Loading summary
Transcript36 lines
- [00:00]
Jordan Cooney
The Voices of Search Podcast is a proud member of the I Hear Everything Podcast network. Looking to launch or scale your podcast, I Hear Everything delivers podcast production, growth and monetization solutions that transform your words into profit. Ready to give your brand a voice? Then visit iheareverything.com welcome to the Voices of Search Podcast. A member of the I Hear Everything Podcast network, ready to expedite your company's organic growth efforts. Sit back, relax and get ready for your daily dose of search engine optimization wisdom. Here's today's host of the Voices of Search podcast, Jordan Cooney.
- [00:43]
AI is transforming search results overnight. Google's AI mode is changing everything we know. Search behaviors shift consistently. SEO professionals face unprecedented challenges. How can we adapt to this new reality? Since early 2025, AI overviews has grown by 115%, fundamentally reshaping the SERP experience. This highlights the urgent need for SEO strategies to be redefined in an AI generated search world. Here's the challenge. Traditional SEO tactics are becoming obsolete. Google's LLMs rewrite the rules daily. Search visibility requires new approaches. How can you maintain rankings and search experience when everything is changing so rapidly? This is the Voices of Search Podcast. I'm Jordan Cooney and joining me today is Kevin Indig, independent advisor who helps companies navigate complex SEO landscape challenges. Today, Kevin will share how to adapt your SEO strategy for Google's AI driven future. Kevin, welcome to the Voice of the Search Podcast.
- [01:55]
Kevin Indig
It's good to be on here with you. Thank you.
- [01:57]
Jordan Cooney
Pumped to have you back on the show. You've been on the show before. We're going to cover a ton of ground as we always do. Kevin, if it's okay with you, I'd love to just start off with maybe small personal story. Where are you now? What's going on with your life? What's exciting? Tell us a little background.
- [02:14]
Kevin Indig
Yeah, a lot changed. We just moved to Germany about a month ago. It's obviously a big change for us and we're slowly settling in. We had some starting challenges. I'm not going to go too deep into some of that stuff, but there's a lot for us to be found here. Closer to my family, for example, our daughter is growing up here, having a lot of fun and right now we're just all sweating in this European heat wave. It's kind of crazy, right?
- [02:41]
Jordan Cooney
So until AI can solve heating problems or childcare issues, we're really in a bind. Right? But happy for you. On a serious note, really happy for you, happy for your family, happy that you're back in Germany, closer to mom and family. And I'm looking forward to hearing the stories about how that all goes because I think it's a great opportunity for you and your growing family. But we've got a lot to cover today in our show and I really want to kick off our first real deep dive question and really look at like, what's the fundamental changes that we're seeing in the SEO landscape specifically, how is AI overviews and AI mode fundamentally changing the SEO landscape specifically coming out of Google, AIO and even then many other announcements that came from Google. How are you seeing this changing landscape?
- [03:34]
Kevin Indig
It's the biggest shift in search that we've ever seen before. And if I had to summarize it, it probably came down to the fact that this old generational contract is now broken. So what I'm talking about is that there used to be this implied contract between Google and the web and advertisers which said that websites provide their content to Google, Google sends some traffic, and at the same time they're allowed to run ads against the search results. That was kind of like a three part or three way contract. But now that searchers are getting answers right in the search results to a very elaborate degree, that contract doesn't work anymore. Because the majority of searchers don't click through to websites anymore. They already have their answer. They might still click through when they want to purchase something, when they want to validate, but for the most part they don't click through anymore. And that means that the biggest referral traffic channel of all traffic channels is shrinking. It's getting smaller, at least from a referral traffic perspective. And so there are a couple of huge questions implied, but I would say the biggest one without a doubt is how valuable is it to still be visible? What's the value of being cited in AI overviews, of being mentioned in ChatGPT, being mentioned in AI mode? What's the value of that? There's no clarity around that right now, but it could go as far as to say, okay, maybe at some point if there is no value, then we might not Let Google or ChatGPT Index and Surface our content. So anyway, don't want to stray off too far, but that's kind of the underlying question here.
- [05:11]
Jordan Cooney
I agree. I think that the underlying question is what is visibility? What does it mean to be in Google now? Right. Which is why it's such a huge transformational shift for our industry as we think about the fact that today there's just so limited. Let's just Call it information data metrics. How are you thinking about the foundational aspects of what we're seeing in the SERPs and, and having conversations with clients or SEOs or industry people about what they should be looking for, what should they be looking at when they look at the SERP now, what they should be looking at when they type in a keyword or may I say prompt into a search engine.
- [05:57]
Kevin Indig
So my model and look like this. This model changes and adapts all the time. Right. So I want to be clear, this is a snapshot of the moment, but right now the way I'm thinking about it is really in terms of influence and presence. So are you mentioned in the answer of AI overviews? AI mode like ChatGPT? I kind of put it all in the same bucket, by the way. I also think that AI mode like long term or soonish is going to become the default search experience. So I think, you know, right now there is like the way to go about it is to like look at, okay, how present are we? How many people might we reach through impressions? Google gives us barely any data, but they still give us clicks and impressions. They just don't segment it by which surface we're talking about. But basically like a brand marketer, how many people do we reach? What's the sentiment against our brand and product and kind of what content can we create to be more present and potentially influence the purchase decision of people, similar to how ads would influence that as well. Right. So I think it becomes more of an influence Game metrics have to change. I mean, metrics of success have to change. The ways of creating content have to change slightly. Not that much now, but probably more in the future. And so I think the whole approach changes. But it's very hard to write the playbook right now because so much is still changing. New models are coming out all the time. ChatGPT5 is supposed to launch this summer. So I think we're like in the grand scheme of things, we're also in a big transition period that could last anywhere from the next six months to the next, I don't know, six years. But that makes it hard for everyone to adapt. Right.
- [07:37]
Jordan Cooney
I want to ask a specific question about AI and how AI is changing the way we search, changing the behavior of being a searcher. Google has since launching AI mode, been very persistent at expanding their AI experiences in the serp. So notably, like right now when we're recording, we're seeing even in mobile, the deep research or ask AI to do more deep investigation on a particular topic. So I did one recently. I'll give you the example because I think an example really helps with this. I was on my mobile device, it was about 10pm at night, I got sunburned and I didn't know whether or not you could put Neosporin on a sunburn. And so I typed into Google, can you use Neosporin on a sunburn? And Google gave me an AI response with good bullet points and all this information. And then it said, hey, do you want to go deeper? Do you want to? Do you want to. It asked me if I want to go deeper. So I said, yeah, sure. You know, I'm not quite convinced based on what you've told me, whether or not I should put this on my shoulders because it hurts right now. So I go deeper. And what was really fascinating is it gave me a lot more clarity. It broke things down into two buckets, why it's safe and why you can use it. And then what are some of the potential risks? Fortunately for Google, it didn't tell me that I was going to die like most other articles do when it comes to medical advice. But joking aside, the reality is, is that like, this is going to become a new way that consumers behave. Right. I just did it here. I wanted more information. I want to understand something better. So when it comes to being able to influence, influence that experience, whether I'm Neosporin, I'm WebMD, I'm some research institute that writes about medications and whatnot, how do I think about influencing the initial AI response? How do I think about influencing that going deeper phase? And what, and how do you think about that when talking to the industry, when talking to businesses or websites that are trying to, to gain traction there.
- [09:52]
Kevin Indig
Can I ask you if you clicked on a, on a link or in a citation?
- [09:55]
Jordan Cooney
I did, yes. So actually I did two things. One, one was totally an SEO thing. I started scrolling through all of the citations till I found Neosporin, which unfortunately Neosporin was like way, way, way far away in the results. Like, I think it was like the 10th or 12th option in the carousel of, of pages. But the first one was incredibly helpful, actually. It was like university, it was like some University of North Carolina. And they're the ones that kind of broke down most of this information and actually talked about how specifically Neosporin helps in the healing process. So you shouldn't really put it on the initial blisters or anything like that, but once your skin is trying to heal. Not to get into too much detail here, sorry for the personal information here, but once your skin is trying to heal, Neosporin can be very helpful. So anyway, that's what really convinced me me to end up using it.
- [10:46]
Kevin Indig
To be candid, why did you click on a website, on a link or on a citation?
- [10:51]
Jordan Cooney
Trust. Purely trust. I was like, thanks AI, but I'm about to put something on my body, I don't know if I really should. And then that page obviously was some medical school or whatever. For me it was a lot more trustworthy than just reading off of a bullet pointed list of AI stuff.
- [11:14]
Kevin Indig
And that's exactly the answer. Together with a research partner, I conducted this usability study where we actually recorded 70 people out of the US across different age ranges. We gave them eight different tasks to solve, like research this medical question, find a tax accountant in your area, buy these earplugs, like all sorts of tasks across different search intents and we had them comment their experience, their thoughts on that kind of stuff. So very in depth usability study. And one of the major findings is that trust really is a big incentive to still click through. Right? So you're not like, I mean, you're exceptional and unique in many ways, but right this specific way it's a pattern, right? So people really do that because especially for medical stuff, they don't necessarily trust AI all the way. And so that is still a way to kind of influence users. Now specifically, what I mean here is it becomes a more holistic game. So now in order to succeed in this new AI search world, you also need to ask yourself questions like how can I build that trust before people search? Right. How can I identify that? Jordan Cooney is part of my potential target audience. For some companies that's much easier to very narrowly define a couple of Personas or their audience. For other companies that are more utility based, it's probably much, much harder, it's much more problem based. But anyway, who are the people I'm talking to? How can I influence them in ways that it builds trust before they search, so that when they're ready to search, when they have a need, they seek me out specifically because people actually do that everywhere. Classic search AI modes, AI overviews everywhere. The other thing is there's obviously a lot of context about what types of searches we're talking about. And if it's a basic yes or no type of question, there's no influence in that. Maybe if it's like should I buy X? But other than that these questions are not worth it. So there's a whole kind of type of Topics and intents and prompts that are not worth going after because let's just answer them off the bat. And people are not, they don't want to double check, but everything else is basically creating content, optimizing content. Some technical stuff you need to kind of make sure you take care of. And a lot of that reminds us of very classical SEO. And I would argue that the biggest thing that's actually the most impactful thing that's new and somewhat unique to AI search is that people now search or better said prompt in much more detail, so they're able to express with much more clarity what they're looking for. And on the other hand of that marketplace, you have sites and content providers that for the last two decades have been trained by search engines, mostly Google, to create content for shorthead keywords. Right? So the biggest difference that I see is being able to identify what are the very specific needs from very specific people that are relevant to my business and how can I create content for exactly that.
- [14:11]
Jordan Cooney
Let's talk about that, because my example is, is a very useful one, right? Like, I doubt that like Neosporin is sitting in their labs right now going, boy, we need to make that sunburn version, right? Like, that's just not happening, right? Like, the majority use case for, for Neosporin is probably scrapes, minor abrasions of some sort. They're not sunburn, right? Like, let's be honest, aloe vera is probably much more suited for that particular use case, right? Just using the human brain of like, what is the applicable medicine or remedy for this particular problem. Let's talk about key metrics. When you think about understanding what users are searching, what are the prompts, what are the keywords, what metrics should SEOs be exploring more closely as this search landscape or this discovery landscape is changing, Discovery is changing, how we look for things is changing. My example is one of those that's much more specific than what is probably normal for Neosporin. How should we be thinking about the metrics behind that?
- [15:14]
Kevin Indig
Yeah, such a good question. Because I think there's a couple of different approaches and there is what I'm doing with my clients. So the most common approach right now is to basically use classic keyword research as a starting point and then look for all sorts of questions and long tail variations of your keywords to create content for with the goal of hopefully being visible for some prompts. But the problem is nobody really knows what people prompt. Nobody has information. I would even argue that OpenAI or Claude or Google, even if they see the actual prompt data. It's probably not that useful unless you normalize it and abstract it to a high level intent. So, so what I'm doing with my clients is we're actually going to the people, we're actually conducting interviews. So Neosporin would reach out to you, Jordan Cooney, and say, hey, we noticed you're a customer. Or maybe not. We want to do an interview with you. Maybe there's an incentive, or maybe not. And we want to ask these types of questions like, what do you think of Neurosporin? Where have you heard about it? But also, how do you use it? Where are you unsure about using it? What else are you using besides Neurosporin? Maybe aloe vera. So you ask all these questions and the beauty is that with AI you can analyze and interpret them very effectively and at scale. So we're turning these interviews into intents, into problems that people are trying to solve. We're tracking problems based on these intents and problems that we're identifying and we're building content. Sure, you can monitor your citations, referral traffic, llc, AI crawl requests to your website. There's a lot of hard data that you can then capture. But then at some point we go back to these people or to other people and ask them, okay, have you heard of Neurosporin? In what context? What do you think about it? Do you think there's something better? Et cetera. So we're literally conducting user interviews and we're conducting brand awareness or brand recall studies. And my maybe controversial take is that I think that's required now of SEO teams to do maybe in some cases product marketing teams already do this, or ux, like product UX teams. But the reality is that usually they don't ask the exact questions that you need as an SEO team to then infer problems that people are trying to solve. But I think that's what it takes, though. You need to think more like a brand marketer who has a direct channel to customers and a target audience. And you need to collect quality data at some small scale, pair it with all the large quant data, and then draw the picture of what works and what doesn't.
- [17:47]
Jordan Cooney
This is why I love you, Kevin. This is absolutely why I love you. Because I couldn't agree more. I think it is fundamentally a shift from the old school. Rip down a keyword list from any old tool, sift through the list with my brain, and pick a couple that make sense and then go write copy to a totally new method of user research, customer Research, doing actual interviews and investigation with the end user, the people who are engaged with the product, service, experience that you offer and gaining insight from there to then go and develop content, develop experiences, develop a better web. It's not a third party download of a CSV anymore. That's just done like it's over. And I think it's a hard. This is an important key metric because it's a key process that we've used as SEOs for the better part of a decade now. And it's been a foundational aspect of how we do research, how we make decisions, how we prioritize, how we do all that work. And now it is totally shifting and adapting. And SEOs need to do it very quickly.
- [18:57]
Kevin Indig
Yes, yes. SEOs need to do it very quickly. Teams are still struggling with that. You know, the first response when I pitch that is usually, okay, let me check with other teams, what they're doing. And then, yeah, sales is having some conversations with customers. Okay. But let's push. What if you had a conversation with one customer a week? That's like 50ish, right. If we round it down, it's like 50 conversations a year. It's not that much. Maybe do half an hour. So 25 hours a year spent on deeply understanding who you're talking to. And it's funny because in a way, it all goes back to what marketing used to be. We went to this crazy, you know, the meme of the Jedi who starts, and then there's this huge curve and then it goes back to the basics. Right. It's kind of where we're at right now with SEO as well. It went through this roller coaster where there was and still is performance marketing. And performance marketing works incredibly well. But it also has led us to completely detach ourselves from the people who are actually selling to. And now we're going back to actually talking to these people and asking, so how does that come across and what are you trying to solve and who are you? Right. Don't forget that LLM search is becoming so much more personalized also. So this idea of rank tracking, it's so much further removed than it had already been in the past.
- [20:12]
Jordan Cooney
Correct. That's 100% right. And I think that's one of those are some of the shifts we need to really take into account is that some of these old practices are probably a huge waste of time. And with very limited time in a very fast paced, moving landscape, you got to start measuring the right stuff. And these customer insights, client insights, interviews, those are the places where you're going to gain the most insight and knowledge, no question. I want to go into a practical approach like how we start to do some of this work, right? I want to get your perspective on what is the new playbook for visibility. I'm using the word visibility specifically because I want to erase this concept of rankings. But what is the new playbook for visibility in the new world of search discovery?
- [20:56]
Kevin Indig
So you need new metrics that reflect this new shift in user behavior. So I already mentioned you want to get an understanding of the reach. How many people can you address? 2. In what way do you address them? Is it impactful for them? Is it important? Does it change your sentiment or not? And then three, how does it compare against your competitors? A set of metrics that I like to track right now, first of all is impressions. Impressions out of search console on Google. You can also get social impressions and all that kind of stuff. There's nothing yet for LLMs really. But starting with impressions on Google, only for position one, because that implies that it's probably coming from an AI overview. There might be, you know, it's not always true, but we're trending towards that place where it's probably always true anyway. Impressions on position number one, I call them influential impressions. That gives you a high level idea of like how many people do you address. And then two, I like to look at share of voice, which is essentially very tactically, how many spots are there for search across the AI overview, Like how many citations are there and how many classic search results are there and how many of all of these spots combined do you occupy versus your competitors? So that would be a percentage that you can track over time and hopefully grow. And then three is these kind of like sentiment survey, qualitative type of data points, right? Like what are people thinking about our brand? How do they feel about it? Do they recall it blindly or on a set? And I think these three alone together make for a really good start. I also like some of the visibility metrics in third party tools. I mean, come from a similar shared voice perspective. But I think even that is a decent starting point. But the key to me for any number is always how that number stands in relation to other numbers. That's why I like to use triplets or quintuplets or just numbers that are closely related that are looking at very similar things that keep each other in check so that you're never just leaning too far into one side like, oh, too many impressions. Okay, now we just care. Impressions create crap content, go in the wrong direction or now we just care about shared voice. So maybe you'd be your competitors, but you don't have a lot of reach with your target audience. So I think that kind of relationship is critical.
- [23:10]
Jordan Cooney
So as you think about this set of key visibility, kind of understanding, what are some of the actual practical things that we need to be thinking about as SEOs now? The practition of using that information. And I think this is changing very fast too. We were joking as we were prepping for this that the old tactics of just writing four blog posts and throwing it up for your software used to give you some traffic, right? You'd squeak out a little bit of traffic and you could say, hey, hooray, we did something really well. Those days are gone. That doesn't work anymore, right? So what's that playbook of understanding to then practicum of doing work on a website, doing work in the SEO landscape. How are you seeing that shift? What are we doing differently? How are we scaling our practice on websites? Couple other questions that I'm just going to keep throwing curveballs at you here. Maybe what do we do about tech SEO? What do we do about core web vitals in this new landscape? How are we looking at the practice of doing SEO?
- [24:18]
Kevin Indig
So this is kind of layer of bots or agents, whatever you want to call it, which are the AI crawlers that is important to pay attention to. So from a tech SEO perspective, there are some things that are still important, like can these AI crawlers even crawl all of my website? Right. JavaScript, huge problem for AI crawlers. So those things matter. Server response time matters for any crawler. Right? So you can probably do some optimization there. And then not to forget what happens when the humans still come to the websites. We're not yet at a point where it's just agents and the humans remain in ChatGPT or something. So humans still come, so they want to have a good experience. That whole kind of tech SEO layer remains. But then from what else are we doing? What's different now there's this whole kind of part about content creation, but also content refreshes or optimization of existing content. That is so much more important because don't forget the way that LLMs work. And also Gemini, which is underlying to AI overviews and AI mode is that they have a training data cutoff. So there was a point at which we're trained on data and then any answer they give after that cutoff that relies on up to date information that then becomes a web search or is based on a crawl that the AI crawler has done. And so I think it's important to understand that LLMs rely on very fresh content. The content freshness, keeping content up to date has gained a completely new quality. I forgot where I saw this stat, so apologies for not giving the proper credit here, but I saw an analysis recently that said that 85% of the content that is cited in LLMs is less than 2 years old. And so I think maintaining the content you have gains a new quality. And as you already alluded to, shifting the types of content that you create. So I think, for example, what is articles or explainers or big guides probably make less sense. Original data or data stored stories, digital pr, all that kind of stuff makes a lot more sense now because you essentially want to try to contribute new information and insights to any conversation. And just writing another guide in times of AI is just not that meaningful. Maybe if you're a really big brand in a space, you have a chance to pop out a guide and then get some citations because you have that authority. But for the most part, companies have to think about different content types that, that provide relevant and important net. New information can come in the form of a study or a report or analysis, that kind of stuff. And they need to think about how they're being represented on the web. So stuff like reviews have been important before, but now they really influence LLMs and their answers and so review management, reputation management on other platforms, all these things, like they become important now. And I will also say this is very different from SEO before, where SEO is a lot of people. Some people will now say we've always done that. And yeah, sure, some people have always done that, but the reality is that the mainstream has never thought about that and now they need to.
- [27:31]
Jordan Cooney
Yes, I want to talk about that for a second because I think this is super important. This is the crux of where we're at as SEOs, right? You and I, Kevin, we've worked together for a long time. We've worked at the same companies together, we've linked arms and attacked SEO problems together. And so we've got a lot of in the trenches experience along with the strategic and the directional experience to help lead organizations. The practice of SEO is changing super fast. And I fundamentally think that there are new sets of skills that we have to adapt to. You mentioned a few of them in there, Reputation management. I mean, SEOs used to think about that. They'd be like, I just need a 4.5 star, it'll be good enough to get Me into the, you know, the, the featured snippet area of this. Like that is the old way of thinking though reputation management in and of itself is a, a set of work. It's a discipline of tasks over time. What are some of these new practices that we've got to get better at? What are the new things, the new tasks we should be really, really thinking about?
- [28:33]
Kevin Indig
Totally. So platforms, like third party platforms are now, you know, like much more in scope for SEO. So I think YouTube is the one that has always been the closest to SEO because it's a search engine. Reddit is one that people have a lot of feelings about, but that is also critical now. And third party review sites like G2 for software or Trustpilot and E Commerce and there are many others. Right? They're not the only ones. Those are also important. So the question is now like, okay, cool, so what do we tactically do? Do we just pump out YouTube content? In part, yes. Do we just engage on Reddit? Yeah, sure, all of that, obviously in a measured and proven way. You just don't want to blindly rush to these platforms. There needs to be a system and a structure. But the way that people explore on the web is not just through Google and it's not just on one website they collect. They're on all of these platforms all the time. Right. There's this interesting paradox which I need to find a term for, for that I coin or something. But the fact that all social networks are growing, right, like that's kind of interesting because you would think that people only have so much time and sure, like linear TV time, yeah, sure, that's down. But that cannot be the only. Right. So people are everywhere kind of all the time. And you want to account for that as an SEO, because again, how people know you, how many people know you and how they know you, how they think about you, that matters now before the perform a search. And so in an ideal world, we'd be very active on Reddit, we'd be active on YouTube. We would look at what are people saying in their G2 reviews and then how can we make these changes or reach out to people or educate them better or improve our onboarding or improve features. All these things in an ideal world matter. And in the practical world, it's very hard to influence these other teams. So for example, very specific example, I'm not going to name the company, but there's a company that puts out a piece of software and that piece of software, after a certain amount of time or usage is becoming Gated, so you need to pay for it. The challenge is that that gating often happens when people need the software the most. So there is this kind of like, I got to pay for it now that I need it. So there is some sort of a frustration and friction. Absolutely. And that kind of dial is often being turned up whenever there is a new earnings report coming up. Right. To improve the numbers of the company. So you have these two different colliding incentives and that leads to a lot of bad reviews because obviously people are frustrated. And so that is, as an SEO, you see that in a cool. This reflects in our click rates or just the way that people search about us versus competitors. You can see, for example, that that company, their brand search volume over the last five years just is like a tail slide. It's just a slope. Right. Like just downwards. And so, I mean, that also means that they lose more revenue and therefore they're more likely to dial up the gating again. So you get into this like, spiral, which causes more negative reviews. But then as SEO, you look at these negative reviews and like, you basically have to go to the CEO and you have to say, look, we need to forego future revenues to provide users a better experience so that they don't leave these terrible reviews and so that they're more likely to maybe recommend us or talk about as well on Reddit.
- [31:50]
Jordan Cooney
That's right.
- [31:51]
Kevin Indig
That is a mammoth task and probably impossible at some companies. Right. So let's be real here. The ideal world kind of collides with the practical world because it's probably not going to happen. And so there are some of these situations where we need to be realistic about what's possible. But that doesn't change the ideal, that doesn't change that some of these things are necessary. And so like in every case, when the playbooks change, it's usually going to be young companies and new startups that are going to roll up with a different team formation that are going to bake this in from day one and then attack some of these incumbents and probably win.
- [32:26]
Jordan Cooney
Yeah, the disruptors. Yeah. But I do think fundamentally you're addressing what Google has released. The only real information they've released recently with respect to optimizing for AI is most of the stuff they put in there, by the way, is stuff we already knew. Oh, make your website fast. Oh, you know, use a site map. No shit. And so like that's, that's the basics. Right. But the one piece that they really gave us, that was, that was something to think about. Was the utility of multimodal elements. And I think this is important because it goes to your point, which is like you just gave a full cycle example of how a company is suffering from reviews and how those reviews, reviews trickle into Reddit and how those then Reddit and then you people are complaining on YouTube. All of a sudden you're all your multimodal experiences are negative.
- [33:14]
Kevin Indig
Yeah.
- [33:15]
Jordan Cooney
And that's, that's a total detriment to your future growth. Right. You're spending all this money maybe in paid channels to generate leads, but then those leads are then matriculating into an organic negative yield. Right. And so like it, it goes into this whole multimodal set of expectations. Thinking about that, thinking about multimodal. What's the optimization strategy moving forward now for SEOs?
- [33:40]
Kevin Indig
Yeah, totally. I mean it's to first of all understand the touch points with your audience. There are these kind of like universal platforms like Reddit. I mean, Reddit is on path to becoming the most visible site on the web, only Wikipedia is bigger. But then there are some of these very punctuated platforms. So maybe if you're Nike, right. You're trying to recover from missing the last five to 10 years or missing the boat over the last five to 10 years. You're competing against on and Hoka and whatnot. There's probably going to be a couple of running forums that are going to be important for you to kind of regain your fans. And so you should really think about those. So you need some sort of like an engagement approach as an SEO. Right. SEO is a demand harvesting channel, but that's not enough anymore. You need to find your audience and you can do it again. First party research, like customer interviews. And then you need to be present there with them and engage with them. Right. You need to again translate a lot of that into content. I mean, Reddit is a goldmine for great content ideas because it's mostly based on real unincentivized people, but then also like video. I mean, we spoke a little bit about YouTube. One thing that we found in this study, in this usability study that I mentioned earlier is that people absolutely prefer videos for tasks. Right. So if they're trying to figure out how to do something, like how to, I don't know, apply sunscreen on your back when you're by yourself, just go back to your previous example.
- [35:01]
Jordan Cooney
Too soon, too soon.
- [35:05]
Kevin Indig
Trying to casually slip that one in there. Yeah, plant the. Yeah, exactly. But that would be, you expect a video there and people will often either start on Google or directly go to YouTube. And so I think that's a great way there to be present for brands. Sure, there's still some production effort for YouTube and videos, but it's come way down. And so you want to think about when does their experience or the part of the user journey require a video, or when is the video a better way to consume that experience. By the way, AIoViews mention YouTube videos a lot more, and YouTube generally is a big source of training data for LLMs. So I think it's important to not underestimate it, especially because we don't have that classic ranking system on YouTube. It's all about relevance. Do you have the answer that an algorithm is looking for? And do you have some engagement men? I'm not even sure about how much engagement actually plays into the relevance aspect. So, anyway, long story short, figure out where your people are, provide content in the format that makes the most sense for what people are trying to achieve, and then measure that with the metrics that we laid out earlier.
- [36:18]
Jordan Cooney
Last, real kind of deep question here. A lot's changing in how consumers are discovering content information, maybe in these multimodal experiences, companies investing more specifically to be more tactical in certain places, like your example there around video. When it comes to instruction. I want to get your perspective, Kevin. How are these new LLMs and these AI search experiences, ChatGPT perplexity and the like, how are they changing this search landscape? Are we as SEOs in a point of real competitive differentiation between Google and these platforms, and how quickly should we be investing in them?
- [37:03]
Kevin Indig
It's a tough one for Google, and I think the biggest sign of that is that AI mode basically looks like ChatGPT. Right? Basically. And again, my. My theory is that AI mode is Google's end game. That's basically what they're going to drive more of the search experience towards. And Sundar Pichai has said that on podcast interviews. And so the fact that it looks so much like ChatGPT kind of shows how big of a threat ChatGPT is to Google. And so I think we have to be careful because on the one side, ChatGPT users are growing fast. Referral traffic from ChatGPT is growing fast. ChatGPT is the biggest LLM referral source. It's growing fast. Users are pretty qualified coming from LLMs, and at the same time, in proportion, the amount of referral traffic and even referral users is still tiny. Right. So I think it is a great time right now to monitor, to go for some quick wins. And I think it's not a good time to throw your SEO strategy overboard and just say we're now just going after LLMs, right? This is going to be like a sidecar situation for the foreseeable future where you need to be present on both of these tracks or search engines and LLMs. The good thing is that in the majority of cases there's nothing you would do right now for AI search that would be harmful for Google search, maybe cloaking, but how many serious companies are really doing that? So right now the Venn diagram is, is very, very overlapping. In the future, I think that might change to some degree. But right now I think what's beautiful about this is that it blows new wind into the need of streamlining new SEO basics. Jordan, you've been in this forever in SEO, right? I've been in this for a little while myself and man, some of the basics are still not sitting tightly, you know what I mean? And so if this is a wake up call to finally streamline your basics, then I don't know what else is. And I will also say that another opportunity in this whole question, like how much should I invest in LLMs versus Google? Part of that question is also how much can I automate myself, right? I think there's this question in the room of like, who benefits more from AI? Is it kind of search engines and LLMs or is it SEOs? And right now I'm not sure. I think SEOs have a profound opportunity to use AI to automate a lot of workflows, increase their efficacy by at least a magnitude, if not two magnitudes, and almost be so far ahead of Google and LLMs that they need to watch out that they're not getting flooded or spammed or I mean, they are already. But for Open Web and SEOs to win this game, it's very interesting. If you look at some of the interviews with Google leadership, like Liz Crofton, read some of the recent interviews, they acknowledge that there is this kind of cat and mouse game between Google and SEOs to some degree that also exists for LLMs, because of course we're trying to figure out how these work so that we can influence them. And implied in that is do we maybe get the bigger weapons here to kind of influence LLMs that LLMs cannot defend themselves against and we're kind of winning that race to some degree. So I'm fluctuating my answer there, but I think it's a really interesting one. And for me it's a push to just say, like, look, dare to automate some things. Play with this. Build your own tools. Really lean into AI, not just in terms of search, but also for yourself.
- [40:36]
Jordan Cooney
100%. Okay, that wraps up this episode of the Voices of Search podcast. A huge thank you to Kevin Indig for joining us. Very grateful to have him and his insights on the show today. If you'd like to contact Kevin, you'll find a link to his LinkedIn profile in our show notes or at the voicesofsearch.com you can also find and learn more about Kevin@growth-memo.com which is a newsletter that he publishes. If you haven't subscribed yet and would want a daily stream of SEO and content marketing knowledge in your podcast feed, hit the subscribe button in your podcast app or on YouTube and we'll be back in your feed. Okay, that's all for today, but until next time, remember, the answers are always in the data.