Generative Engine Optimization Basics

Welcome And The GEO Book

0:01Howdy, welcome back to another fun fold episode of The Unknown Secrets of Internet Marketing. I'm your host, Matt Bertram. We're better known as the best SEO podcast. That's how you find us on all the handles and you know direct match, exact match uh domains still do work. But to bring someone on that uh can can really speak with this with authority, who's actually written the book on GEO. Uh, I wanted to bring and welcome everybody to the show, Ross Higgins from Siege Media. Uh welcome to the show, Ross. Yeah.

0:33Appreciate it, man.

0:34Well, cool, man. Uh, I know uh Brighton just got over. Um, you're out there in San Diego. Um, you know, there's there's been a lot happening, but I think the biggest news is your your your your official book, you're you're you've now made the record books, okay. That that there was there's certain books in SEO, right, that are cornerstone books. And and I I believe uh geo uh is gonna be one of them. So how does it feel to have written the defining book on geo as like it's still being written?

1:09Yeah, well, I mean, we'll see on that if it's the defining book. That would obviously be nice. Uh, we got the keyword rankings also, definitely thinking about that, but not a lot of competition thus far. Uh, but yeah, obviously it's nerve-wracking as things continue to change. I try to write it in a way that hopefully holds up and it is evergreen, not make too many definitive statements uh throughout. So just reread it this week again. Feel feeling pretty good about it so far, but we'll see.

1:38Yeah,

What GEO Actually Means

1:39so to set the table for for everybody from your viewpoint, um, what is the definition of geo?

1:49Oh, the definition of that's been a big debatt, right?

1:52Like it's been a big battle out there of like, is is it different? Is it the same? What is it called? Like all that. So I just wanted to get your yeah.

2:00You think I always hated the what is section of SEO articles. I do have a definition section in there. I can't say that I purposely maybe I should have more thoughtfully made exactly what the definition of it is. I think the most kind of like pertinent, uh I mean it's influencing LMs, these large language models and the practice uh of doing so. Um and to your to your maybe you're leading here, Matt, is how that overlaps with traditional search, I guess is the the bigger question. Maybe we'll get there.

2:35Yeah, well, you know, how it was framed up to me was think about traditional search as like good data hygiene, right? It's good data hygiene, and now you're talking about a new technology that's different than search engines. Maybe it gives you a similar output, but it's a different technology altogether, as a new layer on top of it, where it's building different kinds of insights, it's looking at different things, and it's giving a uniquely different output of how it got there. So you have to do things a little bit differently, and and I think it's like a new layer on top of traditional SEO, it's not one or the other, it's not just good SEO, but that's been a debate out there in the blogosphere, social media that people are fighting over what it is. And I'm just like, like, like there's things that you can do that are working to rank, like just ship it, like go for it because there's a land grab right now, right? Like that we you just kind of mentioned that too. I I went after LLM visibility, okay. And so yeah, so I had a book, yeah. And so so I I I got the trademark on it as well. So um, yeah, so like there's a lot that uh like barnacle SEO, like I don't know if you're familiar with that term, yeah, yeah, yeah. Grabbing onto something that that you can get cited in because what I'm seeing more than anything else is the LLMs are citing methodologies or theories that you that it's referencing maybe somebody or a client or something like that, but it's not listing them as getting credit for it. And I I'm um that's a big question that that I'm working around right now. I don't know if you have some thoughts around that.

4:20Uh so just to restate that back to you, they're doing a data study, and then that data study is not being cited correctly back to them. And is that what I'm hearing?

4:28Uh yeah, it could be a lot of things. It could, it could certainly be uh a data set, it could be a methodology, it could be a quote, it could be it could be any data point um or cluster out there where you're influencing the LLMs, but you're not getting cited, right? Like so you're you're getting uh you're not you're you can tell that it's pulled from your data, but it's not giving

Consensus Beats Self Declared Best

4:55you the credit.

4:56Yes. Yes. So that uh a common reference point I like to give, which maybe is the direction you're going here. We had a client who is not showing up in AI overviews despite being mentioned several times below. And what we looked at for that situation, one of the things I noticed was they were calling themselves the best overall for that thing. And I think what LLMs are doing are they looking for consensus on something. So when you call yourself best overall, you are doing a few things that are challenging. One, you're not clearly carving off a slice of that thing. So say you're best for small business, something like that. You, if you're in the search is project management software, but you're actually a small business solution. If you say you're best overall, LMs normally will give segmentation around that bigger thing. And you're not giving yourself a clear spot in the list that they're ultimately gonna do. And to build on that, your the other people building lists will have the same challenges. So they're probably not calling you best overall. Only you are calling yourself that, normally. And then they may be challenged to describe you consist one consistently, they're gonna pick their own framing. So, by nature of that, you're gonna have inconsistent framing on these other results. So now the LM is gonna merge all these. They're gonna, you're the best overall, someone else is calling you best for something else, third person's calling you best for something else, suddenly you don't belong in that summary. And that's what I see a lot is like a lack of consensus of like we need to be old school view of we need to be one, while it's more like we just really want to be included, included positively for the customers that are best for us.

6:37Yeah, I've seen I've seen that change. I've seen that change, and you're right, like it's moving to kind of listical consensus and the authority or authorship of the person that's saying it. And it's looking at a lot of different factors to decide the the amount it trusts or it wants to reference you. And and I think that that's that's fairly fluid. It's funny, I even have right here in front of me um it's uh a little reference point, and this is a couple couple months old. It's already changing, right? It's like the word best triggers 69.71 percent of brand mentions, words that trigger brand brand mentions, trusted source, recommend reliable prompts, including recommend often mention public organizations like FDA, especially when the prompt includes words like trusted and leading, like this is like relevant, but also it's a little bit stale, it's a little bit dated about they don't like that anymore, right? Like, you don't you can't say you're definitively number one without that consensus, because if you're the only one saying it, or even on your properties, you're saying it, but you don't have a third party saying it, it's not gonna hold any weight. Um, and so so I I totally agree with you there.

7:57Yeah, third third party motions are uh I think we haven't said it yet. The most definitive, in my opinion, not traditional SEO thing there is. It's like affiliate, even reviews on a Yelp if you're a local business. Uh those the third-party review sources most commonly are the newer motion that I think indirectly impacted SEO previously, but now it's a direct thing. It's like for geo, in my opinion. It's like those sources, and you're right. First party LMs are getting smarter, they're ignoring that more often. The counterpoint, I'll sometimes say if you do them well, you're still ranking potential, but you're contributing to the snowball of other people mentioning you. Like for Siege, reference point I like to share is uh our homepage actually was ranking for geo agency in and out. It's not currently, at least from where I'm Googling, it's still there. But we were ranking for long enough that I think that made almost everyone else list us on their list. And now we're in the AI overview, and who cares if we're I mean, I'd like to be ranking, don't get me wrong, but I it's still pretty, it's because we contributed to the snowball that we now have that that visibility.

9:15Man, you opened up a can of worms in a couple different directions. So, like, um, from a from a ranking standpoint, there's a lot of different AEO, GEO tools out there to rank. Uh, we built, well, we built two of them so far, but we measured over 127 uh different points. And I think it measured 11 times over uh the 127 words that we were using. And so, like when someone's checking the rankings, it was only the same, like under 0.6% of the time. It was it always was showing something completely different, and so I think you're right on the snowball. I think also um on the listicles, uh if you have the first listicle, it's like kind of setting that frame up. But you're right, I feel like those have started to fade. Uh if it's if it's a self-referencing listicle. Um, but the review degradation I've seen really happen. Like if you have old reviews, they don't carry any weight. It's kind of like social posting, it kind of falls down what's happening. Like, and I mean, you've talked in the past about PR and how important PR is and how PR kind of owns what is happening from a standpoint of the SEO can only do so much on page, like it really becomes the digital PR, what's going on to influence those results because citations, unstructured citations are incredibly important. Again, going back to the kind of kind of the uh through line that we're talking about with third-party

PR Awards And CRO For Trust

10:55references.

10:56Yeah, I mean, there's a few different ways I think PR is influencing things, and depends on your definition, too. This is another hairy definition thing. I think some people are describing like a best ex article on men's health as PR. Some people call that digital PR. I think it's performance PR, could call it affiliate marketing. Uh, there's overlap that's hairy, but it's it's still impactful. Yeah, I mean, multiple ways. Data sets get cited a high percentage of the time. So that's a good additional data, like traffic source that doesn't currently exist. But for directly impacting transactional, one of the things uh I'm working on in a deck right now that I'm seeing that's impactful is essentially the contribution towards CRO. So if you apply for awards for your company and then cite those awards on your homepage, and then do a press release. I see the press releases in citations a decent percentage. It's not high by any means, but I think it's a pretty strong cyclical process to be a PR agency that helps your clients apply for those reviews or just be aware of the awards or not reviews, awards that you get. Highlight the product-related ones on your homepage. That's just good CRO. And the argument I heard Annie Crusadina of Orbit Media Make, who's a great marketer as well, is like good GO is good CRO. So you're like trying to convince the LLMs, just like you would try to convince a customer. So you're you we're out there fighting to just do the same thing that humans just did themselves. So if we think about our homepage and the attributes on there, we would probably want to tell people, hey, we were named the best in this category for this product. We have 3,000 five-star reviews instead of making them hope they go find it, just like do both, hit them in the face and hopefully they find it too.

12:49Yeah, I mean, you want everything. I was building a lot of doorway pages at one point, like not doorway pages, like let's say that there was uh a conversion bucket on the page. So it wasn't a doorway page, but it was a thinly sliced page that was based on a template, which AI's FYI, everybody that's listening, do not work very well anymore at all. You're you're really selling to just a highly intelligent like human. Like we've trained these things, they're just highly, highly intelligent. Um, and they they catch this sort of thing, but but always the reference point was from Google was put everything in the body of the page that helps them make the decision they need to make without going anywhere else. And that's what um I've seen, if you wanted to put it in a nutshell, makes pages rank better than anywhere else. Like you don't want them going all around the site, you want to put it there. At one point for me, there was a real debate on okay, where do you put and how do you link to to press, right? So, like if someone's mission to you, like now having that canical page of all the places you've listed at, I find to be super helpful. And I remember, you know, years ago, it was a real debate for me of like, if someone mentioned you, do you list it and link to it? How do you reference it? Where do you put it? Like, I I wasn't clear on all that a couple of years ago, but now it's pretty definitive. Like just having like a a one-pager sheet is is really quite helpful uh for for the LLMs because well, they forget stuff pretty easily. So putting it all in one place that they remember it well seems to be working quite well and helpful.

14:40Yeah, yeah. I like I like that playbook. I have been seeing it way more often as well. I mean, I haven't A-B tested it. Uh, I would hope putting the absolute best stuff on your homepage, which is probably what you would do, is enough. But to your point, it it never hurts to add more context and help your customers get more context of just all the many ways you're amazing. Uh, and hopefully you're sub-segmenting those awards to places they belong. Like if you have all those awards and you're like best for PR, you have that on your PR page. That's probably thinking kind of CRO-minded, um, hopefully, what they'll what they'll go to discover and then make a decision on you

Schema And LLMs.Txt Expectations

15:17with.

15:17So, so then that leads me to like two other points, right? Like, so schema, um, like there's been a recent debate, it's odd to me that like schema doesn't work. Um uh I but people are saying it. So schema and like lm.txt files, like what are your thoughts around those two things?

15:37Yeah, uh, not super bullish on either. Uh I I think I'm gonna pronounce his name wrong, so I won't try here, but there, yeah, there was a good study that basically was showing they are looking at lms.txt files, but not really like using it for ranking. It's essentially like organizing information, but not helping prioritize it. I have like a couple paragraphs. I almost considered not including it. The consideration that could be three years from now is for SaaS clients in particular, it's helping companies uh with the MCP layer. So they're being used to help LLMs access uh MCPs, which could be the new version of like downstream that could be an indirect effect right now of like we're helping people that now they're gonna love our product because they have this MCP they can easily find, therefore, other positive brand things happen, and we actually do show up higher in LLMs because of it. That that seems to make sense to me. Yeah, yeah, go eat the schema. Yeah, just to finish the thought on the two questions, schema similar. Uh I see it as like I think most of the time it's if if it delivers a rich snippet in traditional search, there may be indirect effects that create lift. And it may help help organize pages generally. My general thought process is if it uh is used and is needed to be relied on, we often have something wrong on the front end of a site. We're not kind of displaying the information correctly, with the exceptions being those kind of rich SNP elements that very clearly organize and feed the information. I think there are exceptions with like e-commerce characteristics, the different product features, the weight, all those variables seemingly could back into a very long-tailed query that someone would make and could impact that. Uh, I don't think anyone have, and correct me if I'm wrong, I'm not seeing a lot of very clear studies right now that is it's driving a major impact, but uh I yeah, whatever.

17:42Yeah, no, I I I see the same studies. I I've been using more schema around like entity SEO and how how to define that entity and what it knows and the relationship to things, and building just kind of that data later that I mean you can build a a knowledge graph or whatever as well, but it's it's not um to your point, you're right. Like if you put it on the front end and you do it right, then like it should show up. And I find I'm doing a lot of not new builds. Um, I'm doing a lot of like somebody comes with a site that probably needs to be uh rebuilt, and we're doing like a remodel and we're trying to like organize data that also the client is very particular about like their their nav map or what they want to do, and the weighting of the link equity is just everywhere, and you can't do a lot of stuff on the visual standpoint, right? Like, if in a perfect world, I totally agree with you. And I'm I'm trying to go, okay, you don't have a digital footprint at all, you don't really want to change a lot of stuff. We're having to go through a lot of layers of approval to get this done. I've found schema to be really helpful to try to send those signals to the LMs. I I feel like even um, you know, and there's not, you're right, there's not a lot of data around it, but the the the LLM file, like even like the like what was it, the people file. Remember the people file too?

19:14Yeah, yeah.

19:16I I I think it kind of goes back to authorship, but it's just trying to direct traffic of how much you should take weighting of whatever else you're taking. I'm trying to give you that direction because Google with LLMs, I feel like it doesn't matter. Like people are even starting to like create pages and put them in draft mode, and that's like a thing. I I don't have any data around that. Isn't that yeah? But it's like the LLMs are not um paying attention to no follow, just like Google's not paying attention to no follow. So I feel like it's looking at everything, and and it's it's just those little signals that are helping direct it to maybe you want the crawl budget to get to the right place, like so. It's indirect, it's almost like dwell time, right? Like dwell time is not a ranking factor of Google, but all these other things equal dwell time, they're just calling it something different, it's semantic to a certain degree. And and I I I think that there should be some more data, hopefully, that comes out around this stuff. But I mean, I I I don't know, like schema.org is really, really powerful if you know how to use it. And um, I don't know. That's that yeah, it's interdependent.

20:34If you if you don't spend a ton of energy on it and it can be done quickly in the in the midst of other updates, I mean, by all means, I think it's a it's a it's a good thing to do. Um I I I just hope no one's kind of leading with the reason you're oh I don't know how I'm saying you're doing this, but the reason you're not visible is that you don't have schema implemented. That I think is oh yeah, yeah, yeah.

20:55Yeah, no, it's like well if you're if you're side by side and you're looking for the the the the break, like your your side by side. Yeah, yeah. That's where I kind of put it. I mean, what let's flip it around.

Reviews As The Core GEO Lever

21:09What are the things that are proven that you're bullish on that people need to be doing, actionable steps that people need to be doing when they're thinking about geo that are gonna move the rankings that are undebated?

21:22Yeah, uh review volume. So absolutely auditing and optimizing the review process. So there's every business is different, so it could be local, it could be bin size, it could be enterprise. Um, on local, one consideration is and some of these are best practices, but amazingly, some people are not doing them. Are you responding quickly? Uh, are you flagging fake reviews? Literally, your score could be a point lower because someone gave you one star and it's a fake review. You just didn't request to take that down because you have such sample size. Are you responding to people that are giving you five stars and seeing Could they actually contribute elsewhere? Maybe that same mate of like super happy customer, you could keep a pulse on Reddit or other communities that are showing up in LMs and send them that thread when it goes live and say, could you review me here too? That's one route. Um, and that could flip your the the previous response mechanism could flip your stars potentially. Maybe someone was a one and now they either take it down or they give it a three. All that is pretty, pretty helpful. On the B2B front, it's very similar. You can use that context. Uh having a workflow, it's harder to get reviews. So, really, it's kind of understanding do you have the right workflow? We ask clients, what do you give us one to 10 in terms of uh NPS score, net promoter score? And anytime we get a 10, we have a trigger, we bonus our account management team. Anytime they generate a review for us, so we have the operational structure, and you could do that for your own kind of like people in the weeds. People need to be getting paid when they generate a review for you in on your employee on your employee base. So that's how you get more of them. Um that's just some of it. I could probably go for a lot, but yeah, yeah, any immediate thoughts on the review side?

23:12Uh, I I think absolutely you have to have a system in place to generate reviews. I think a lot of people are weak in that. Um, I I've had personal experiences where we've been able for clients to get reviews taken down that don't meet um the threshold. Uh, there seems to be a lot of negative SEO out out there that's starting to happen, even though, of course, there's no such thing as negative SEO.

23:36This is the new negative geo, probably. Actually, uh, hopefully I don't give anybody any idea.

23:40But but but um I I think that uh again, going back to the review degradation, um, I you you need you can get someone to update your existing review and it will carry strength. And it over time it it loses it its its weighting or its strength. And you see that even on social media posts, you wonder why they move around. Um, and getting the review with the the right keywords and the right mentions in it, but again, that response that you talked about is on Google's platform under your profile, that is like additional context around that review that will help rank and for the LMs to pick it up. Um, the thing I would add to that that is unfortunately SEO continues to change, um, the question-answer section of the uh the GMP reviews where I they were gonna they were talking about sunsetting that for years and they finally did it. Um, and that was a really nice strategy to build kind of uh a digital architecture on on on around that. And now now you really have that review lever as as the main thing. Um, I mean, expanding to like mentions, uh, and what is the definition of you know, really getting a backlink uh versus getting a mention versus okay, yeah, we got listicles, we took we talked about that. But but again, that that PR component, and and you were even, I I think you kind of brushed on it, but awards awards are so critical. Because I'm finding in some of the really competitive categories, if you're trying to like stand something new up, you got to prove that you're a business and that you've been operating for a certain amount of time. And and I feel like it's one trust threshold to be like, well, we've won these awards, so we're gonna check that box. But if if you're not presenting like a real business, you're not getting the the fan out distribution that you're looking for. So length of time in industry and building that footprint and strengthening that seems to be something that's at least working for us.

25:47Yeah, no, I agree. I mean, on the review front, one additional detail you reminded me of is I mean, ChatGBT is not directly using Google. Uh, they're indirectly using it, but they're referencing Yelp more often. I see, and you're you're more local expert than I am, I believe, but uh they're using um you people tend to be using Yelp as an afterthought or having it as an afterthought as a small business owner. Like, I see more and more kind of more dormant profiles as more people are spending effort to get the reviews on Google. So it becomes an afterthought. And now the LLMs are primarily using Yelp.

26:24Well, G2, right? Like, like there's G2, there's different and different industries for lawyers or for marketing, like clutch, like anywhere you can get a review, even Glassdoor, right? Like people people look at that. Like, how how do you treat your people? Um, and the LLMs look at that as well. Anything that's out there, it's a 360 view of what's happening. I I think Google is still carrying a lot of the weight. I mean, that distribution seems to be changing uh pretty pretty quickly. Um, but but I mean getting reviews everywhere I think is is more important than that consolidation. Going back to your consensus statement, like like you want good reviews everywhere. And and again, everybody listening to this podcast, like shameless plug. I need reviews. We're at we're at 4.3 reviews, and we have 316 reviews. I need like 50 reviews to get over 4.5. So, like, please leave a review if if you like this conversation.

27:26It's hard to get podcast reviews. I'm with you. Yeah, we've been doing it for like 10 years, and we have got like 17, I think. Um, I need more. You got to call to action, you gotta go in there with like you just did. So you're well done.

27:39So, guys, everybody go tenant check out content and conversion. Ross's got a great great review to talk about.

27:44Conversation, conversation. We got to get the mentions correctly. The citation, yes, you're right.

27:48I apologize. Yeah, yeah, yeah.

27:50All good.

27:52Um, what else do you think would be valuable to add to this conversation?

27:56Yeah,

Community Mentions And Reddit Reality

27:57I mean, the third party we uh I'm really emphasizing reviews. There's indirect versions of that, of course. So uh I still think Reddit is powerful. I've been reframing it recently to community management, and that's a little more future-proofed. And you could also call it reputation management. So the review cycle is reputation management, but community management is like where are your customers going and engaging, or where could they go? And monitoring all those platforms, positively engaging on them, and if possible, actually doing a build function of building a community there if it exists. So that's a Reddit version. If your audience is on Reddit, I think the value of having your own community is real if you have a pretty, pretty large audience size. So if you're national, maybe not local businesses, but national, uh, I think definitely. You could be running ads at at those areas as well. So I think that's a playbook that not a lot of people consider. Like if you see like your business is in Austin, you could run paid ads if it's relevant to most of the Austin subreddit. The thought process being if you acquire more customers on the community that are showing up on LMs and you have a good product, therefore you will get mentioned more positively as a byproduct of that is sort of the bet. And of course, hopefully you just have positive ROI short term. All right, yeah, yeah.

29:19Yeah. So

Do Ads Indirectly Boost Visibility

29:20the two things you said that I want to go down both of these rabbit holes if we have time. Okay, so there's been this big debate, and and even like when I'm debating with my my own LOM team. Okay, so I have a bunch. We've actually started to create personalities around some of our LOMs because people on the team or even clients don't want to talk to my AI, and so we've had to build uh personalities around it. Um, uh a buddy of mine actually built his around Dwight, so he had a bunch of uh you know infographics and images and kind of personality, but but but when he told uh one of the the the LLM told one of the upper management to um you're pre-fired, uh that that didn't that didn't go so well. Um, but anyways, uh back to ads. Sorry, I totally am ADD. So um everybody has debated, and when I was debating my LLM, it's like ads don't indirectly affect GEO, SEO, whatever, whatever you want to, however you want to look at what lens. But I feel like getting the right if you're if you're targeting exact match keywords, you're getting to during the audition period, you're getting the right kind of person to see that page quickly. It helps the machine learning to understand better what that page is about and how the right people are going to interact, therefore showing it to more of those kind of people. So I feel like it closes the gap of SEO. Like when you say, Oh, how long does SEO take? I feel like running ads, if you run it in a way to support what you're doing with SEO, it can help pages rank faster. And that's how I've used it a lot, not like direct response conversion so much. That's a whole different strategy. But but there's been this big debate around using ads versus traditional SEO. And I've always found when you pair them together, and it's not that you have to spend a ton of money, you just need to spend enough to get to that threshold of here are the people spending ads, here are the people doing SEO. If you could just get in that next category and you have good SEO, you'll get some velocity. Um, has been my experience. I I'm just curious how you view ads as it relates to GEO SEO.

31:46Yeah, uh beyond the Reddit one, I mean, uh I think that's fair as an indirect mechanism. The the byproduct to me would be click-through eight lift because you're appearing in both locations or essentially brand lift. So if you have brand lift, you're gonna click more, you're gonna potentially have more dwell time. That over a long enough duration will lead to positive uh lift. Uh, of course, you gotta have you can't fake that forever. Uh, it's it's generally based on real things that further compound real brands because that tends to be correlative, right? You can you can afford good SEO and you can afford good PPC, you're probably pretty good. Not always, but it tends to be all right if you can run multiple marketing channels in a competitive space. Uh, so that probably all generally means you're deserving to have that click-through rate lift. And of course, Google is not a bad uh fan of that, but it's interesting on the LMs too. I know um we haven't gotten to the point of like running ads for clients there.

32:55We're doing it one, we're we're doing it with two clients, one client, but it's been in a small segment on chat GBT, and I don't have any definitive data one way or another.

33:04No, I have yet to hear something positive about it. Of course, it's gonna, I know it's gonna get better though. Um, all this stuff has. So I ignore it your peril, I would say. Yeah, direct impact cards are are harder to harder to argue, like the way it's placed, anyways, right now. It could change how it's placed, but um it's like an afterthought in the in the prompt. Um, it seems like good marketing, of course. You want to show up when you're referenced if you're getting asked, probably not not hurting.

33:34So, so chat GBT from a recent study said it like it's only grabbing the first 200 characters on every page, so it's it's building its own index and it's just building it in the first 200 characters, which you know it's like a little snippet of what what the page is about. I think that's gonna be exceedingly valuable because a lot of people cram a lot of important stuff further down the page. Um, I don't know.

34:01I just want to I see that extractability, scannability. I think that's one byproduct of FAQs, interestingly. That uh I wonder if that's a correlative thing. Like we all put FAQs or generally 200-word kind of snippets. It's not at the beginning of the page. Uh I haven't tested this, but I would assume if there's like jump to links, maybe that's sort of quasi-functions what you're is what you're describing. Like you're giving them it very quickly. Uh, shorter posts are are are are happening as everyone kind of knows these things are occurring, but scannable content I think is never a bad idea, where you give the answer up front. So that will that will continue to be a best practice for sure.

34:43I've been putting it in schema, but uh okay. So Reddit, Reddit, let's talk about Reddit for a second. Um, Reddit kind of got downgraded a little bit. I mean, Reddit was the hottest thing that they did the big deal with Google. Uh, everybody was jumping into Reddit. They were, I mean, I have a client, reputation management, okay. Like, we're trying to get on Reddit as the client posting a response banned because they didn't have any karma. Okay. Like, I'm like, they're they're really, really strict on on Reddit. Um, how do you where do you see that going? Um, because I think it's still quite important. It's definitely that and Quora like need to be players. I updated my Quora account, by the way. Quora, if you're listening, if you're listening to Ross here, like I basically re-turned on my account after like a number of years, and I tried to switch the email and you banned me, and I don't appreciate it, and now I can't get access to my Quora, but because it's really important right now. But where do you see Red Reddit and Quora going? Um, also, I feel like some of the dumbness, I maybe that's the wrong word, of the LLMs uh early on, we're like referencing all these people on Reddit and Twitter, uh because like it's mirroring humans, right? Like so, yeah.

36:04Uh I have the feeling it was always uh the framing I would use when I saw it very high in citations, it's a breadth function, not a depth function. It's kind of like, did I include that source that I hadn't otherwise included? That seemingly was what it was doing, but now it's even dropping that from the calculation to your point. My core recommendation for that that could be actionable for people is if you have non-automated uh review collection workflows, monitor the places where these reviews happen on those communities. And then when you see a thread appear that's relevant to you, then go to those five MPS or 10 MPS people that love you, the five star people, and you could send them an OR function where you say, I would like, I would love if you could also give us a review on Yelp or Reddit if you have an account. Not everyone's gonna have an account, it could be weird to some people that we asked to see their Reddit. But if you give them that optionality and you send that to enough people, pretty good chance you'll get authentic reviews from people and the right people. And say Reddit's big enough, there's enough users we should encapsulate, encapsulate that core might be a little harder, honestly, because so few people do, but that's I I've seen that more rarely in my my citations recently, anyways.

37:20Man, the undercurrent of what you keep saying is guys, you got to build a community, you gotta you gotta build a community, you gotta buy the the the avid listeners, the avid followers, the people that love your product and brand. Like you need to nurture that. Um, you were talking about social earlier, Ross.

YouTube Influencers And Social Proof

37:38And um I mean, socials are now getting indexed in in Google. Uh I'm finding socials being a lot more important. Uh, LLMs are now recently, at least from what I've been seeing, verifying against social of what people are saying, who's saying it on different handles. It's just adding another layer of that kind of optimize everywhere complexity. Um, I'm finding socials to be a lot more important. Also, if socials are advancing and the website's outdated or vice versa, um, that becomes a data point that the LLMs are starting to look at. Um, I'm just curious, any more context around socials you want to add?

38:20Yeah, so YouTube comes top of mind. I mean, it's in some ways, it's yet another review lever, but it's normally done by an influencer. It could be by an organization, but most of the time it's like a third-party affiliate. You can find these through marketplaces. So there's like Impact, uh, partner stack in B2B, uh, Passion Fruit with two O's in B2B as well. They may have some B2C overlap. You pay these influencers up front typically to create a video, like the best CRM software, the best uh running shoes, things of that nature are normally require an upfront payment. It's it used to be affiliates would just do it for their affiliate link because they get so much traffic, but that is collapsing. So now a lot of them are asking for sometimes six to ten thousand for a video. You could do a package of like four videos up to 20k sometimes. You could do a comparison article, you could do a best ex article, you could do your review article, you could even do a tutorial of your product if you have like a complex product. I've heard that being really fruitful. YouTube is by far the the biggest weight that we're seeing, but the indirect versions of that too that we're thinking about. Someone's creating a video, one of those influencers, definitely have them package it as an Instagram reel or TikTok reel, and but and like get that search everywhere component. I see that's the that's what I see in my LM trackers from those networks. It's essentially just a a snippet of a larger video a lot.

39:57Man, I uh I I have I've really seen the rise of and and I love how you're talking about PR. And I want to kind of land the plane there because I I feel like that is really the core concept here is building that community, building that audience set, and how are you gonna reach those people? You're not necessarily gonna reach them on the level you want to create that um standard deviation impact through SEO alone. Like you have to add those bigger visibility funnels to get those spikes, even though they're not um sustained, right? You gotta keep you got to keep doing them, and over time you you're gonna stitch together nodes of uh of clear like brand equity uh across that. And I think YouTube uh checks that box speaking, checks that box press releases. I think check that box. Like you you need to get these engines running on a on on kind of higher cylinders. I mean, what are the things that that you've seen to get the biggest bang for your buck when it comes to general PR or digital PR or however you want to frame it?

41:14Yeah,

PR Strategy As Search Traffic Drops

41:15it's uh interesting timing for our conversation. Was just reading a uh study, and maybe you saw this. 20 news publications are down 28% in the last two years in traffic from search on average. So just getting destroyed. Less publication. I'm working on this for a deck on PR, less publications are here, many of them are going under as a byproduct. I'm concerned about the future of that. That doesn't mean that their PR will not always be a thing, but I think we need to reframe from our classical SEO minds of like we need links on these third-party sites to how do we distribute and promote content? And social is one of those. So one of the things we're working on is paid amplification on LinkedIn, uh, X, like working with influencers to distribute these data studies and being okay with that outcome. Of course, we can't measure a link in that case, but if we truly are building a strategy, that's worthwhile. Uh, and if we're using the data for other reasons, like we're updating content that fresh that LMs love, they love freshness and that data, they love data. There's a lot of other value beyond like the the past goal, which is more and more fleeting of like I want an asset to go viral. That's pretty challenged right now to have a city study go go viral, uh, like it like it used to anyway. So that's how I'm thinking about it. It's kind of that it's not going away, it's super powerful for sure. We have digital PR as a service, but I I think that's how we have to go is kind of leading into those social signals as you're talking about, and thinking of promotion is more than just did we get a backlink that was follow back to our website.

42:58Okay, last question for

How Smaller Brands Can Compete

43:01you. So, like big brands competing against small brands, right? So you're having to before we could tell people focus on one channel, like just get this channel, own this channel, build the community there, like get all the reviews there, like you saturate that, and you'll capture and own that niche of the market. Um, also, there was the strategy of you could be an expert in so many different things, but now you really only have the leak equity to be an expert in what you're truly the expert in. So, like double down on your specialty on your niche. Um, which that I've been having a lot of those conversations with clients of like you we we should close the door on this because that one big thing that you might miss, we're missing the bread and butter over here, and we can't we can't be in two places at once. But also, like when you're talking to smaller brands or brands on the rise, um they they they're really Sensitive on where their dollars are going to go, where the the attribution's going to be. And SEO's always been hurt because we don't have clear attribution, uh, like paid. And so the conversations like, well, where do we put our money? Where do we get the biggest bang for a buck? And if you're building a proper playbook for a brand, uh it really is 360. You you kind of need to be everywhere. And then what that's leading for me to say is we need to build operating systems, we need to build automations, we need to build crons, we need to, you know, we need to have human in the loop or human origin in what we're doing, but but we need to go back to the workflow and and leverage automation because you kind of need to send that signal on these different channels. How you do that. I'm not saying that that's right, I'm saying that's kind of our approach. That's been my approach. Is if you don't have the budget, we need to build your brand set. We need to build like your messaging, like your like I mean, right? Harrow got bought up, or I don't know exactly what that is now. So there's just limited time, right? Like just like this podcast, it's one to many. You can reach a lot of people, you can uh chop up the content like we're gonna do and distribute it a lot of places. How do you how how do you advise like medium to small brands to compete with these bigger brands with these large budgets as the shift happens underneath everybody's feet?

45:44Yeah, uh it's a great question, not an easy one. One of the things that is potentially more optimistic is thinking about your positioning. Like, why do I win deals over someone else? And a thought I've had, I was like, it's possible in a certain market that if each brand ran their kind of prompts that matter to them in LLMs, if they're properly positioned, they would all have high visibility scores in their own prompts because they're defined correctly for who they are. The challenge is when someone who is not going to win against Salesforce goes into those prompts and says, I want to be the best CRM software, that is not going to ever win for them. And that's where they will lose. So I think the reframe would be lean into who you are. And yes, build and add like little incremental things along the way. I think that's that's fruitful. But if we're modest about where we are in this moment and we position very strongly against that, I think it's a lot more feasible to win and then slowly expand out, add a few new things, invest incrementally, and don't jump to who you're not. I think that is more feasible. And 100%, you you are right that there are automations and things. It is challenging. You got to do a lot of things to win, even in that smaller world. Uh, but hopefully we can set up some workflows that will make that faster, especially on the small business side. It's all relative to what your competition is doing as well. Um, so as long as you're best in class there, hopefully you'll be best in class uh where you're sitting.

47:21So, so if you're listening, what I'm hearing Ross say is it's really important to go back to the basics and define what your brand is and map that out. I think a lot of people try to run too quickly, or they say, okay, my my target market is SMBs, like small business, small businesses. Well, what segment inside that, how are you positioned to really win against somebody else? And that's going back to that specialization as we talked about. But building that brand entity, I think is is really, really critical because that that's how you create that digital twin of who you are online. Um I I I think even when you look through the PR lens, it's going back to like what's on brand for you. You you can't and you can't win everything. Um so, Ross, last question. Sorry,

Attribution For AI Driven Leads

48:15I said last question before. Last question, attribution, just kind of a tail-end question. How are you approaching uh tracking in this world? Um, because when I'm talking to executives or when my team's talking to executives, they don't want to spit, they know they need to spend money, right? They know they need to spend money. Now they know that they they have a time clock here with GEO, not everybody's figured it out, and so they're gonna invest in it. And I'm seeing increase in investment in it, but they want to know what they're gonna get for their money, they want to know how they're gonna measure it, and that's where they tend to fall back into like, well, let's just run paid.

48:59Yeah, yeah. Uh a few different things I just heard this week that are continuing to build some thought process here. Our our first most simple suggestion has been telling people to add self-attribution in their forms. So if you look at our website and our forum submit, you'll see how did you hear about us? And it's open-ended. So you kind of leave that for them to input and it'll give them context. We'll give you context about emerging channels. Like we had Claude come out of nowhere in our LM tracker. The if you do a drop down scrape in you, yeah, yeah. If you if you do an LM, if you do a drop down, the risk is they'll pick the first option, they'll randomly pick things. If you kind of make them input it and make it optional, if you're worried about it hurting your conversion rate, that that can help. If you're really worried about conversion rate, maybe your B to C a post-conversion multi-choice, I think is good where you have AI assistant. Uh, this is just make it so you they have to click a button. I think that makes more sense in that case. Ideally, you kind of have it visually. I think Pangram and AI to detector, if if you ever want to check that out, has a great post-login call action of like six boxes that all look amazing. That's a great way to do it. And then in the data itself, uh, a lot of data is often in the wrong bucket. Our new analytics head has let me know. Like it's not attributed as AI assistance, and that's problematic for credit. Second idea I got from Eric Wu of RAMP, and we're gonna have some content coming out on this that was really interesting, is basically any non-home page URL for 90% of businesses to the homepage, the direct traffic to that is a bot. Or sorry, is AI is someone coming from AI? So if we basically group anything that's not the homepage into AI, because we could probably look at before I think it was October 2025, or I'm gonna get the day wrong, where AI really took off, that will represent itself in the data. So now we could carve out a whole slew of probably blog URLs, product page URLs that people realistically won't bookmark. I mean, use your brain for what's bookmarked and what's not, like career pages I saw on some, probably they maybe would bookmark that. A calculator they may bookmark. So you may have to remove some, but do a custom thing. And then the final thing he, I don't think he's perfected, but I'm interested to dig dig into this is people coming from AI are gonna operate differently than people coming from other channels. So if we look at our direct traffic and compare the prior period to now, we may be able to show see a very clear segmentation difference that we could argue should be grouped over with us in AI assistance. And then if you put that all together, and this was his argument, it's actually for them higher than it used to be. They're ramp, they're growing super aggressively. So they may be lucky in that way. Uh for most people, it may be still down, but it's a much rosier story if you group it that way.

52:06I like that. I appreciate you sharing that. I um I've been starting to dig into like the funnel and trying to understand once people get onto the site, where are they going? So we're we're using a combination of like Microsoft Clarity, and then I'm doing UTM codes inner site to like look at where people are going. I'm trying to map out like what the customer journey looks like, and I'm getting all nerdy on it, but it's because also like if we go to the site, like you have to add a lot of filters, your bots going to the site. I mean, I think something like over 49. Well, I'm making up data points, guys, but I I think my AI told me that 49% of all the traffic on some of the sites is like us, you know, yeah, that we're like working on, and so um that makes the the data really noisy, and and so it's it's really about understanding where people are coming from and what they're doing, especially like we have a remote team, so so it really goes back to analytics, it goes back to the basics, it goes back to to the brand.

Where To Find Ross And Wrap Up

53:12Um, so Ross, so we're we're kind of wrapping up here. Uh, is there anything else you want to add to this conversation, as well as like how do people find out all the awesome stuff you're doing, checking out your book, um, you know, where you're speaking at next, your podcasts, all that sort of stuff.

53:29Yeah, uh appreciate you having me on. Um, yeah, I post on LinkedIn, Ross Hudgens, pretty pretty high frequency. Yeah, the book, if you search GEO book, hopefully we're ranking, or maybe not, it might be some geography book. Search generative engine optimization book. It's a little long-winded, or AI SEO book, or just go to our websites in the top nav.

53:48Ross and G I G E O. That's yeah, that's the easy one.

53:52Yeah, Ross Geo book. That will show up, hopefully. Uh, but yeah, it comes out October 12th. So uh excited about that. And speaking at Brighton SEO, uh, sorry, I getting mixed up. HRFs Evolve October 12th. So right on launch day in San Diego. Very cool. But yeah, uh I had a lot of fun being here.

54:10Awesome. Well, well, guys, if you want to grow your business with the largest, most powerful tool on the planet, the internet, keep listening to this podcast for more tips and tricks. Uh, until next time, uh Ross, thank you so much for coming on. My name is Matt Bertram. Bye bye for now.

54:27Thank you.