iGaming Daily

:
Learn how OptiMove Positionless Marketing is changing how iGaming teams operate. Discover how operators are using OptiMove's Positionless Marketing platform to launch personalized CRM campaigns, dynamically change casino lobbies and bed slips, and create engaging, gamified experiences. Learn more at OptiMove.com. To see how this approach comes to life, OptiMove Connect returns to London on March 11th and 12th, 2026. It is the only user conference where marketers from around the world share real-world results of positionless marketing driving, efficiency and ROI. Register at connect.optimove.com. Hello everyone and welcome back to iGaming Daily by SBC, the marketing part where we talk anything and everything connected to SEO and marketing. Today we are talking to Konrad. He's a marketing director with a massive SEO knowledge and SEO background. Welcome, Konrad. Thank you for joining us. Hello, Ivana. I'm great to be here again. It's my second time on this podcast. Yeah, we don't nearly get enough of you. So we are very happy you agreed and we are very happy you finally found time for us. And we are going to be talking about very, very important and interesting topic and it's how our search is changing, how we behave differently in SERPs because the SERPs has changed and we have got this thing called AI overviews and how us as marketeers try to stay visible and try to stay relevant. So my first question is this zombie of marketing called SEO, is it dead or is it still surviving? Yet another change. are we still irrelevant? No, we're still irrelevant. We're not going to maybe speak about the geography today, but we're going to speak about how SEO is evolving and is definitely still relevant. The important thing is because there was a lot of uh information overload, maybe incorrect information the past year is yeah, SEO is evolving as it has done in the past. There is a major shift. I think this is one of the biggest shifts because the representation layer, how the whole search engine result page is changing. AI mode takes it kind of a complete change. AI overviews, you mentioned them, is kind of slight change. But yeah, we are seeing definitely a shift, a change. And yeah, of course, this as well. other players before we kind of used to say Google, you know, is the main player. They are still the main player, but you know, there's ChetGBT, there's Croc, there's Claude, there's Flexity. So there's other players as well in the market taking the chunk of the market share. Absolutely. So when you search for information, you as a marketeer and as an SEO, do you go to ChetGBT or do you go to Google? Do you read AI overviews? Because I started to go to ChetGBT quite a lot and I read the AI overviews. but I still go to a normal result just in case it's lying to me. What do you do? How do you behave? So um when I actually, when AI mode got released, I kind of shifted a little bit back more to searching on Google. um Of course, we are curious. We are in the SEO world. We play with a lot of models. So yeah, I'm sure everyone had the discovery phase. What is the output uh on Google? What is the output on GGBT? So yeah, AI mode actually is quite interesting. And especially you go on Google, you search, you get the AI overviews at the top. Which kind of, before Google as well had this kind of position zero, a result with features, this kind of takes it to another level because you get an actual real answer, generated answer in the search and the result page. And then you can click as well from there and go to AI mode and continue to search. um Searches is now kind of more natural because actually the prompts, they understand each other so we can do a follow up prompt without going into much detail. The contest is there, the intent is there. We need to understand as well because a lot of people say ASCO is dead. A lot of this kind of uh technology is based on previous concept. So even before, when users used to search a traditional query in Google, you know, there was kind of a query parsing from Google query refinement and as well expanding. So Google as well, sometimes was changing to make it look more realistic and provide more accurate results. um But yeah, now um with AI and with large-fangle models, there's quite a shift. know. Do you love it or do you hate it as an SEO professional? I I'm sort of like... Sometimes I love it, sometimes I hate it, but in the end, I'm just learning to work with it. Yeah, it does provide these challenges. And I think one of the biggest challenges that it provides is it's kind of, it was kind of, kind of not being important anymore in the industry is that AI bots now really don't run their JavaScript, at least most of them. Google is really good at running their JavaScript. So one of the headaches with AI uh engines is that um Sometimes there's not enough visibility because they cannot crawl the business website good enough or fast enough, or they don't understand it because you know how the web is plagued with JavaScript frameworks, Angular, React. yeah, AI bots are struggling, especially the newcomers to the market. And in fact, we're seeing the notion of like certain pre-rendered companies which are kind of now selling themselves to cater for AI bots. So that is part of that headache. Another headache is that they drop in traffic. So we're seeing, of course, when there is an AI overview, these sessions that are without clicks increase slightly, but then with AI mode, actually, they increase even more. So we need to see how we need to survive into the ecosystem. I do see that. we need to make sure we plug into certain results. It becomes more time consuming to have full content ecosystems from the whole journey. Yeah, and you now need to make sure that you cover all the topics, all the angles, even if you have, for instance, a help desk, because certain queries, certain maybe even customer support queries, DLLM, you would prefer that they are sent to your own website, other than maybe an affiliate or a third party site. So it is a challenge. It is a challenge. But yeah, it's a new way to search. yeah, of course then when it comes to automation and what you can do, yeah, it's a new way. It's the future. You touched up on content and we discuss this a lot. Almost everyone touches up on it. Do you believe that as we need to create new types of content and more content, as you said, because we're going to be on your website and third party website just to be relevant because you don't know what's going to quoted in AI overviews or chat GBT. Do you believe we can do this quality content with just AI tools or do you still go for experts? you still go for, we need to have an expert understanding our topic so we actually are shown, we actually are relevant. What's your view on this? Yeah. In these models, um especially in the beginning, they were really hallucinating. And slowly, slowly what the technology adapted themselves is if they ground them um to knowledge basis, to vector data basis, so basically search indexes, um they hallucinate less. So um the main strategy of optimizing our content, it still applies. Of course, sometimes in our industry, we have a deplac notion of Google still matches the query to a keyword. So as well sometimes even keyword density survives. Those notions should have been there for the past five, seven years even more, because we know how Google evolved to... using vectors, embedding models. So optimizing became, content became a little bit more complex. It's like how we used to do before, but we need to understand where and why the certain language model is citing. We used to say for us in the past, I don't know, two or three years that let's answer the query. in the first paragraph and at the beginning rather than going like we used to say before, going in with a backstory and then replying the query at the bottom. So for instance, LLM Studio like that you answer the query at the beginning of the content. So yeah, I'm sure there will be tools that have worked with the profile simulation, but yeah, it's basically we need to make sure that we optimize the content for entities. And as you said, our VZP our, how we interact with the results is different. So how do you measure it? Because it used to be, we are position one, we report on position one, we gain position two, we, we know this keyword is now position two, we are proud of it. But this is sort of like, if you have got AI overviews and they hate results, being position one is not exactly the position one used to be. And you said clicks are going down. So what do you report on? Is there a new reporting model? Are there a new KPIs that you would set up for you and your team? Yeah, this is a good question. So it is an evolving space. There a lot of tools that are tracking large language models, AI mode, and HLGBT. So basically how we used to do before, you put in a keyword and you scrape, let's say, google.com and UK, and you got the rankings. Now, of course, with prompts, it's very similar, but the difference there is that there's a of personalization. um And of course there's memory of previous searches. So it becomes really unique to the user. So almost you can think of it like users might get something different for each prompt. So it's now kind of not affect like before we used to say maybe ranking number one, maybe there was some personalization, some localization, there's differences. But now even if these tools show you that maybe you are being cited at the top, Yeah, every user will see something different. So we're seeing as well, especially in AI mode, these changes in the layout, that kind of now you don't have the template links anymore. You almost move instead of ranking websites to almost citations. are basically, there's almost a citation to your website from the generated content of the large language model. So we need to see how this evolves. If it's gonna be, we're measuring. citations as well, mentions, brand visibility. The kind of, this industry is still evolving around this space. And we will talk more about mentions and brand after a short break. You said brand mentions. I very much advocate for very long time that we should go with brand rather than exact match domains, like online dot whatever, casino, casino, casino. I always advocated for brand because even before AI, brand gave you stability. But I think through these AI overviews and AI, these and AI dead brand is more important than ever and mentions are sort of new links. Where do you stand on this? Again, very divided topic. Some people say, no, you can better do with online casino.123. Some people say brand all the way. So where do you stand? What's your point of view here? Yeah. Beyond the large language model, sometimes brand as well is much easier with certain Google algorithms. So like the Google core update and the helpful content update, you know, it favors brands. Brands signal sometimes it's much easier. So yeah, I do favor in kind of having brands, but you can as well have brands which are kind of using almost exact match domains. And you can still build a strong brand signature if you use uh exact match domains. Of course, what these models are doing, of course, there are kind of two folds. We know that uh they train on vast amounts of datasets, right? So if, let's say, Gemini, for instance, which is the Google's large language model, is browsing the internet, checking credit mentions. If they're going to find the mention to the brand and the mention to your entity. Yeah, that's almost going to be really strong, almost maybe even sometimes more powerful than any. It depends, of course, in which context. you work with Reddit and do you love and hate it? I mean, I have very much love-hate relationship with Reddit every time I think I mastered it, rules are changing. Do you work with Reddit? Is it something important in your work? As you said, Reddit is becoming more and more popular. It has got love from Google. If they are mentioned there, at least your brand, it can help you. So do you include that in your work? Yeah, I think nowadays any, any where you can be mentioned is important. So Reddit is actually a data set where LLM's really use, you know, actually I think there was in the news that Perflexit is being accused of scraping Reddit where they even shouldn't be because they're licensed to Google. So yeah, if, if you notice that the models are using Reddit to train, yeah, you need to be there. You need to be maybe in another. websites where you have reviews. So yeah, it's very important. And as you said, it's not just about links. It could be even be mentioned nowadays. Wonderful. So I think we talked about it in million ways, but what would be your top three things, how to optimize for LLM? So top five, if you can, but top three at least, that would be a takeaway because we just mentioned so many things right now. If you can simplify it and give us top three and explain us how we get to be visible in oh any AI, either chat GPT, AI overviews, wherever. So LLMs love us. What would you do differently than normal SEO or is it a normal SEO? Yeah, in the foundation, the pillars are normal SEO. um if a user is searching, you know, if there's a prompt, a lot of these models and as well Google does query fan out. So what they do is there is a prompt, query fan out, are basically, you know, a number of searches. And These searches now don't just are kind of, they're not just closely related to the query. Even for instance, if you're looking at, know, uh best betting sites, for instance, they might look, for instance, um the query fan out goes like, you can just see license online sports book, you know, and you will come over for this may be event and so forth, which may be close. So um the optimization now needs to cover that you kind of have that content ecosystem, right? Do you have mentioned those phrases? Do you have those in your headings? So that is now an important factor. And of course, we as well mentioned um earlier that you need to be visible in the training data. So if, of course, you're not visible anywhere on the web, you're not going to be cited. You're not going to be mentioned because you're not going to surface. Because the training data is where you need to be visible. And of course, then, with query phenols, they are going using RAC technology and they're busy rounding themselves to search indexes. That kind of mimicking searches that solve your basically you would rank because they have as well ranking quality criteria, that information is returned to the problem and then there's the generated user to the ad stack. So I think if I would summarize in in two points would be, you know, we need to be visible in the training data and we need to do SEO to make sure we're ranking for the main terms and we have the right content. Basically what we used to do before, even with link building, mentions and content optimization. think there is one very, very important point you mentioned here because last few years it was like, okay, AI overviews are taking away from this sort of like how to and all of these things that they can answer. So a lot of people started to shift more to the commercial, know, bonus and convert now and all of that. However, your strategies, not exact opposite, but almost exact opposite, covering all of the educational data to be visible in AI and not caring that you might not get click from it because you get visibility from it, you get new KPIs from it, right? Am I correct saying that? Yeah, that is something that we're gonna see traffic go down, maybe even certain queries. It's really gonna go down. and we're not gonna get enough clicks. But unfortunately, there is the kind of the Google algorithm like the traditional one that needs that content, you know, to survive into this ecosystem, you still need to publish that content, maybe even if you don't get enough traffic to it. Now, we need maybe, the only difference is maybe, Should I merge that content with somewhere else instead of having things siloed on a separate article? Maybe should this content be like longer form content? Maybe we can do it with less content because maybe, you know, it's not gonna rank, but yeah, we need to be there. yeah, I think this is gonna be quite a shift. I think this is why you are director of marketing because people are like, oh, let's just get the clicks, but you are trying to get the visibility and stay relevant to the Google, which obviously I think that's the correct thing. Just a cleanup, bad content, obviously, but create content that Google wants to at least cite, right? You want to be visible. You want to be recommended as a source and you are working actually with a good brand. It's almost like a match domain, but became a brand, life's course. that is sort of, doesn't happen too often when we have got, you know, not brand becoming brand. Cassino.org is another one. I think lot of the success of this is thanks to you and your visionary approach to SEO and marketing. We have got last few seconds, so what would you say, how to survive 2026 in an SEO and in an AI world that is coming? How do we stay relevant beyond 2026 as SEOs or altogether digital marketeers? What are the skills and what is the approach you would preach because you seem to be Not in our time, you are already in 2028 with your thinking, I would say, because most people are like, I'm sorting problems I have now. You're of, I'm sorting problems that are coming my way. So how do we prepare? What is your crystal ball saying? So I think the first thing is that I'm seeing struggle and there's a lot of measures is something that I mentioned before is a lot of sites nowadays, especially in the online gambling sphere, you know, uh because, you know, sometimes you have systems merged with other providers and so on, like sports book, games and slots. There's a lot of JavaScript frameworks, to be honest, the whole web is JavaScript. So we need to make sure, and you need to make sure that your website can be crawled. And I would say crawl with the website itself. There's a lot of mention, for instance, I think this week or last week of showing the AI bots markdown. but that's really kind of flattening the content that you're losing the representation layer. Even if you look at certain like Germany APIs, there's actually the API is context aware. So when they parse the content, they do understand the context. yeah, make sure that the content is rendered properly by AI bots and Google. I think that's the first thing that you need to do as a traditional SEO. And then the second thing is, you know, as I mentioned before, there's this notion of, know, I put a number of keywords, frequencies of keyword matching in the content. And, know, it, it does the way optimization. No, basically, if maybe I use an example, I actually saw this on the web kind of the relationship of the topics that how we see it and how Google sees it might be very different. For instance, you know, chat jukti for Google. rather than being closer to, for instance, German as a language, whether it can be closer to the USA, was maybe it based there. So yeah, then how you will structure the website based on basically Google sees it from a kind of vector point of view, you need to kind of cater for this. And I think this is the notion that, you know, unless this needs to evolve, you know, we're so keyword focused because that's what we understand as a visual layer, we kind of see the language. But yeah, behind the scenes, there's kind of everything is translated into numbers and vectors. uh yeah, the industry, think as SEOs need to shift into optimizing into that manner. Fantastic. So SEO is not dead. Once again, we're surviving. We are just, what do you say, changing. We are morphing into something else. Endless story of SEOs. Thank you very much for sharing oh all of your knowledge, letting us know how to survive 2026 and beyond. and we look forward to hearing from you in another episode. Welcome. I hope our listeners will enjoy this episode. I know it was quite technical, but it is a hot topic at the moment, so we are here to demystify everything that is going on. Thank you.