Google AI Mode Is Almost Default: What Marketing Teams Need to Do Now
- Jessica Bush
- 2 days ago
- 21 min read
Season 1, Episode 03 | Conversations with Vibe & Logic | April 13, 2026
Google AI mode is expected to become the default search experience within weeks, yet most marketing teams either don't know it's coming or lack a plan. In this episode, Stephan Bajaio and Trevor Stolber explain what AI mode is, how it differs from AI overviews, what it means for your traffic and measurement tools, and what VibeLogic has spent over a year building to prepare for it. If you run a marketing program that depends on organic search, this is the conversation you need to have now.
Watch the Episode
What We Cover
00:02 What Google AI Mode actually is
01:23 AI Mode vs AI Overviews: the real difference
01:50 When it goes default: what changes for marketing teams
04:03 How AI Mode behaves differently across query types
06:30 Is SEO dead? Stephan's response
08:06 Why SEO should have always been an intelligence function
10:37 The content has already been taken: what that means
12:07 Is this actually a smart move by Google?
16:31 What if the rollout is slower than expected? Does any of this still matter?
19:13 What VibeLogic has been building to prepare
19:53 Vector Alignment Score: what it is and why it matters
21:00 The PAIS framework and how it fits AI Mode
21:50 Query Lytics: Google Search Console on steroids
22:41 Why Bing is suddenly worth paying attention to
27:12 The gap between having SEO tools and actually getting results
30:26 What to do in the next 30 days
Full Transcript
[00:00:02] Jessica: Google AI mode is expected to be the default in a matter of weeks, and most marketing teams either aren't aware or have no plan. We've been building for this for over a year. That's what we'll be talking about on this episode of Conversations with Vibe and Logic. Let's get into it. Trevor, we've been talking internally about AI mode for a while now. For our listeners, can you describe what it actually is and what it looks like when you're in it?
[00:00:28] Trevor: Yeah, absolutely. This is basically the evolution of search and where we've gotten to with AI search, and it's not to be confused with AI overviews, which you get in the regular Google organic search results. This is a new search mode. They call it AI mode, and it's much more akin to a ChatGPT or LLM-based search. There are no 10 blue links anymore. What it does is generate an answer based on a lot of data that it pulls and gives you a contextualized answer back instead of the traditional search results. That's the new search experience that Google is testing and rolling out.
[00:01:09] Jessica: And just to be clear for the listeners: AI mode is different from AI overviews. Most people are familiar with AI overviews. Can you draw a distinction between the two?
[00:01:23] Trevor: Yeah. The AI overview is still quite similar in the way the answer is generated, but in AI mode it's much more of a conversational-based search experience. Whereas an AI overview is still your typical search, and where Google sees a need and where it matches the query, they give you this AI-generated response done in much the same way, but the search experience itself is different.
[00:01:50] Jessica: Stephan, Google's strongly signaling that this goes default in early May. When that happens, what changes for a typical marketing team?
[00:02:00] Stephan: Well, I think a lot of things could change, and I hate to be the SEO that says it depends, but the reality is we don't necessarily know. Do I think that Google would just flip a switch and move over to AI mode completely? Probably not. They would probably announce that they were planning on doing it, give us a heads up as to some semblance of when, and then roll toward that. Why? Because we're going to have to look at Google Analytics differently. All of a sudden the metrics are going to change drastically. What can we track? What can't we track? What does it look like when it comes across as traffic? Rank trackers and all of the products that do that sort of thing, which a lot of SEOs and marketing departments have come to depend on, will have to change the way they're analyzing the data and the search results, because that will be a completely different set of results to understand and look at.
So traffic is definitely going to be questionable as to what happens. Will it still be organic traffic? What does that channel bring you? Your metrics might go a little bit haywire, and in addition, your ability to tell what's actually performing and what's not. We don't even know how Google Search Console would react to that. There's a lot of big questions out there.
It used to be that Google would just drop something big and then the world would have to respond. They've made changes to that approach over the years. So I think this is more of: if they're going to go this direction, they're going to give us a little bit of a roadmap to help us out, because helping out the SEOs is ultimately helping themselves. I think Google won't do it haphazardly, at least I hope not, and they'll approach this with a little semblance of: here's the base layer, let's go from here.
[00:04:03] Jessica: Trevor, you've been watching this closely for over a year. When you look at it across different query types, informational, consideration, decision stage, does AI mode behave the same way across all of them?
[00:04:16] Trevor: No, it really doesn't. And nor does regular organic search. The SERP is very intent-driven. They try their best to match the user's search intent and they meet that with SERP intent. Those are two different things and you have to understand them both.
I've been looking at a whole bunch of different queries and they are variable. One other thing that gets discussed a fair bit is the fact that they're not always going to produce the same results either, because there's some personalization, some contextualization of your previous queries that influences the output you're going to get, which is a big problem for tracking as Stephan alluded to.
Typically what this is going to be: top of the funnel, informational-based queries largely, I predict, are going to get pretty decimated in terms of actual clicks and traffic through to a website. That doesn't mean they're not valuable. It doesn't mean you shouldn't show up. A lot of our approach to AI mode and AI search in general is just to be part of the conversation and making sure you show up everywhere that you should.
They still cite resources. I did find a query recently where there were literally no citations whatsoever. They are sort of consuming the knowledge of the internet and using it as their own, which is really kind of interesting.
There's also a big push for these AI searches and LLMs to integrate a commerce layer, so you never even have to go to the website to complete your purchase. That has big implications on the revenue model and the business model for them, as does where and how they intersect ads with the results. I'm sure they'll figure it out. There's one thing Google has always been very good at, and that's figuring out how to monetize search behavior.
So we're watching this closely, and we've been building systems, tools, and processes for this for well over a year, back when it was in its infancy in Search Labs. We're well prepared for it. It's going to be interesting for sure.
[00:06:13] Jessica: There's a version of this conversation where people hear AI mode and immediately think SEO is dead. Stephan, you've been in the industry long enough to have heard that before. How do you respond to that?
[00:06:30] Stephan: This isn't a small change. I don't want to call it cataclysmic because I think that's too much whistle-blowing, but it's on the magnitude of "not provided." If you've been around for a really long time, you know what that means. Google used to provide the actual queries that led to your site in the string they'd send over, and we were able to tell how many people clicked on a given keyword exactly and tie it back to specific pages. Ever since then, and even before then, Google would add or remove something and everyone immediately yelled, "Oh my God, it's the end of SEO." This is potentially the evolution of, as Google wants it to be, a different kind of search engine. Does that mean SEO dies? No. People are still searching. People still have needs. People still want to find stuff. Therefore, an SEO is necessary. Call it search engine optimization, call it search experience optimization, call it whatever you want. All the acronyms in the world don't change the fact that people search, they have needs, they want to find things, and somebody has to help companies organize their data in a way that helps them be found. That's really all a person tasked with SEO is tasked to do. Now, where that gets searched and how it gets searched, that's just getting more complicated. But essentially it's the same thing. Different engine.
[00:08:06] Jessica: You've said that SEO has always been sitting on top of a really valuable data set and that the industry treated it as a channel instead of an intelligence function. Does AI mode change that, or does it prove it out?
[00:08:22] Stephan: I think it proves it out more than ever. It's scary to think, to Trevor's point earlier, that you could come back with a search result in AI mode that doesn't cite anything. Sorry Google, I don't think that works. You didn't just come up with that information out of thin air. It came from somewhere at some point. You can't just ingest the internet with no guardrails and then claim it as your own. So I think there will be pushback, both legally in that space and also in how sites decide what gets crawled. All of that will evolve over time.
Search data is incredibly valuable. In SEO, we've been trained to always be the result, and in this case I'm kind of glad it's happening because it's forcing us to recognize that being the result wasn't always the answer, no pun intended. Sometimes it was about understanding what was showing up that influenced people's decision-making. A brand is not the best, unbiased version of its own products. Other people saying what they think, whether it's on review sites, in forums, in social, in articles: everyone's gonna have their own bias, but it's not going to be the same as hearing you're great directly from you. We've always known this. We just never really paid attention to it as much as we probably should have. That's the stuff that makes for a consumer's understanding and trust. They don't build that trust by showing up on your website. They've already built it by the time they get there. That's where we have to focus our time and effort.
[00:10:37] Jessica: There's a line that you both have used that I think is worth talking about, and that is that the content has already been taken. What does that mean and why does it matter?
[00:10:49] Trevor: It's already been ingested. Google has this big index and the LLM is just an extrapolation of that source. A lot of the AI searches end up going out and doing regular web searches anyway to pull the data back. But to me it's really like featured snippets. This is not a small change, this is a big shift in the industry. SEO has been proclaimed dead many times. I don't think SEO is dead. In fact, I think it has a big future. But it really feels a lot like when featured snippets came out, which was the first time there was a piece of data in the search result that didn't end in a link. That brought up the rise of zero-click search. Now that's a strategy. Now we try and achieve those featured snippets. It's being part of the conversation. The data has already been taken. It's already in the LLM. There are technology solutions to stop that from happening, but they've got most of it already. For me, this is just a progression of search: starting with keywords, then phrases, then phrases became topics, and now it's really just part of the conversation. That's what I see.
[00:12:07] Jessica: Stephan, I know you have a point of view on whether this is actually a smart move by Google strategically. What's your read on that?
[00:12:16] Stephan: Right now Google is a definitive search engine. The moment they become a configuration of Gemini with a search experience, which LLM is a better search engine suddenly becomes a real question. And now you're comparing Claude and ChatGPT and Google to each other. Right now in the average consumer's mind, those are different things. Yes, people do searches in the LLMs, but not in the same way they do in a search engine. And so if you start bastardizing the experience into a chatbot one, which I actually don't even think is the best example of an AI technology being put to use correctly, I think what you do is you take your most valuable monetizable asset and you leave it for Bing to be the only one serving up traditional results.
I understand you're losing some market share. It hurts. You're not used to that, Google, I get it. But to go off and over-index, pun intended, I think it's the wrong move. I actually think it hurts you in the long run and equates you to other products when you shouldn't be.
[00:13:51] Trevor: I agree with a lot of that. My approach is more just based on the technology side of it: we've got to be preparing for it anyway. And it's actually really aligned to what we've been building with Web Presence Intelligence, even before AI was even a thing. The approach of being part of the conversation, seeing where all of the options for visibility are, is just good marketing anyway. It plays really well with AI mode because if you want to show up in AI mode, you've got to be part of the conversation. It's not just your website anymore. Whether or not they actually roll it out as default, it's already there. A lot of people don't notice it yet, but it's already there and available. And even if they don't make it default, the solutions and techniques you use to show up in all of these other places where you can find visibility, being part of the conversation, is just good marketing anyway.
[00:14:50] Stephan: Yeah, and this is maybe a different exposure of the same thing. Google was always trying to figure out relevancy based on other websites, who was having conversations and who was linking to you. The idea of understanding what sites are authoritative on the web and which ones are talking about you, and how that authority then moves to a particular page based on a topic, is nothing new. All we're really saying is: change the dynamic in understanding what else is showing up in their path that is influencing the conversation. Understanding what is most likely to cross paths with your audience, and not looking at it from a navel-gazing perspective where you must always be the answer, but rather you must know where answers are being provided and be considered as part of those conversations: that is a much more holistic, multi-channel vision of the world. And frankly, that's more like how people actually consume information and search.
[00:16:31] Jessica: Let's play the other side of the scenario. What if it doesn't become the default, or the rollout is slower than expected? Does any of this still matter?
[00:16:51] Trevor: Oh yeah, absolutely. Whether it does become the default or not, a ton of people still go to Reddit. A ton of people still go on social channels. There are niche-specific forums, there are videos people watch. Those are all places you need to be showing up regardless of Google. If Google went away tomorrow and those were the only places that existed, you should be showing up there. It is just good practice to make sure you're being visible everywhere.
[00:17:22] Stephan: Yeah. Whether it's Google, whether it's an LLM out there, ChatGPT, Claude, Perplexity, or whatever we haven't even thought of next, it's going to point people places. Those places are not always the end point. Recognizing those results as opportunities for influence is fundamental. The problem is that we've lived our lives in marketing silos and channels where individuals don't really command multiple channels except at the CMO or VP of digital level. They're kind of looking through a very myopic lens. And that's not really a way to influence your audience. The web doesn't work that way. Whether some new version of Google rolls out or Google stays with its multi-result type approach and continues mixing in AI overviews and other things along the way, which they have been doing, there's still a huge opportunity that has been ignored for too long. I'm kind of happy this has happened because it's forcing marketing organizations to look at online marketing differently. You cannot keep living in your channels and expect to drive results the way you did yesterday. It's just not the way the new web works.
[00:19:13] Jessica: Trevor, you've been preparing for this for a while now. Can you tell us what that preparation actually looked like: the things you've built from scratch and then how you work with tools that already exist?
[00:19:33] Trevor: Yeah, absolutely. There's been a lot that's gone into this, a lot of technology solutions, and a lot of that comes from understanding at a pretty deep level how this content is understood and interpreted, because it is different from an index. An LLM-based search is very different structurally, so the tools we need to understand it are different. We've been building a lot of solutions that can deal with that.
One of which is our proprietary metric and method for understanding how keywords and content align together, which is our Vector Alignment Score, or VAS. A lot of the tools and techniques done to analyze content are what's called lossy: they do dimensionality reduction, you lose data, you lose information. The Vector Alignment Score doesn't do that. It properly and fully understands the vectorization of the content and then helps you understand how a set of keywords aligns and scores to this content versus another piece of content or another set of keywords. We can look at that in summation and understand: in order for this content to be relevant to these groupings of keywords and conversations that we want to show up for, it scores this. And if that's not high enough, we need to make some content adjustments.
Another technique we developed is what we call PAIS, which stands for Problem, Application, Industry, and Solution. Because the industry needs more acronyms, right? What this does is break things down into the more conversational approach, which is very much aligned to the Web Presence Intelligence approach of breaking things into topics and personas. What kind of questions are people asking? Not just how do I rank for this keyword, but what are the actual problems people have? What are they asking? And how do you position your content to help solve that problem along the journey?
We also have an internal tool called Query Lytics that really helps understand these questions and how people use them. When we can do that, we have really good visibility into how queries interact with the content that we have.
And then you asked about the tools that already exist that we use to augment that data. There's a whole host of things. You've got a bunch of commercially available tools. The key thing is understanding how to interpret that data, how to build it into your workflows, how to take that data and analyze it correctly. There's a huge amount you can do in Google Search Console, which is a free tool. Not many people can really get the full use out of it. And even where Google Search Console has limitations, you can solve those with different platforms and APIs.
One big potential winner out of all of this is Bing. Bing Webmaster Tools, which you should definitely go set up if you haven't already, now gives you AI analytics. They show you which pages are being cited, which queries are being cited, and how you're performing in AI. Google doesn't give us that. So that's a really useful, informative data layer available for free, and not many people are leveraging it.
[00:23:27] Jessica: Let me ask a few follow-up questions on each of those tools. Vector Alignment Score: you're saying it tells you whether your content is likely to get picked up before you even publish. What does that look like in practice?
[00:23:43] Trevor: Yeah, it doesn't necessarily guarantee it'll get picked up, but what it does is tell you how good a fit this piece of content is to the keywords you're trying to target. Often the keyword research is an afterthought from the content, or you have a bunch of keywords and then you're like, okay, well this piece of content needs to fit that. But what we see often is they don't all play together well. What the content says and how the page is positioned in the title doesn't necessarily gel. The Vector Alignment Score helps you understand: does your content really speak to not just the keyword, but the summation of the score for a broad set of keywords, which tells you how well that content is likely to perform for that keyword set. And when I say keyword set, I mean the topics and conversations that people are having.
[00:24:39] Jessica: For the PAIS framework, how does that fit into AI mode? Is it something that helps marketers respond to this change in a different way?
[00:24:54] Trevor: Yep. What this does is help you understand what those conversations are, and it dovetails really well with our Web Presence Intelligence approach, which is based on topics and personas. You have to understand those. It's not like "best iPhone" or "top 10 iPhones." It's the actual problem that someone has. And that's very conversational, which fits really well with the LLMs. But they're things, they're questions you should be answering anyway. If your content doesn't answer your audience's questions and you haven't put that content in the places where they're searching for answers to those questions, you're not showing up in the conversation. That's what it all comes back down to. You have to understand the conversation and you have to make sure you're showing up in it.
[00:25:55] Jessica: Query Lytics sounded really interesting. What would this tool surface that another SEO tool someone's already using might not?
[00:26:07] Trevor: It's basically Google Search Console on steroids. You get to see how the different queries are behaving with the content that you have, and you'll understand: we thought this page was actually about this, but really our users are telling us it's actually more about this. That might suggest you need new content for that term. It might highlight conversations and questions that you didn't know were being asked. There's a huge amount of data available if you can mine it in the right way, and then you have to understand it as well. This gives an additional contextualization layer that you don't get with Google Search Console: this is what the content is doing, this is what the audience is asking, and here's either where it does or doesn't intersect.
[00:26:57] Jessica: Stephan, from a positioning standpoint, a marketing leader listening to this probably has a Semrush subscription they're not fully using. What's the gap between having access to tools and actually getting results from them?
[00:27:12] Stephan: Well, that's a pretty big question if we're moving toward AI mode, because we don't know what those tools will ultimately show. There's a large gap in terms of: will we still be able to analyze search results the same way? Will those search results change to be super personalized? Right now, results were always personalized, and we have methodologies to kind of look at how given search results would come back without that bias. How do all those products deal with that? There's an entire industry of third-party products looking and watching Google and trying to show you what actually shows up and how you show up. All of that information is kind of going to be up for grabs. But listen, none of these companies just showed up yesterday. They've seen countless algorithm shifts. They haven't been sitting on their hands saying this'll never happen. They've been making plans.
I would say, CMO, Mr and Ms CMO, whoever you are out there: know that your data is incredibly important, not simply because you need to report performance to the board, but because you need to set a strategy that takes your audience's understanding and what they're going to see into account. The better a job you do at that, the better you'll be doing marketing.
[00:30:26] Jessica: So for the people who are listening, running a marketing team at a B2B company, sitting on a decent chunk of organic traffic: what are the two or three things they should actually start doing in the next 30 days?
[00:30:33] Trevor: Yeah, I can definitely take that one. First, get Bing Webmaster Tools set up if you don't have that gold mine of data available there. Next, you've got to really understand your current AI visibility: where you're being cited in LLMs. You need a visibility report. That's something we do. There are a bunch of commercial tools out there, but we evaluated a lot of them and a lot of them are really just generic, rushed to market just to have a solution. The solution we have, we're using an API with additional data that actually gets down to: are you showing up for the right personas? And that brings me to the next part, which is you really need to be thinking about topics and personas. That is core and we built a solution around it. Do those things and you'll be better off than most people.
[00:31:27] Stephan: Also, go get your schema in order. Go take a look at your actual content and make sure it's structured in a way that makes sense and can be ingested by the LLMs, not just Google but every LLM. Think through it as: am I really answering the core questions? Do I know what the main questions people are asking when they're looking for what I do? Am I actually answering those questions or is it marketing fluff? Marketing fluff doesn't have a lot of mileage moving forward, because all of that tone of voice, I hate to say it, isn't necessarily what Google or any other LLM is going to carry over into their answers. They're going to try and strip that and provide it back in supposedly egalitarian, unbiased terms. But it's kind of funny because we went to Google because we were looking for bias in the first place. That's the whole point of a search engine. Regardless, they're going to try and bring that information back without necessarily always citing you or always putting a link back. So your content matters now more than ever. Even a single page, even a paragraph, can be its own thing on the web. Go back and really analyze your content, your journeys, your personas. It's a fundamental activity. It doesn't feel like it's going to get you going forward because it feels like looking backwards, but sometimes you have to prepare the stuff you have for the future you want.
[00:33:13] Jessica: You mentioned Bing a few times, which is surprising to me and might surprise our listeners. Can you go a little deeper there?
[00:33:25] Trevor: Yeah, I mean, if AI mode drops, Bing is going to be your signal center for how users behave. It still gives you that layer of data that you might not get from AI mode. A lot of people dismiss Bing because it's only a certain percent of traffic. But Google's market share is declining and Bing has a meaningful chunk that gives you a good signal, especially in B2B. In certain industries, Bing paid ads are actually really effective. They're like a quarter of the cost of Google typically, and the conversion rates are usually better. What you find in certain industries is their tech stack is built on Microsoft. Typical corporate users all day are on their work computers, and when they search for something, whether personal or professional, they're doing it via Bing. B2B actually has a really solid hold in the Bing space and can drive not just paid but even organic conversion metrics you would not have expected. And the kicker to me is the irony: if you have Google Search Console set up, you can go to Bing Webmaster Tools and authenticate your property with Google Search Console. That always makes me laugh.
[00:34:57] Stephan: I'm just going to go over the top on this one because it's super important. People are trapped in Microsoft Teams, people are trapped in the Microsoft environment. If you've gone down to that little search box at the bottom of your PC and clicked on a result, you're not going to Chrome. You're getting Edge, and then immediately they're trying to convince you that should be your default. The reality is there are a lot of people who don't think like marketers. They don't care what search engine or what browser they're using. They just want to find what they're looking for. So when you're in that experience, everything is tied to Bing. Often in B2B, you've got people who all day are on their work computers searching for things, and they're doing that via Bing. We have to get out of our preconceived notions and recognize that Bing has the potential to be a real game changer for a lot of people if they play their cards right.
[00:36:31] Jessica: Last question, for both of you. When AI mode actually becomes default, what's the one thing each of you will be watching most closely?
[00:36:40] Trevor: From my perspective, it's just going to be what data is out there. How do we know who came to the site? How do we know our content was part of the result? How do we know someone came in via that method? I would love it if Google would give us a similar level of visibility as Bing does. It'll be really interesting to see what happens. I sort of can't wait because we've been building for this for a while, but it will definitely disrupt the apple cart when it drops.
[00:37:17] Stephan: I really can't imagine that at their conference, Google just drops, "We're doing this, we flip a switch and it goes." That feels very irresponsible, frankly, because there are literally billions of dollars that flow through the secondary markets that have to do with this. I think it would be too haphazard for them to just flip the switch. However, we've seen them make significant changes before. I think core web vitals is a great example where they gave you abilities to measure and things you could look at beforehand. If they do this from a judicious perspective, recognizing the amount of change it really looks like to those on the backend, like us, explaining what can we expect in Google Analytics and Search Console, giving other platforms a chance to prepare: there are a lot of question marks and I think they're going to have to answer a few more of them before they just flip a switch.
[00:38:27] Jessica: Awesome. Thank you both. Every time we record one of these, I leave with something I want to test or dig into a little more. As for our audience, here's what I leave you with: open Google, turn on AI mode, and search for the question your buyer is asking when they're looking for what you do. See if you show up. If you don't, that's the gap and that's what we fix. We'll see you on the next episode.
Comments