The SEO Playbook for the AI Overviews Age

SANDI (00:03):

Well look at us back for our season finale of Season one of the Chocolate Pill. 

ERIN (00:09):

Should we let listeners know? And it's so funny, I was listening to this other podcast and they were like listeners. They're like, sorry, we're jumping ahead, listener. 

SANDI (00:21):

Appreciate you. Yeah, more specifically. Hi mom. Hi mom. Hi, I'm Judy. 

ERIN (00:28):

But we were just chatting and we're like, we need a seasoned finale. Should we just record one now? And we're like, okay, there we are. Everybody. 

CHRIS (00:36):

I think you guys sell yourself short. There's probably some guy in his basement watching it right now. Oh God. 

SANDI (00:42):

Don't talk about my ex like that, 

CHRIS (00:44):

Chris. Oh, ouch. 

SANDI (00:47):

Actual footage. Yeah, actually pitch 

ERIN (00:51):

Ruby. But yeah, welcome to the season finale. Everyone's like, thank goodness of Chocolate Pill. And I believe we wanted to talk about what AI and SEO. 

SANDI (01:04):

Yes. Sing it. B-I-N-G-O-A-I-S-E-O. 

CHRIS (01:09):

AI in general, I think relatively right, just in terms of what it's doing to the whole industry and what we're seeing in terms of how it's changing marketing, I think. 

ERIN (01:19):

Anyways. 

CHRIS (01:20):

I think that's interesting. 

ERIN (01:21):

It is interesting. And Chris, you just did a lot of research and deep dived, deep dove into it for hours, for hour 

SANDI (01:30):

Deep. He plunged. He plunged into this plunged 

ERIN (01:34):

Feet first said first for some of our clients because they're battling this, right? Because everything's changing with ai. And so you did a ton of research on it, which I feel like puts us and the people we work with ahead of the game because I think a lot of people are still figuring it 

CHRIS (01:52):

Out. Yeah, I think so. And it's more like treading water, right? And I think that's how a lot of marketers feel right now is they feel like they're treading water. 

ERIN (02:01):

Less of a plunge, more of a tread. 

CHRIS (02:03):

Yeah, absolutely. So with SEO in particular, it's been really interesting over the past six months, even from when we did our last podcast episode about it, about SEO. A lot has changed. So the models are evolving really, really rapidly. If you look at what Google is doing with Gemini, if you look at chat, GPT, if you look at cloud, all of them are evolving at such a pace that it's hard to keep up. So how do you that? And so even what we thought a few months ago is outdated by this time. So how do you approach that? So it's really interesting. I think you can actually use those models to help you do that research. So if you are a marketer with any background or experience, you could take sources. If you use a perplexity or you use even the other models on a deep learning mode right now, you can cite many sources, compare it to your past approach, and then validate a new approach or direction and even give you some ideas. And if you can do that at scale, you start to have more confidence in what you're doing. So that's part of what we've done, I think with clients, is to look at that and look at, I think for us, I looked at hundreds of articles over the past, maybe since last November or so, and trying to see a direction, trying to see ideas of how you leverage those large language models. A lot of the informational search, I feel like I'm talking a lot. Am I talking a lot? 

SANDI (03:30):

No, you're good. Keep going. Keep going. But I do have a question for you, but finish your 

CHRIS (03:32):

Thought. Okay. I was going to say with so much is changing now, there is zero click. So all those little blue links, you're not getting the traffic anymore. They say up to what, a quarter, a third of informational queries. What is this are being done within those large language models now, and they're not coming back to you. So a lot of businesses, like we said before, I think we'd mentioned last podcast, a lot of people have lost their businesses between the algorithm changes and AI within the past six months. But I think there's some really interesting opportunities as well with all of those changes. 

SANDI (04:12):

So question about that. If you are an SEO manager right now and just trying to figure out what the heck to do, why are my numbers plummeting? What am I doing to get people to the site? How am I going to reach my performance goals and how do I think about keywords and long tail keywords? And it wasn't too long ago that our advice to clients was make sure that you have the q and a, the FAQ content on your site, and make sure you've got glossaries, you can catch people who are looking for specific terms. I think what I heard you say just a minute ago is so much of that has changed and people, once you make that query, people are finding themselves in the LM. So they're finding themselves in a place other than your website. So what's an SEO manager to do at this point? 

CHRIS (05:11):

Well, I think the most important thing is don't throw the baby out with the bathwater. You have so much content on your side already. You need to learn how to repurpose that. And so there's a transition from all of that keyword work you probably created, pillars you probably created supporting content around that, all links to that pillar. That's really common with a lot of marketers, content marketers, is to be able to do that well. How do you adapt that? Right? And so there's a couple things is to see how these large language models look at things. They just don't look at the keywords anymore. Obviously, I think when we've covered this before, they talk about intent. So there's the informational queries. What is this to the how do I do this to the, how does this compare, right? And so I think it's really interesting. I think the next step for a lot of marketers is to actually create a matrix, I think, of your current content and then go out there and score it and to be able to see, okay, these are really the content areas or the topics in general that we feel that we should be an expert in feel we should be found for. And just audit yourself in terms of not just do we have content there, but do we have content that matches the intent for all three of those different areas and score it? 

ERIN (06:30):

Yeah, because like owning certain topics, but then it's also structuring the data in a certain way. Google's looking for certain things. You want to talk a little bit about that? 

CHRIS (06:42):

Yeah, I mean, when someone goes to a large language model, all the answers are contained within it now pretty much. So anything from a definition, which we know that the 40 word definitions right now are being picked up a little bit easier. And some of the longer ones, which that's even a change from what we've seen prior to things like how to some deep comparison or listicle type things that lend itself well to that comparison. So all of those, you're never going to get picked up unless A, you have content that's aligning to that intent and then B, that you're creating the schema. And maybe that's not the right word. The JSON ld, that's the kind acronym for it. But to be able to create the structure within the H TM L to say, Hey, this is what this page is about. And there's many types. I think what we did six months ago was we said, Hey, let's apply this to our blog pages and our articles and things like that that we're producing on a regular basis. We may not have done that on an organization wide basis or a product basis or some of the other pages on our website that may have more weight, that may have more traffic that we're not telling these large language models what they are. So that's a miss, I think. 

ERIN (08:05):

Yeah, and taking advantage of other areas. I know we had talked about this a little bit, AI systems, they read Wikipedia, LinkedIn, yeah, 

CHRIS (08:19):

The backlinks. That's really interesting. We kind of went through an evolution where we had these, what would you call them? 

SANDI (08:27):

Backlink 

CHRIS (08:27):

Farms. Oh yeah. Oh my gosh. You'd pay 9 95 for whatever, and you get these backlinks to these farms and those days are long gone. But I think other approaches that marketers that take it, Hey, let's look at the top 10 results for this topic area or this phrase, and then let's go down the list. Hey, if we get one of these top 10 to link to us, that'd be great. Let's reach out to 'em. It's probably still valid. But what we're finding is that these large language models, they have sources for their data. They have places that they go for data. So wouldn't it be better to go out and to see how you could bolster the content that you have within those places? So someplace like LinkedIn, for example, and you're looking at the topics that you want to be included in the conversation's already there, whether it be an individual or a group. And if you're not involved, that conversation giving thoughtful answers within the context of that, oh, then you're maybe missing the boat because that's one of the places that I think is really important to be mentioned by these large language models for sure. 

SANDI (09:34):

So this is where I'd ring the bell for product marketers, right? Oh yeah. Because typically product marketers, we tell our story on the website. It usually unfolds from the homepage onto the products and solution pages. If you're B2B, we, but we've traditionally sort of been a little bit of hands off on social unless it's related to product launch or maybe some competitive kind of campaign where we've provided content. But what you're saying now is we need to actually apply that same discipline to the other channels, which means we need to show up in those channels. And also means we also need to think work probably much more closely with content, Erin, on just making sure that the story that we tell in the audience, we want to tell it to kind of meets the moment with respect to how we want to position product, how we want to talk about it. 

SANDI (10:26):

Really, I feel like what it does is it almost expands the aperture of the buyer journey beyond your company website. Because usually if you think about buyer journey, it's like how are we going to pull them into our website and tell them a story and keep them here and then convert it into quality conversation MQLA response form, a demo request, whatever you want to call it. But Chris, what you're suggesting is no one's coming to your house to hear your story. You now have to evangelize. You need to actually go the other way and be in all of the other places where the conversation's happening and know that you can't not sacrificing one kind of platform, for lack of a better word, for the other. You're in fact telling your story not only at home on your website, but in all of these other places. And 

CHRIS (11:15):

For 

SANDI (11:15):

That, that requires a ton more thinking with respect to how you tell it, what language you use, the appropriate mode for it, right? Is this video, is this narrative, is this, who is telling the story? All of the, I feel like it's the discipline of the message from product marketing with all of the art and creativity and data backed insight that come from content and SEO basically we were built for this. You guys, you thinking about integrated, integrated marketing. This is just another application for it. So I guess I think I've talked myself into out of being scared about it and now saying like, oh yeah, we're literally taking the show on the road with A-I-S-E-O. No. 

CHRIS (11:58):

Oh, absolutely. I mean, you talk about monitoring those sources, right? Well, if you're motivated within your current job, if you're not motivated, you tend to do the bare minimum. If you're motivated, you say, okay, what are these sources? We should really be involved. That conversation, that's a natural thing to do. What is less natural, I think is adapting to doing things in a new way. Erin, you could probably talk to, okay, let's have a single source content methodology, create one piece of content, and we're going to put it out on all these different platforms, but they all require different formats, right? 

ERIN (12:34):

And that list has expanded because it used to be like, okay, these few different mediums that we all as content marketers got used to like, okay, yeah, we'll be here at X, Y, and Z. But now when you think about, okay, well, okay, gosh, it's not just social and video and anymore. And then I was like, okay, where are they? Okay, now AI models are now looking at Wikipedia. AI models are now reading this, and it's like, where are they reading and where should we be? And now there's a paid play too, right? I mean, now it's like you see ads being inserted into AI overview, and then it's like, okay, well now we need to get into there as well. 

CHRIS (13:14):

Yeah. That's a scary thing. If you don't have a strategy for how you're paid and your AI search strategy is evolving, you're really missing the boat, right? There isn't something that's evolving there. Do your landing pages have schema on it? Right? Are you structuring your landing pages in a way that those are findable, that the traffic that you're driving from those AI overviews, the paid ads that are starting to appear, are you leveraging it back to those pages? That's a lift, that's a potential lift and an inclusion that you might not have otherwise. 

ERIN (13:47):

Right? Exactly. It's like I think between the three of us, or I know between the three of us, we're excited about this. Oh, absolutely. 

SANDI (13:57):

I mean, I was nervous for a minute, but I came around. You're around. We 

CHRIS (14:03):

Got, yeah. 

CHRIS (14:04):

This is a case of one of those where the entire channel, the entire demand channel has imploded, and you have to relearn everything. And the lessons and what we thought we knew a year ago is different now, and you just have to be able to adapt. That's the scary part. The cool part is that I think there's an opportunity within the next 12 months or so, especially for those companies that have a deep dataset. If you're a company that has a lot of data specific to your niche industry that's unique, oh my gosh, you're sitting on a gold mine. You can create that or go through the beginning stages of creating MCP, which is like an API for ai, large language models. If you can create that early, get it out there in terms of being able to leverage at GitHub and these other places where it gets picked up. That's interesting. If you could do that, if you can then create those search helps within those different platforms, and you're creating a framework around when you talk about this type of information, here's a way that we're going to go about and do that. Here's the way that you search for that within the platform. Oh, that's on marketing. So all of a sudden you have these plays where marketing is now working with engineering, which FINRA software company. That doesn't happen really often. When's the last time you worked with engineering in a software company? 

SANDI (15:35):

Yeah. Very, very technical product. A very, very long time ago, cutting edge stuff where only the engineers understood it and there wasn't enough product management to play middle play the middle. But yeah, it's rare. It's rare and exciting. 

CHRIS (15:55):

Yeah. I mean, becoming the trusted source, I think is the holy grail. And you don't do that unless you think a little bit differently and how to work cross-functionally a little differently. And I think we're talking about working in a more integrated way on two levels. One is within your own marketing team, how do you have paid advertising content and a traditional SEO role work better together to accomplish these things? And then how do you work better cross-functionally within an organization? And that's really, I think the companies that figure out how to do that faster will come out on top. 

SANDI (16:30):

Yeah, totally 

ERIN (16:31):

Agree. Yeah. Agreed. So we will keep our finger on the pulse ear to the ground. 

SANDI (16:41):

It's ear to the ground. Air to the air. Air, yeah. Just get down there like the pope down low. We're going to kiss the ground what we're trying to say. Yeah, poorly. Cut this out, please. Let's, let's hope our podcast editors kind to us today. I know 

ERIN (17:02):

They'll keep it in. He'll keep it in you 

CHRIS (17:04):

Guys. I keep thinking he's going to give me the villain edit. I'm really interested to see how this works out. 

ERIN (17:09):

I know what personality going to give us all. 

SANDI (17:11):

I have to record my confessional interview about Chris of like, Chris always wants to talk about SEO, just like all day. 

CHRIS (17:19):

I know. It's a little boring. Sorry guys. 

ERIN (17:22):

We should do that on the real world. We'll have our confessional. Oh my God, Sandi. One times out of here. That'd be amazing. That wraps up our first season. Do we cry? Do we have cake? 

SANDI (17:45):

What do we do? No, I think we leave it as a cliffhanger. We leave it as a cliffhanger. We'll come back with season two when we will all had facelifts and better makeup and lighting, and we're going to Chris's back screens. There'll be a new intro song. 

ERIN (18:01):

I know. We need a 

SANDI (18:02):

White lotus cliffhanger. We should have wrote, we'll have different hair going to go ahead and get extensions. 

CHRIS (18:06):

Are we going to have a summer break montage? 

SANDI (18:10):

Possibly. 

CHRIS (18:11):

We 

SANDI (18:12):

Could do that. Possibly. Yeah. Mine's all going to be the dogs. It's like, where did I take Ali and Ivy? They had the best summer ever. I just carried treats and poop bags. Summer. That's my summer. Amazing. But yeah, maybe we'll pick up on it on AI when we return for season two, because as someone told me, which was kind of ominous, it's only going to get better. It's really only going to get smarter. It's not like us where we start to lose things. So there'll probably be even more changes, and then hopefully, well, not, hopefully that's the wrong word. By the time we're back, we should also have that decision that could potentially be a landmark decision about Google and it's paid ad business and search and just what that does, because every time they change the algorithm, it ruins our nights. So this is probably going to be a 10 x kind of change in the 

CHRIS (19:06):

Industry. Absolutely. And not just SEO. You guys already use it in very progressive way with other areas within marketing. Right. And a lot of marketers are very afraid to share those things. They're like, oh, this is proprietary. If I share this and our competitors get it, they already got it. Both. They got it. They probably do it better than you. 

ERIN (19:26):

Yeah. That's why we're super transparent and we will show everybody all the tools that we use and what we do, because all in this together for sure. And in the interim, between the season and next, we'll still also be publishing articles and putting out posts just to keep everybody in the loop as we learn things. We'll share it with you and we'll all figure it out on this messy journey together. For sure. 

SANDI (19:50):

Yeah, likewise. Share two things. What'd you do on your summer break if you had one, and what did you learn in between? In between then and now? 

ERIN (19:59):

Yep. No, those are good ones, for sure. Well, but yeah, we're tuning in to our first season. Yay. Yeah. We're excited to keep this party going. So thanks again for joining us on our ship Barking on our cruise 

SANDI (20:16):

Storming the castle. Thank you for storming the castle Storm. The castle. 

ERIN (20:19):

Thank you for storming the castle with us on Chocolate pill.