in this purely AI tools has two different sort of ways of showing up. In Google, you get the AI overview first, then you have the ads, and then you have the organic results of the classic SEO stuff. In chat GPT, it basically bundles up all the answers into, as you've seen, it bundles up the list of options that it has into a couple of sources that it seems that are best suited for the buyer in question. And there's no longer this long tail of answers, where it's basically much more congested into one. And of course, this is a chat GPT that's the ads changes, for example, this slightly. So you still have even though you're not might not be organically favored in the search results. This might be also your way of climbing into the results. So if that's how AI finds the answers, it's important to understand, well, that there's two different sort of tracks that they're providing, but SEO and GEO are not competing with each other. It's basically GEO is just an evolution of SEO. A lot of the work is basically mostly the same. And as we'll cover, and I'll cover slightly later on, you'll see that the traditional SEO, some of the stuff that you do in traditional SEO is still very valid for GEO as well. But then you have a certain additional task list of GEO goals or GEO stuff that you need to do in order to guarantee that you're visible in the in LLM results as well. And this is this a lot of the work is the same, but you also find that a lot of the results are also the same. So I mean, in most of the fields that I've studied, there's a high correlation between good search engine visibility and good LLM visibility. So if you're among the top results in Google, it's quite likely that you're going to be among the top results in LLM search as well. So this favors sort of brands and websites that I've been doing long-term SEO. There's a sort of slight variance into the methodology of how you do things. Your goal is, or how you measure things is probably different. So in traditionally, you were measuring rankings per page, for example, seeing looking at clicks and or number of organic traffic you were getting. Well, with the AI-driven GEO, clicks are of less importance because just just so much less of them. So it's harder to make any, any, harder to make decisions based on number of clicks because there's a few of them. So you're more looking at mentions, citations. So whether you are cited as a source in AI search for the relevant searches or prompts in your field. But yeah, foundation remains the same. Your goal is to build a site that's easy to crawl. It's well structured, access and shows authority in the eyes of Google and in the eyes of AI search. So you're trying to, not trying, but being an expert in your field. One important note to make is that I haven't seen too many of examples of in Finland of this, but AI crawlers are pretty bad at running JavaScript. So if your content is hidden inside JavaScript and not visible in raw HTML, you might struggle to get sourced. But likely if you're using a modern CMS tool and not a custom site, you're unlikely to run into this problem, at least in the websites that we've looked at. Good. Then, beginning to this, that's the introduction. So far, before we jump into the topic of LinkedIn, I think what we could do is have a look at the results of the poll that we did in the beginning, just to see like how many, how many, how much time have you guys spent so far in optimizing your, doing optimization work for LLMs. I'm gonna show the results results that should be visible in your screen now. So, looks like most of you are actually, have been spending quite a lot of time on this topic already. So, good time to hear. We can, especially towards the end, it does get a little bit technical towards the end. So, good to know that you have a lot of background in this field already. So, hopefully you get more out of the, out of the, the practicalities that we will, will discuss. But clearly the, the audience that we have here is already quite, quite familiar with the topic and have been jumping head first into the world of GEO this year as well. So, more than half of you have spent more than two days this year in working on GEO. Good. Ben. Louise. Yes. Tell me about LinkedIn and its role in, in the new customer journey. Yes. So, first of all, why speak about LinkedIn as a, as a source to, to optimize, if you want to improve your LLM visibility. As Jaco already mentioned, these LLMs look very much to also bigger external platforms. So, of course, they look into your website and we should focus on optimizing this, which we'll cover later. But LinkedIn is, is one of the major sources that the LLMs also look for when they try to, to, to answer your questions. And there is right now a lot of different research on, is it the most or the second most cited source? And it's among the two. So better safe than sorry. I chose a quite big study, which has analyzed 9.5 million AI citations. And here it's the second largest platform just followed by YouTube. So, so LinkedIn is, is one of the channels where we can actually do quite a lot. So that's what we, we want to focus on today. So if we jump straight into, into the next slide, we have an overview of what we will cover today. Right now we have a split between the AI citations where 75% goes from personal posts and 25% comes from company posts. And of course, most of you, marketers out there are in total or almost total control of the company page. And here we can do quite a lot. And that's unfortunately only the 25%. But if we try to engage our expert and colleagues, we can also influence the 75%. So today I'm trying to cover both. So we're going through five different things that you can focus on when you, when you try to improve your LLM disability via LinkedIn. So let's, let's just jump into it. The first one is to choose the answers or the questions you want to answer. So basically we should try to have a strategy around what topics within the LLMs do we want to improve our visibility in. Maybe since you already worked most of you quite a lot with optimizing your LLM visibility, you already know the themes or the prompts that you would like to show up in. And if you already have done that, you should mirror this to your LinkedIn setup. So if you already do prompt tracking or some of your colleagues have set up prompt tracking, look into this tracking in terms of which themes is it that we want to focus on and then mirror this to your LinkedIn setup. If you have not done this yet and not done the analysis, these themes often come from the leadership in terms of which topics or themes do our company really want to be present on in the different LLMs. It can also come from your sales department in terms of when we talk to potential clients. What have their journey been? Which uncertainties do they have or which areas do they look into when they want to find a new partner or collaborator? So there are different ways that you can do this. Look into your already set up prompt tracking or have a speak with your leadership team or your sales team in terms of finding out which themes that we want to focus on. And try to build the content around three to four themes. So we cannot choose a hundred different themes because then we will not show the the LLMs that we are an expert within a hundred different things. Choose three to four different topics that you want to focus on and really build your authority and your content around these three to four topics. As an example of prompts for for our case as a B2B lead generating agency, it could be like which agencies are the best for lead generation in the Nordics? How can B2B companies get better leads from LinkedIn and stuff like that. And if those are the prompts that we really want to focus on in our case, then our content and our page should be focused on these prompts and the answers to these prompts. For Obsidian as a case, just to give you some examples, we want to be very present on AI, of course. It's a lot of competition around this area. So we also try to focus on how can we build up our expertise and show our expertise within this field, within AI and marketing. Another thing could be tracking, which we also focus quite a lot on since we are a performance agency. We want to be dominant in the area of server -side tracking and tracking in general. So as an example, now that Jaco also mentioned the chat GPT ads, which are almost just launched, we also focus on how can we set ourselves within the topic of chat GPT ads and tracking. Because if you want to advertise on a new platform, which chat GPT ads is, then you also need to set up tracking. So we write some content about this theme as well. So usually what we see if we try to go through our own LinkedIn profile and the posts that we do, it's often the employer branding post and the very cultural posts that gain a lot of likes and have a lot of engagement. But those are not the ones that we can show the LLMs what we are experts within. So we have to maybe come to terms with some content which on the platform of LinkedIn maybe does not perform that well because it's highly expert content which often gets less likes than cultural posts or employer branding posts. So try to figure out which are the three to four themes that we want to focus on and keep this in mind when we go through the next points. Because the next point that we want to cover is the company page. And this you are fully in control of. So here we can really go nerdy on this one. And why is it important? It was only for the general LLMs, it was only 25%. But within the different LLMs, as an example of perplexity, almost 60% of the citations came from company pages. So within the different LLMs there are also different sources that they focus mostly on. So especially for perplexity, as an example, the company page is really important. And usually it's something that you set up and forget. Because you make your description, you find the leadership in the platform, you have the different products that you want to focus on, the different industries. When you have set that up, it's my experience. And I can also see that for Obsidian ourselves, that we tend to forget about it. So if anything changes in the terms of the industries that we serve, or the leadership, or our company description, it can be that we've now Obsidian has grown quite a lot. So now we actually have 400 employees, that's a good number to update and stuff like that. So really try to have it as a recurring task for you to have a look at the company page, at least once a quarter, maybe once a month, just to have a look at is everything still true? Or do we actually have more products or more industries that we can service now that we can also show on the page? When I did a review on our own LinkedIn company page from the Obsidian site, I saw as an example that for the services, we didn't include AI in the description, and we didn't include tracking or chat GBT yet. And that's the three topics that I just mentioned before that Obsidian wants to focus on. So it also goes for ourselves that we need to focus on updating this page. But I'm sure it also goes for a lot of you guys out there, that it's something that you said, and then you forget about it. So it's a really good idea already today, perhaps have a look at your company page to see if it's actually updated. And another thing that you can quite easily check as well, is that if your your website is actually linked to LinkedIn page, because if it is that by the structured data, the same as Shima, then on your website, then the LLMs can see that this website is actually tied to this LinkedIn company page. So we can see this these two entities as as the same. So that's two relatively easy things that you can go home and implement already today. The next step is a little bit more complicated, because this also depends on the willingness of your expert or or your colleagues. Because as I mentioned before, 75% of the LLM citations comes from private profiles. So it's not something that we can 100% from a market marketing point of view control, because it is in the hands of our colleagues as well. But I think we should try to do it still. And most often, when you think of who can be the thought leaders or who can be the experts that we want to position in front, you think of the C level, or you think of founders or some very prominent people in your company. And that's totally fine. But it doesn't need to be these very high, high level people. Actually, it's only like around 8% that citations comes from CEOs and 7.5% that comes from founders. So there are still a high percentage which comes from people who are not necessarily at this level. So what we should try to focus on more instead is to focus on people who are very skilled within topics that we want to focus on. So as an example from Obsidian, we have Juan, which is like our tracking expert. He's a part of the partner group. So of course, he is like one of the is in the management of Obsidian, but he's not a CEO, he's not a founder. He's not C level, but he has a very strong voice within the tracking environment. And we can build him up as an expert within this field, because he already is. But we can show this and make sure that his voice is prominent within the LinkedIn. In terms of dividing the 75% within two groups, we have around 40%, which have plus 10k followers, and we have the remaining 35% with less than 10k followers. So again, it doesn't need to be people with a huge following or a huge audience already. It can be people who have less of an audience, but have a very strong voice within the specific field that we want to focus on. So try to really pick a couple of people within the themes that you want to focus on and make those the experts and make sure that you engage with them in trying to get them to post more and help them if they're not so good at posting at LinkedIn already, help them with the with content that they can post from their individual profile. And it's very important that it shouldn't be like with a corporate voice. It should be since it is from the the people's own profile, it should be sharing their own view of of the topics gaining or sharing their own insights and their own knowledge. So it's very important that we don't force the company share a corporate voice upon these people, but we should try to encourage them to share their own opinions about the topics that we want to focus on. And it's it's also important and that can also be difficult to actually start a dialogue on LinkedIn. But often the people who are experts within the field, they are also connected to people who are also interested in in this field. So as an example, again, one within the tracking, he also have a lot of people who follow him because they are interested in this topic already. So he has an easier way of actually starting a dialogue within the tracking field, which could be difficult for us to start from a company profile, as an example. Two important things that you you also want to focus on when you focus on on these experts is that their profile should be verified. That's quite easy to to get done. But it's very important, again, for for LinkedIn to show that these are people with authority and real people, and they are also connected, of course, to to your LinkedIn company profile. And the second one is actually to also get those people which we choose as experts on your website. So again, Juan, as an example, he should be posting from his own profile on LinkedIn, but we should also show him on our website within the sub pages which focus on tracking again to build the credibility and authority within him, within this field and in him to again create the connection between tracking and Juan, both on the website on LinkedIn and etc. So and you can do this again by structured data, which I'm sure that Jakko will cover a lot later in the presentation. Going to the fourth step, it is to write so the AI can quote you. So here it's it's different things that we can focus on. One thing is to focus on the language, which is especially important for for you guys, because we have both the Finnish but also English. And we want to focus on most of the content often in English, if you want to have reach. So if we want to establish ourselves within a thought leader, we should focus on doing the content in English, because it is easier for the the LLMs to to gain this knowledge and establish you as a thought leader, not specifically for for any country, but within the theme itself. But when we want to focus on local prompts, so as an example, the best digital agency in Finland, then it's important that we write it in Finnish as well. So if you have prompts specifically for Finland, and you only have Finland as a market, of course, you should focus on only Finnish. But if you have Finland as a market, but also have other markets which are interesting for you to to focus on, then most of the content should most likely be in English, but you can't do specific Finnish content for the Finnish specific prompts that you want to be visible on. So that's in terms of language. Then there is also different kind of content, ways of doing content on LinkedIn. Most of you are probably most familiar with the standard posts. And most, and maybe some of you have also tried the articles. From my experience, articles does not on LinkedIn have such a good performance always because it's long content, it's more in depth, and it's perhaps a bit more niche. But the LLMs would love this. So if we want to focus on really building authority, we should perhaps also focus on trying to leverage this format even more, the articles format, because you can write more content, you can go in more depth, you can establish yourself as an authority within the field fields that we discussed. And then there are some good small notes that you can do. As one example, like write the title as a question. So actually write the title as a prompt that you want to answer. That again, makes a life easier of the LLMs. Make it too long, didn't read. Publish with a relevant cadence. And this is quite important that we want to focus on establishing also a cadence, which is often and not every two months, but actually once a week, perhaps. And even more with posts, which you can post maybe two, three a week to gain this cadence in, yeah, up to two to three times a week. For the posts, the regular posts, those are more for reach and engagement and where you can focus on promoting the content that you have in the articles. So here you, there are also some small tweaks that you can think about when you post. It is to have keywords in the first line. It is to have, not use hashtags that's, yeah, we don't really need to do that on LinkedIn anymore. But to focus each post on one idea and to make this quite clear. So we don't divide the posts and the articles into three different themes, but focus the content piece on the specific theme that we want or the specific prompt or question that we want to answer. So try to use both of these content pieces, posts, which I'm sure you already do, but try to get the cadence up maybe two, three a week and then also to try to leverage the article format even better. And then I have some stats in terms of what goes again in posts that get cited often. And that is that you use bullets. Again, it's very easy for the LLMs to read. It gets also the reader, a very quick overview. Then it is to have clear H2 and each H3 headings. It is to name other companies, other tools. So you set yourself in connection to other, yeah, either companies or tools. And it is really also to include hard numbers. One of the things that LLMs love is that if you have unique insights or numbers that they can use to give the the prompter an answer. So if you have some specific insights that you cannot find anywhere else, it is easier for them to cite you because you have something unique that they can also show the the person who actually asks. The last step is something that focus on a little bit more of the paid part as well, because what we also see is that there are a correlation between posts that get a lot of comments and a lot of reactions and actually them being cited by the LLMs. So the LLMs needs to have some kind of verification from the LLMs already that this is relevant content. And it can be difficult for especially the experts if they don't have a high engaged audience. If you speak about tracking, it's maybe also some people who don't comment that much on LLMs posts already. So it can be difficult to start a conversation only from the individual person's LinkedIn profile. So what we can do is that we can use the posts that we already have, either from the company side or the individual side, and put them into our paid campaigns, either as thought leader ads if they are from individual people or as a part of our regular business as usual, set up on LinkedIn, if they come from our own company page. So focus on trying to get the best performing organic posts into your paid set up as well, to give them a better chance of actually getting cited by the LLMs and boost this engagement, both in terms of reaction, but also in terms of comments on the posts. Yes. And I think those are the five steps that I wanted to cover today in terms of LinkedIn. And I'll hand the speaker back to Jakob. So really interesting, interesting stuff and compliments quite well what we're going to talk about next, which is going to be the stuff that you can do on your own website, basically. So I'm going to separate it slightly into classic SEO actions that are still key today, and then going to talk about some of the new actions that you need to do in order to be shown in LLM results. There's now these classic five actions that you would do are now paying off not only in classic SEO, but in GEO as well. And the first of all is that if it is, of course, the technical foundation of the site and its crawlability, so that most of the content is rendered on the server so that it's available and visible to the LLM searches is that you have a clean site map and you have a fast loading website and you're not logging into content with the metatags. And of course, as I said before, that you're not blocking any content off with or hiding any content behind JavaScript. And this technical foundation is also quite easy to review and analyze using technical tools. As an easy step, you can, for example, ask Claude or ask any LLM tool that you're using to basically review the technical foundation on your site and suggest any improvements as a very simple step. Then next up is authority and links. So, like in the old SEO world, your domain ratings and number of backlinks that are linking to your site is still fairly well correlating with how likely you are to be cited because it's still like a recommendation marker to those LLMs. Is that if your content is being linked to, it's probably of higher value. So, it's still worth doing digital PR to gain more links because these, of course, medias as neutral parties are quite valuable sources to AI citations. Then structured data, as Luis also mentioned briefly, GEMA has become more important than it was ever before. It was already important if you want to be shown in Google's rich snippets before, but now that structured data is more foundational in order to AI to build basically an entity map of what is your business and where, what are all of the different parts that belong to your entity. What are, for example, your products? What are your services? And so on. Who are the people working at your company? And so on. And using schema like FAQ page to highlight what are the FAQ sections of your site is also, has been shown time and time again to be too valuable in order to be quoted by a search. Getting into schema, if it's not a familiar topic to you, it's usually, if for example, you have a WordPress site and you're using some kind of a SEO plugin like Yoast or numerous other SEO plugins as well. That's usually a simple way to get the structured data done correctly is by using such a, using the same SEO plugin that you've used before. They usually come with these features and if not do a well by yourself, then pick it up with their, picked up the discussion with your, with your developer, site developer partners. In terms of content, you should move more and more towards an answer first style of creating content. So having clear heading hierarchies, using questions as headings and doing shorter sections under those headings and using those, basically those paragraphs to answer directly what the, what the heading was, is more now more important than ever. And also for each, each page is valuable that you have a short summary of the, of the topic in natural neutral language at the top of every page. Seems to be a good way to, in order to be cited in the answer. And finally, in order to, for the, as much, similarly as Google valued before, then same with the LLM world. So having, if you have a blog post or article, it should always have a named author. And the more credentials you can give to the author. So telling about their background, why are they an expert in this field and so on. Having dates shown when the content was updated and so on. Having any original data that you can show, maybe you did some research, or maybe you've got some numbers from your own products or services. This like original research. Research is, is very important for LLMs to determine whether your content is valuable, as it was before in Google world as well. Then a couple of more, more practice, practical things for, for SEO. So, or classical things that are still relevant today. So, moving from basically, and moving towards an answer first way of, of writing your, out your content. So, especially in your product pages, or you might, you might feel tempted to start with a very marketing, trying to convince the speaker kind of, kind of voice. But as more and more of the research phase is moving, not to your own side, but to the LLMs, it can actually be much more beneficial to write in very neutral answer first style. So, instead of saying that, explain, giving a very elaborate explanation of who you are and what your values are, it's better to be more practical and say that we do this and we are good at it doing because X, Y, and Z. Then, the similarly, in, instead of, instead of, instead of using headings that are more vague or more you focused or you as a business focused, moving towards writing headings that are questions and headings that are more detailed into the, into the field that you want to be sourced in. So, instead of using headings like, what are this solution costs and how should I pick apart supplier, for example, these kinds of headings work well in helping AI understand what the content is and, and perhaps even citing directly the answer that you're giving. Then, finally, just guaranteeing that, that basically, your content is visible to, to AI search and not hidden behind JavaScript, for example. You can do this in your browser, whether it's Chrome or Safari or whatever, turn the JavaScript off and go to, go explore the content of your page and see what it looks like. You can, of course, also just prompt your LLM to go to a website and see, go to a specific page and see where and tell you what they, what they're basically seeing, whether they are actually be able to see the, see the content of the site. And this is especially valuable if you're using an older CMS system or that you have a custom built website, because there is more likely that you have some, some non-traditional ways of, of, of showing your content. Then finally, the organized building schemas I mentioned before that they've been, become more and more important. And probably, if not the most important, then at least one of the most important ones is, is your organization schema that you should have on, on basically the pages, all the pages of your website. So stating like using the organization schema in the code of your website to tell me what, what's the name of the business? What's the URL? What's the link of the logo? What's the link of the logo? In some example, having a description of what the business does, having your same as links. So linking to your LinkedIn page, linking to our social media profiles as well, as well, linking to any possible Wikipedia pages that you might have, or Wiki data sources, or whatever external sites that you, you control and are trusted as well. It's a good way of ensuring that AI gets the details of your company correctly. And you can, of course, also implement, there's a lot of other, other stuff that you can include under your organization schema as well. Like for example, your addresses and so on. So that can be important to provide the right content for LLMS. Then moving on to the new GEO actions that you need to achieve. So for example, first of all, the most important one is that you need to provide the AI crawlers access to a site. So if you have a robust text file, and it's currently blocking AI crawlers from accessing your site, you are very unlikely to be cited in the answers. And for example, I just audited a website of a company a while ago, and they had set up these, they had blocked the AI bots from looking at or crawling their site about six months ago. And you could clearly see that they were a source before, but then the results started slowly declining. They weren't quite sure why they had made that decision to block those. So they started to explore within the organization who had made the call and how to reverse the call that they had made. So you need to have that in the robots text, not to be explicitly telling the bots that please don't call my site. Of course, if you have content on your site that you don't want to be crawled, and it might be a good idea to place those restrictions in place. But that's usually not the case in marketing. Then placing an LLMS text file, much like a robots text file. So it's a sort of a robots text file for AI systems, which is a short curated version of your site. So it tells you the basics of your business, links to the most important pages on your website. Of course, the importance of LLMS text files is debated, but it takes less than, probably less than 30 minutes to create one and put it in place. So why not do it because it's really not at least hurting your visibility. Then ensuring that you are building a wide enough entity. So entity by meaning that you're connecting different examples of where your business exists online to one, to the same entity. So for example, you have a wiki data page, wiki data entry about your company, possibly even a wiki data page. And if you're a SaaS business, for example, using G2 or Captera or whatever these sort of yellow pages for different SaaS products are. And linking them back to your business. So to make sure that wherever your product is, is being discussed and providing AI with more and more examples of what your business is and showing the credibility of your business. And this, of course, yeah, as I said, moves to citation seeding as well. So finding out external sites where AI is using as a source, whether it's LinkedIn or whether it's like G2 or Captera or whatever it is in your industry. So making sure that you're also visible in those locations on those websites. So then one seems like a quick hack or quick fix to be a part of the conversation is to write that comparison kind of blog post. So comparing your product to the product of competitors or writing a blog post that lists various different kinds of competitors in your field and showing that benefits and downsides of each alternative seems to be a good way to be sourced. Because you have to be a little bit more neutral when writing that post. So it seems to be that AI is valuing that content. And then finally using FAQs that match real AI questions. So don't just write FAQs based on marketing questions, but try to be neutral when creating those FAQs. So going into more detail about, for example, pricing, for example, availability and so on related to your company and related to the field in general. So that's a good way to be sourced as well. And below in the bottom corner, you have a list of different bot names that you can, should be whitelisting in your robot text file or at least not blacklisting. And we'll provide an example of robots.txt file later on that you can directly use. So basically here you can just basically copy a robots .txt file that allows AI agents to crawl your website. And there's also like a simple way to create your LLMS.txt file. So by telling first of all who your brand is and then providing links to your services, to any resources you have, your About Us page, which seems to be more and more important. So basically guaranteeing that LLMS have the capability to access to your website and understand what the most important content there is. Finally, yeah, as I mentioned, like LLMS are not only looking at your page and they're not only looking at LinkedIn, but they might be looking at other sites as well for whatever to find out, understand what's credible content. So being present in any industry media can be really important in order to be cited in these results. Maybe you want to pitch an article to industry media, or then if it's like a yellow pages kind of sites like G2 or CapTera, as I mentioned before, can be really important, especially in SaaS businesses in order to be cited in the results. But basically looking at the answers that you're getting from AI is a good way to see what kind of websites AI is currently citing there and then making sure that you also have visibility in your business in those same websites. And finally, we're running quickly out of time. So we're going to have a quick look at measurement before closing the webinar. And I know, but still have a couple of time to answer your questions if you have any. But finally, how to then measure your basically visibility because it is a little bit more trickier than classical SEO because you don't have clicks that you can just monitor. So basically you have to do a little bit of manual work in order to understand. You might have to do a little bit of manual work in order to understand how well you're doing. So basically doing sort of what we call a query visibility map in order to basically list certain queries or prompts questions that customers might be asking. And you can determine the volume of each of those using tools like Ahrefs, Brand Radar to see like how many are looking for, for example, best digital marketing agency in Finland. And then seeing whether we are mentioned there, whether our competitors are mentioned there. And then using this query map, for example, updating it monthly with doing the same queries in different systems in order to see how your, your basically your visibility is evolving based on the things that you are doing. And yeah, you can do this in manually, but there are ways to basically do it using Ahrefs Brand Radar, which takes a lot of the work of manual work off. So you just define the prompts and it does automatically runs those prompts through the system, you know, to understand like what's my, how many, how likely is it, is it that my company's going to search results and how many of my competitors are shown there, for example. I see a lot of companies that I discussed this in a previous webinar, I got a lot of questions about it. So how can I track the, the AR traffic from LLM referrals in, in Google Analytics? And so it's basically going into, into those custom reports and creating a group or segment for AR referrals. I'm not going to have time to go into detail about this, but send me a question if you, it's still unclear how to, how to set this up. But it's not super hard, even though, even if I am not one, but I was still able to do it quite easily and have been able to do it. So that you can pick it up and see how many, how many referrals you are currently getting from, from AI search. And interesting that, that note that I've made done with, with majority of my customers is, is that seems that probably 80, 80 to 90% of AI referral traffic is coming to ChatGPT in Finland. So it seems to be like a really dominant presence compared to other AI tools. Not sure if that indicates that ChatGPT is also the market leader in, in AI search. Good. So, and we'll end with a, with a small ad. So if you want to understand what your situation is currently, we'll be happy to do, to do an LLM audit to you guys for free. No, no string is attached. It doesn't, we'll just produce report, review it with you guys and show you what your situation currently is. Are you allowing LLMs to crawl your site? How well is your schema markup set up and whether there's room for improvement on your schema markup? Whether there's room to improve on your content when you've written out? Do you have answer first content and so on? What's your authority situation? Are you being, do you have a Wikipedia data entry and so on? We'll be happy to do this report and just send me an email and we'll be happy to provide a report with you guys and walk you through just to show you the quick step of, of a quick way of showing what their current situation is and what your grade is. It's a pretty harsh grading system that we have. So I haven't seen too many companies land in A. Most of the results have been C or D so far. Good. Finally, we've gone over time. So I'm going to have to leave this as something that you can review after the webinar. But it has a quick timeline of things that you can get started with in order to fix your visibility in LLM search results. So first of all, just to list of quick wins. A lot of them are sort of technical stuff to just to ensure that your baseline is fine. So you're allowing LLMs to crawl your website. And so you have a schema set up, right? The most important schema set up, right? You have picked the right themes that you want to be visible with on produce content on LinkedIn and so on. And from those on building towards a long term plan on how do you want to be or how are you going to be constructing yourself as the thought leader in your industry and as the most visible player in the LLM search, which appears to be more and more dominating the landscape of B2B buying. That's yeah, that's pretty much it. Sorry, we went a couple of minutes over time. But still, if you guys have any questions that you might have, still have a few minutes to answer them. I'm not sure if Luis has jumped directly into another call, but just shoot a message in the chat if you have one. There's nothing. Yeah. I'm curious about the last question that are of the questions we often get list. What do you think Luis is it possible to pay for visibility and AI answers? With chat GBT ads. Yeah, I guess. But I think it is something that you should focus on organically first, have the good foundation in place before you start doing anything paid. So go through the checklist that was just mentioned in the previous slide and make sure you have a good foundation. And then you can see to see if you should boost some of your LinkedIn posts to try to gain visibility or pay for the chat GBT ads. I don't see any questions coming in. So I think it's safe to call it a day just to say thank you on my behalf. Thanks a lot, Luis. And thank you for all participants for joining the webinar. And let me know if you have any questions beforehand. We have another webinar next week as well. So you'll find it on the Obsidian website and feel free to join. Thank you.