AEO: Get ChatGPT to Recommend Your Brand

This episode is brought to you in part by Swim. Here's the thing about wishlist apps. Most of them just sit there. A customer saves a product, and then nothing happens. Swim actually activates that data. When someone wishlists a product, you can trigger price drop or back-and-stock alerts and feed that intent directly into Clavio or your CRM. You're not guessing what people want because they've told you. Plus, customers can share wishlist for gifts and your team can view them to offer personalized service online or in-store. And unlike card abandonment, wishlist data is permission-based. These are people raising their hand saying, "Hey, I want this. Just not right now." Swim's been around for over a decade. It powers 45,000 stores and installs in about five minutes. You could try it for free today at getswim.com/curt. That's g-e-t-s-w-y-m.com/curt. [Music] All right, today's show, we've got a repeat guest, an SEO expert. There's no shortage of them on the show because I love them. I personally enjoy SEO. It's a fun topic. And two years ago back April 2024, we had Patrick Reisson. And at that time, he was like, "Hey, did you know you can get Chatchy-P-T to work in a Google Sheet?" I said, "No, I've never done that." And I did that. And oh, wow, that was like, that was the moment that really opened me up to the possibilities of what you could do with AI and automation. And so, you know, credit to Patrick Reisson for pointing me in the right direction and really opening up a Cannell Worms there for me. But two years later, he has not stopped. He has continued with SEO and AI in that research. You know, not long ago we had Matt King from Domainon and he was talking about a then very emerging field, GEO, which is like SEO, but for AI. And now we've got, you know, I think that's going to be a recurring topic here. But I'm happy to bring Patrick Reisson because he DM me recently and said, "Hey, I got a case study where we got a coffee brand to rank in Chatchy-P-T pretty consistently." And that's the thing I mentioned because I've had clients asking about it. And I recently read an article from the BBC. I don't know if you heard this. It's in BBC Tech, where a journalist got Google's AI overview Gemini to say he's like the ultimate hot dog eating champion. And it was just by adding, I think it was just by adding structured data to blog posts on his own site. And so, like, it's very much feels to me like, wow, SEO kind of got rebooted here. And a lot of what became black hat tactics now is just a thing you could do, you know, potentially. But I've not attempted this. I want to hear it from the expert who has achieved success with this. Patrick Rice, thanks for joining me. Thank you, Kurt. And I appreciate your appreciation. Sorry that I overloaded you with Chatchy-P-T and Sheets. But it's been a while. I mean, when we first talked, yeah, we were, you know, nerding out about G-P-T and Sheets. And now, I mean, you have like open claw agents that are talking to each other and like, the world just keeps changing. And then on our side, agencies been growing and we've been lucky enough to work with a couple great brands. Two in particular that we've been able to actually have results for within the GEO sort of space. Lots of my friends are also doing all kinds of tests and WhatsApp groups where doing a similar thing to what you mentioned, where it's like, if you have some kind of longer term query, you can get the AIs to say all kinds of crazy stuff. So really early days, but a lot of fun and a lot of like, unknown in the future. So it's an exciting place to be and I definitely think people should be considering it. So I did not mention the name of your, your agency or business, but it's pretty easy. Patrick Riceco. We keep it simple, PRC is what the team calls it. But yeah, we're Patrick Riceco. We actually have eight people now, so we're a good little boutique agency. And that's a fun size. Yeah, I like it. It's been really beautiful like just being able to mentor people and like having them be really excited about everything going on. So I've been having fun. So since you're last on, we were talking a lot about tools. I love tools. You also got me hooked on screaming frog as well. Good. It's an incredible tool for especially for like migrations and then just checking for pad links and Shopify stores. And since then Shopify has added support for signatures, which is a way to say to authorize a link scanning tool like screaming frog to go through your site high speed without triggering a block. If you just attempt this without the signature Shopify will rightly say stop you go, too many requests, you get a break now. Yeah. And in Shopify's favor, that's actually great security. When I first saw that on screaming frog, I was impressed by their security. Because it is kind of hard to crawl a Shopify site. So the signature is really nice. Yeah, now that we have that, no longer hard, it was very easy. But assuming you're comfortable with screaming frog, which it's a little technical, but for this kind of work, I just followed a few YouTube tutorials and then I was over the hump, I figured it out and I've been using it consistently since. But that, you know, that was April 2024. What's different now? How have your tools changed? Back then it was, it was SCM rush, screaming frog, and keyword research. What's that tool stack look like today? Yeah. And then another one was Matrixify. So I'm big fan and Matrixify. Oh, we all are. My number one favorite Shopify app. Like what would you do without Matrixify? I know it's in it. I kid, but it's a utility, but it's such that you can export import almost any entity, like product collection, customer, whatever from your store in a spreadsheet, update it there, analyze it, whatever, and then even update it, push it back into Shopify. And it just gives you a lot of options, has good documentation. And app we all know and love. My tools stack has changed with really only two main tools that I would say are like very impactful, which is first SEO gets. SEO gets a shout out to them, small tool. What it does is that it pulls in all your Google Search Console data. So Google Search Console best data for SEO. You get all the keywords, all the sort of data for Google. SEO gets will give you that data, make it way quicker, give you all kinds of cool views, extract all kinds of knowledge from the data. So a really great tool. We spend every month. That's like how we choose our opportunities, how we analyze success. We annotate all of our changes in SEO gets so we can see before and after, negating brand search of whatever we want to do. So SEO gets is a big one. Lots of opportunity there. And then moving into some of the AI stuff. This is brand new for me really just in the last six months. But our favorite tool right now is profound. It's very expensive. So the alternative that I've been using as well as prompt watch, kind of just depending on I think they all kind of do a fairly similar thing. So I've tried gum shoe, peak prompt watch, profound LLM's refs in maybe one other, but those are the main ones. For me, I believe I picked peak just because I like the pricing. I think I was comparing prompt watch peak and most of the one you said was the most expensive profound. Oh, profound. Yeah. I saw profound pricing and I threw up on my shoes. That class of tools. What's it called? What's it do? Yeah. So it's AI visibility tools. I would say don't get too caught up in the AI visibility measurement tools. I think that they're very useful for a few key things. But I do see even myself kind of getting too wrapped up in measurement. Measurements very difficult in AI right now. Attributions very limited. It's tough for executives and people that want to measure stuff. But it's very hard to measure this stuff. The class of tools though do work for one main metric. Or KPI, which is visibility. Ignore all the other metrics. The tool will report on for the most part. The main thing is visibility, which is if I open up my chat GPT or my Google or whatever AI I'm using. And I ask it a question like what are the best coffee brands? And I ask that question just over and over and over again on different days, different times, different accounts. I just keep asking that question to the AI. And then if I took all of those answers and then I extracted out the brands from each one that it recommended, which are the ones that show up most often. So what is the percentage that your brand shows up out of all of those answers? That's pretty much the only metric we can really get an idea of other metrics like what position it's in like are you the first brand dimensions or the fifth brand fishkin just did a great study on this. The the AI's are so probabilistic and kind of random. But like you can't really judge those pieces of data. You can just be like okay, there's a few brands that AI usually recommends more often than other brands. Like how can we increase how often it recommends? us somewhere in its answer. Okay. Those tools allow us to pick a set of prompts around our niche or our industry and measure that across different AIs. Similar to traditional SEO, we're looking at search engine ranking position. I have a keyword or keyword phrase, probably a long tail keyword phrase. And then when I search it, where do I pop up in those search results? Am I, do I appear at all? Am I in the first 10? Great. I'm in the top five, top three. Am I number one? Oh my gosh. And that was much easier to digest. Also somewhat harder to do, harder to game as those algorithms got more and more sophisticated. So what's different here? The big difference is in SEO, we have a keyword tracking. So you can see over time what position you are. And they aren't pretty consistent. Usually the top three will be more consistent than everything else. So they'll be kind of like, some consensus on Google and being on like, what are the top players? I do think you have a similar thing here with AI. It's just much less a, honestly, it's kind of always been this way, but with Google, it was easy to convince ourselves that like there was a simple, you know, what, what, what number are you from one to 10? So there's always kind of probabilistic because like depending on your Chrome browser, depending on your history, depending on the results to you. Yeah. So AI is just the next level of that. So it's all probabilistic. It's like, like our success here with the recent case study is getting us to be mentioned in our desired prompts, like 70% of the time. And that's like getting any higher than that. Now we're just kind of, you know, inching up a little bit as much as possible, but it's kind of that it's all probabilistic. So one thing I've, the trends we've heard, the pain points for merchants, for Shopify store owners in 2026 are my customer acquisition costs are higher than ever. And everybody feels it and pay my organic search traffic. You know, the somewhat free traffic I was getting from Google down 30 to 40% is not unusual. And so we're looking for alternatives. Is traditional SEO dead? No. I find SEO in Geo to be very similar. But there are some new tactics that if you're looking for more free traffic and sales, that can work well in the AI space right now. And in Geo, that's a little bit different than SEO. But I don't think SEO is dead. I actually think this is a revival of it, a little bit of a resurrection of SEO, because it's been, it's been up and down like always. Yeah, we've heard SEO is dead countless times. Yeah, that's amazing. It's quite interesting. Even I'm like, will it ever happen maybe? But I think as long as AI's are rooted in the internet and information on the internet, it's almost always going to be a factor, right? Backlinks are just other websites mentioning you, your own website, how it talks and mentions you and conveys the information. You know, AI's are just eating this up. Yeah. Let's apply it in a practical approach. You've got a client, purity coffee, and the goal was to get them to appear more often in these recommendations. What was the prompt? What was the question we're trying to get it to appear in? Walk me through that project. Yeah. So we had a list of topic buckets, and I am going to censor some things so that, you know, I don't get too much information now. But we essentially, on profound, you have the topic buckets. So we chose about 10 topic buckets, you know, looking at different types of the product, different attributions of it, sort of how they wanted to show up on those prompts. And it is a sub niche. So it is good to choose something that's a little bit more niche. We did similar to SEO. Similar to SEO, right? If you're trying to compete in running shoes and you're competing with Nike, I mean, I'm sure you're right. Yeah, what are the best running shoes? And you're going to be competing with Nike Reebok, et cetera. Not going to be really able to get your foot in the door. And so then we had about 10 prompts per topic bucket. Just for various ways that people might search these, Varen Fishkin in his study, and I can share you the link. It's Spark Toro did it. Very good. They saw that AI's do kind of come across an understanding of intent, similar to Google, where if you are asking for a particular thing, even if you ask it 10 different ways, typically we're seeing that you get similar brands back. So like if you say, you know, best coffee versus what is the best coffee, like you're typically going to get the same sort of results, it does seem like, you know, there's infinite variety of how you can prompt, but it does, you know, kind of core less over like the same entities or brands that it sees across the internet. Here's a word from our sponsor, Zipify. Hey, the holidays are over. Traffic's down. Add costs are still up. And you're staring at Q1, wondering how to keep momentum going. Here's the move. Stop leaving money on the table with every order. And click upsell automatically increases average order value by up to 30% with AI powered upsells across your entire customer funnel, launch pre purchase and post purchase upsells for every product in your store in one click. Customers get personalized offers and spend more as a result. Unlike other apps, you only pay for those results, not views. It takes under two minutes to set up and it starts working immediately. Q1 doesn't have to be a slump, make every order count. Well, you rebuild traffic. Go to zipify.com/curt for your 30 day free trial that ZIP IFY.com/kru and let's start the year profitable. We have likely players. We have key players that it becomes likely to recommend. Do we have an example prompt? Yeah, I mean, example prop would be something like what is the best coffee? But yeah, so it's essentially something like that. Another example might be, you know, what are the best lightweight waterproof tents, right? So you want to kind of go a little bit more long tail and then it's going to recommend a specific set of answers. And so those are going to be the topic buckets. And then the AI tool will be able to check visibility across those CF how often those brands are being recommended. And that's pretty useful for just getting an idea of your success. And then beyond that, you can see what are they citing. So what is the actual citations for those? That's what I find to be more powerful measurement is useful, right? It's a good measurement to see like how are things working. Even beyond just profound and peak and prompt watch, I would say just Google search console data is still at the best. So what we saw with both clients that we were able to increase leads from AI is that there's two main signals that really were the clearest, which was branded search increase. So most of the time with chat, CBT or Google AI overviews, if you ask something like what is the best coffee or what's the best coffee brand, they'll literally just list out the brands. So most people when they see that they're going to copy the brand name, research on Google for Google AI overviews. If you click it, it'll even go to a Google search of the brand. And that's always going to lead to a sustained increase in branded search queries. So it's a weird measurement. But both times we seem like a 35% in one case, about a 40% in the other of sustained over three months, increase in branded search, confirming with the client that there was no huge promos or any kind of reasons outside of this. And it seems like just every day there's a sustained increase on our branded name. And so that's one of the best first party data is outside of just measurement tools. All right. I have attempted. I've attempted to guess and I've tried a whole bunch of different prompts. Or I'm just trying to guess it's something to get it to recommend me the branded question. I can't do it. But I also don't know what you're optimizing for. Yeah. I feel like I've just tried stuff like best antioxidant coffee, best antioxidant instant coffee. But then also obscuring the thing, you know, because I've got like all the previous context in the history. Yeah. Yeah. And I can share you that data. But yeah, I don't know how to share it. I get it. Yeah. Okay. Oh, yeah, no, we'll censor it. Let's do it. Sensor talk. Try it. Coffee, coffee, all that kind of stuff. Google AI overviews. It's almost like constant. Chat GBT is a worse. That's what we're working on now. It's about about 60%. So it's about 86% of the time on Google AI overviews, 60% on chat GPT, 70% on perplexity. - Okay, this is interesting. I got it. I had to, chat GPT, I couldn't get to do it, but I didn't try it in Cognito. Google didn't do it. When I opened an incognito window in Google, immediately your brand is second result. So it worked. But it is, yeah, it's like anything with SEO, the personalization gets in the way of testing. And that's where these tools are helpful. - Yeah, and it really is. And that's why probability and data is where, it can be a good directional signal. So like in Google search console, like our branded search for that brand has gone up 40, or 35% just on our branded term. Nothing, like over the last three months compared to the last previous three months and last three months compared year over year. - So there's SEO benefit to this approach as well. - Yeah, and like the main KPIs we're using on that project are branded search increase, visibility increase, but also new customers acquired from organic because new customers acquired on organic is probably the cleanest data point because a lot of, as you build up AI visibility, people are going to be searching the brand and they're gonna be acquired from essentially the organic SEO channel, even if it is AI. And then one of our other clients just pretty much just asked people constantly and he's like, and he's like, yeah, a lot of people are coming from AI. But both cases from like a data standpoint, we saw a large increase in branded. And okay, well, the other one I just tried perplexity, I don't know if you use perplexity, the AI search engine, I love perplexity, have never paid for it, but I use it all the time. It, that one, top picks overall, your brand, purity coffee, number one for your phrase. Yeah. (laughs) So yeah, I mean, it depends like, I mean, there's a lot of factors here, but yes, okay. I have confirmed this, I've now like perplexity is the one that did it consistently. But the, so what, what's the strategy here? You know, what is, what went into making this happen? Yeah, so the biggest thing that we've learned through both projects was what I would say is web consensus. So it's a form of link building, which is essentially just, they're using citations. You ask chat GBT, you almost always see like a searching function. They're going and gathering citations and just kind of looking at what the internet says. And so for one example, we did a large scale just affiliate program kind of stuff, where we go out to, you may have heard it before if you're researching Geo stuff. It's like, listicles is probably the biggest thing people have been talking about. Yeah. And so listicles and Reddit. And those are both. This the one we heard a lot of, especially like talking about it a year ago was like, all right, you just century Reddit comments, bam, was the solution, which didn't feel great. No, it didn't feel great. So we've seen out of the two projects, one is focused on more listicles, the other is focused a little bit more on Reddit, but being very careful. And we actually worked with a partner there on some of the Reddit stuff. But it didn't work, which is interesting. I'm mostly curious about sort of what's working. On the listicles, they work well, but both cases you had to be very careful, 'cause like you can just spam everything to death and maybe that work, but do you really want to do that for your brand? Probably not. And so the listicles, the way that we did it was with affiliate programs and reaching out in a natural way, personalized way. Looking at the citations and just saying, hey, if you're reviewing the best of XYZ product, just reaching out to them, giving them a competitive affiliate program and sending them a product and asking them for inclusion in that list. So that was a major factor. Those are listicles as well as publishing your own listicles. And that's where it gets a little bit-- That feels recursive, right? It gets a little bit gray hat kind of thing, but we didn't really do that much, especially for this project. The other project, we did more of that, which is just publishing articles saying, like, what are the best XYZ or what makes this good, all kinds of questions. That email comparisons between brands, so brand X. That's the one we saw in practice, the comparison ones. We have a brand that discovered in Google Search, a new product was being recommended over there by Google AI overview, on search terms relevant. And then like that, and they came to me with the solution. They said, hey, we're gonna write our own article similar to what they did and we're gonna add, we wanna add this thing called JSON structured data. They knew they needed it, but of course, it was technical, so that part I did essentially turned these articles into like FAQ JSON style snippets, and then just drop that into the theme as a custom liquid template or custom liquid section. It's of that, then the theory here is structured data, JSON is just easier for the AI to ingest, so it has a preference for it. - Yeah, JSON is great. Structured data is very good. And when it comes to how to do geo and how to increase visibility for your brand, it comes down to one, which is web consensus, which is as basic as what are blog articles saying about your brand, what is Reddit saying about your brand, what is even like review sites, like trust pilot, BBB, like especially if you're local, BBB is very important, better business bureau and stuff like that. So what are all those sources saying if you think of like a traditional search engine, if you searched your brand and if you searched FastXYZ and where do I buy this, what all these terms, like what shows up, the AIs are crawling all of that, they're extracting all that information, they're eating the internet, so you can kinda think about it. And so like what shows up? - Yeah, that's always been the joke, 'cause like, you know what, what's the response you get from CHIQPT? Well, the gross stuff it ate on the internet. (laughing) - Exactly, and that's part of the problem with AIs too, like, you're not always gonna get the best answer, right? Don't trust everything you read on the internet. But they do do a pretty good job at like getting some kind of consensus. So that's why it's very difficult to, like you don't really need to do, you shouldn't do GEO in hyper competitive spaces, like running shoes, 'cause there's just too much content on the internet about like Nike and Reebok and like, there's no way a new running shoe brand can ever compete with the amount of information that's out there on. - 'Cause you're also like, you're competing against all of the content that came before AI, ate the internet, right? And like, you're not gonna overcome that with a broad enough category, you know? - Yeah. - There's just too much content. - So you have to be, you have to be pinpointed. And if like, maybe you can try to, and this is where you had to take, you know, discernment very seriously, like, maybe you could try to, you know, get in on a very subset of running shoes, you know, like running shoes for people with, you know, like some kind of like, foot disability, right? But then it's kind of like, well, does that fit your brand and so on. So I think SEO has always been this way, which is that you have to be very conscious of who you're working with and what your brand is, and whether this stuff really applies to you. But I will push it back to just what you can do. So once you've determined that you want more visibility, which I do think is a smart idea if you're in a more of a niche where you can be a leader in that sub-niche, then it's really about web consensus. And then after that, it's about traditional SEO. So that's kind of where you were getting at. Like structured data, schema, comparison articles, like, - That's, that's technical SEO. That stuff we would have done and benefited from before AI. - 1000% like if you look at any of these SaaS companies, like they always have like the versus articles, they always have, and this was all done way before, you know, - Yeah, they were doing that, yeah, five, 10 years ago as well. - Yeah, and so like I do think it's funny when sometimes people come to me and they're thinking it's, you know, this brand new thing, but it's definitely an evolution. Like I would find that these, these listicles read it and stuff like that are working really well right now, but it's almost like their new SEO tactics. Like there are new things that work within AI. But yeah, if you don't have a website that's fast, like if your website is exceptionally slow, if it's super JavaScript heavy, like the AI crawlers just as Google will kind of just like, give up like a lot of times that there's a ton of JavaScript, they can't crawl it very well. - There's no change. - There's a lot of, - It's intensive to render and view. Like for an AI, viewing a website is actually quite a lot of work. And so if it's gonna run into a roadblock, unless it really needs it, it's just gonna move on to something else. Here's some customer service emails. You don't wanna wake up too. - I was drunk. - This was a gift, but we broke up. Wrong color, wrong size, wrong me. - Sound familiar? - With cleverific, you don't have to deal with these anymore. - Cleverific gives your customers a self-service portal to edit their own orders. No emails, no back and forth, just simple fixes, fewer support tickets, faster fulfillment, and happier customers. Peter Manning, New York, slash their order support request by 99%. You can too. Get 50% off the cleverific pro plan. Just $49 a month exclusively for the unofficial Shopify podcast listeners. Go to cleverific.com/unefficial and use promo code Kurt50. Fix the problem before it's a problem. That's cleverific. What was the timeline here? From when you said, all right, we wanna rank for whatever phrase to being able to consistently do it. - Yeah, so it was six months. Pretty clean time period. It's really looking over the last three months to the previous three months that we've been seeing a lot of the data and then also comparing it year over year. And both periods look really good. And yeah, so it took about six months of aggravating the data, getting on these listicles and read it and things like that, as well as doing, I mean, like basic text was yo. We've been doing that for eight years. So we improved the schema, we improved the content, we improved the, even the collection page content, all of that kind of stuff. We added FAQs, tons of just typical technical SEO. I mean, all the heading tags were everywhere, like just getting a very clean structure alongside getting more essentially backlinks, which would be the SEO term. But I think that's all very traditional SEO stuff, which we know is like, you'd have engaged in regular SEO, it does benefit AI search as well. So there's this article that you went around a couple weeks ago, BBC Future in the titles, I hacked chat, GPT and Google's AI, and it only took 20 minutes. And the thing the guy does is write an article on his website, just one article publishes it. And supposedly, then suddenly Gemini AI goes, "Oh, this guy is the world hot dog getting championed." Right? Like, have you read this article? - No, I haven't. It sounds pretty funny. - Okay. Yeah, where he's talking about, hey, you can, you know, here's this trick to manipulate what AI is doing. - That's true. Like, I've seen a lot of people do similar stuff like that, but only for very long term queries. - Then, okay, that is, yes. Like he's, he's spent 20 minutes writing an article, and he links to it, it's hotdogs.html. And the title is the best tech journalist at eating hot dogs. So that's like, no one is searching for that phrase until he came up with it right then, that's a brand new sentence, right? You know, the best tech journalist at eating hot dogs. - That's what he was ever thought about that. - Yeah, but to your point, it's like, okay, you're not gonna compete for best running shoe, but best hot dog eating tech journalist, sure. - Right, right. And I mean, I think there are strategies out there that work pretty well that are based on long tail. So, I mean, like you could do, you know, best running shoe for and then x-wise foot disease, right? Like, I'm sure there's probably like a hundred different types of foot, like disorders and things. And then you could create articles on all of those, publish them all to the internet, publish them to Medium, LinkedIn, all these places. You could publish, read it, post on it. So that kind of stuff can work. And maybe it would work well at scale. I haven't tested it, I don't know. We did a little bit with like longer tail stuff like that. But yeah, I mean, it's very long tail. It's super easy if you're gonna say, you know, like, bass SEO that's a ginger named Patrick Rice. - Like, your name starts with P. - I'm for sure you could rank for it. But then like again, you know, to your point, later in the article, they quote, "NSEO consultant." And he says, you know, anybody could do it. You just put your own brand in number one, in a listicle and other brands two through six, and your page is gonna get cited. And that like that was your strategy. So there's definitely, there's other people doing this. Like for sure this works right now and it's out there. But is this, you know, it years ago, blog comment spam was everywhere. Now I think the listicle article spam is the thing we're gonna see a ton of. - So will it be long term is a really good question? I think it depends on how you're doing it. Because even in this project, our approach was not just to post on any old website. Like typically in SEO, you have a ton of guest post farms is what they are called. Right? You just like any website that's like, you know, they'll, they probably already own a hundred websites and they're like, yeah, pay me 50 bucks and I'll post something. - Oh no. - So that's kind of like standard guest post farms. And some of those can work to be honest. Like if they're a good website, they have traffic. So like that, they can push ranking on Google. But a lot of them are really bad. So you have to be very careful. Rather than doing that kind of thing, which still can work in AIA has less filters. We focused on what is currently being cited, what's currently ranking well on Google and what's like niche relevant. So we were mostly looking at people that actually had blogs and good posts that were relevant, written pretty well, not just AI content. Most of the time we were sending them products and I could kind of tell that they have actually written the content. It wasn't just like AI spam that they didn't think about at all. - Yeah, turn out that's the new Turing test is like, oh, it's still using M dashes, huh? - Yeah, yeah. Like you didn't even add any images to this post. You didn't literally do anything. And so the hope there is that we're going about it a little bit of a better way. Right now I do think it's a Wild West where a lot of this stuff just works. Like Reddit spam works very well. And I don't do it like the only way that we're comfortable doing Reddit stuff is if it's totally like gone. It's not just like a wear posting our own post and it's saying we're the best. I'm very careful with that. Even on the list goals, I like to add like this is a sponsored post or this is like, this brand is us, but then here are the other brands. So we try to do that. - Yeah, that transparency should be there. - Yeah. And I don't think the transparency will necessarily hurt the success of the AI result. If anything, it may accept it as more truthful. - Yeah, it's not right now. And I think if you do it in a conscious way, same with like Reddit comments, you can be like, well, what is the best? I mean, I guess we could use running shield. What is the best running shield? You'd be like, hey, like I own this brand, but some other brands that I really enjoy are XYZ or you could sometimes even like you can put it in your name or something like that where you can be like Patrick owner of this brand or some, you know, in your description and stuff like that. But yeah, I would say be careful. A lot of this stuff is working really well though. Like if you just have a bunch of Reddit bots and you post a Reddit post and you upvote it and then, you know, you say you're the best, it can work well. - That's serious. - You should use discernment. 'Cause that's, this is also a lot of traditional black hat stuff, which I've always stayed away from. And it's just the wild west where it does work right now. It may not work forever. Like I would imagine that they're going to have already people have theories, no clue if this is true yet. But they have theories that AI is starting to actually talk to the owner of ClearScope, which is an SEO tool. And he was saying that he's seeing in his data that they're starting to deprioritize self-promotional listicles, which is like if it's on your own website. And you're, so if you write a BAS XYZ on your own website and then you recommend yourself first, like it seems like they're deprioritizing that, no clue if that's true. But I would imagine some sort of filters like this and some sort of iteration on the way it works will have. - Well, yeah, for sure. I think just as Google was able to optimize their algorithm to make black hats, SEO tactics, spam, harder and harder, and if you only search results, they're gonna get smarter about filtering this AI search manipulation as well. But if you could do something and it passes my three checks, my, I call them the Melchak's moral ethical legal, hey, make hey, well, the sun shines. - Yeah. - But yeah, that's how I see it as well. I think as long as you're transparent, and I mean, people have been, a lot of us like PR, like people have been sending product for review, which is gonna be one of the best things you can do, like as a Shopify brand, if you're D to see or something, If you can just send out product reviews to blog posts, bloggers, YouTubers, you know, all this stuff. YouTube's recommended a lot on AI as well. So we're seeing YouTube and Reddit to be the two most popular on AI citations. So I mean, product reviews and all these things have been going on forever. So I think they're useful tactics if you have a cohesive strategy. For sure. So the, you know, if I'm listening to this right now, I got a new store so it doesn't have backlinks. It doesn't have the, you know, AI citations. What are the, what are the first three things I should do? What are the top tactics I could take today to try and prepare? Yeah. So I would say first off, just make sure that you have your fundamental SEO intact. So I would focus on originally just like, do you have keywords that you're targeting? Do you know like what you're targeting? And then are those realistic? So like we're talking about not best running shoes, but like best running shoes for X-rays. Because even though AI doesn't use traditional keywords, they keywords typically reflect some kind of desire or interest within humans, within the people behind the screen. And if there's a desire and interest in a product or in a thing, it didn't really matter how they prompt it or how they ask the question. Like that's what you want. So you need your website to be focused on that, to be fleshed out in it, and to kind of have like a level of identity. So I feel like a lot of people just don't know. They kind of have had it wrong a little bit in the start without knowing like what are they actually targeting with their SEO if they're doing it. So if they're brand new, I would just say stick to the basics like SEO title tags. Like once you find some good set of keywords like SEO title tags, you know, making sure the website functions well has good heading tags. If you're not using it, JSON LD for SEO is fantastic. Shout out. Absolutely. Honestly, great tool. And so I get your basics in place. After that, I would say do a press release. Press releases a good way to get out a bunch of content on your stuff. So that's working with AI as well. So some people can get longer tail queries ranked in AI just from like shooting out press releases like PR web, A and B newswire. There's a few good ones that are pretty cheap magic PR dot com is good. So get out some of that stuff. And then if you're able to, we have a pretty good vendor for some Wikipedia stuff. Wikipedia is very powerful. Yeah. Yeah. And so some of those sort of just external sources that you can get for relatively affordable prices. That's kind of building up your foundation. And then if you are a bigger brand that maybe has some of that foundation there or once you've built it up, then I would start going into like, how can we increase the number of blog articles mentioning our brand? That'd be kind of like link building. How can we make sure that we're on really firm technical foundations, which you would probably have to get some sort of SEO consultant or team in on to make sure that things are really nice on or at least a dev team that knows SEO. And then yeah, start looking at Reddit and some of the social stuff too if you have the bandwidth. And okay, maybe well, are you taking on Shopify clients right now? Yes, we are. All right, that's good. Where do we go? Where do we find you? Paturosericeco.com. So paturgericeco.com. I do have, if you go to paturizeco.com/ai, I am doing AI visibility audits. So that's where we'll check out some stuff on profound. They have their workspaces, which is nice. We'll check out the Google search console data and just get an idea of where the visibility currently is at. And yeah, so we're doing that right now and we are taking on SEO, Google ads and Geo clients. So it's cool. Okay. It's fun for me. Yeah, I bet the, you know, one of the things on my to do list that I put down yesterday and I've not gotten to it yet. We just moved our agency site, ethersuckle.com to Shopify. And that really makes my life easier and what I could do with it. And I've got all these articles and resources and case studies and stuff. I'm adding JSON structured data to it. And I'm just going to, yes, we've got it written in as a meta field. You know, so that's I'm going to do it manually. Like I'm a really hand-tuned in my approach here. Yeah, that's the way to do it. I love it. Yeah, we use a lot of meta fields. I mean, meta fields is just for SEO and all this stuff. It's incredible. And with JSON LD for SEO, you can do so. It's a lot of Davises app, right? Yeah. So that's the show before. Yeah, yeah. And with that tool, you can do custom JSON. It's a little bit complicated. But if you use, you can get it to work. So she has a blog post on it. So you can do some custom JSON. But yeah, JSON is a great, like for people that, especially if you're on a budget and you want to do some AI stuff and SEO stuff in general, they're big wins and it's schema for sure. Oh, yeah. That's another thing we're doing for our retainer clients. It's just looking for those opportunities where it's like we could, like we already have the structure data built into the theme, but it's got to be all dynamic. And so what if, you know, then I pair that to meta fields on, you know, like I hear and I don't have to do it for the entire site. But like, hey, I have a hero product and it's got an FAQ. Okay, well, why don't I make a separate JSON section that gets that FAQ in there as structure data so that when someone asks Google's a question about this product, AI goes, oh, you know, I've got a primary source in a way that I easy understand and it becomes that much more likely to surface that response. Yeah. Yeah, most definitely. It's fun. Plus, you know what? It's more, it's more spreadsheet work for me. It's some more matrixify action, right? It's the tools I love, love to use. So, okay, I'm here to give you more spreadsheet work. Yeah, for, you know, for me, that's fun. It's just more toys. Okay, I've got some work to do. I have spreadsheets to mess with. Jason to write and I'm excited. The SEO was fun for me 15 years ago when it was, you know, a little more wild. And then over time, you know, there it was just like, hey, make great content and, you know, keep your head down and hope it works out. Now we're back to, well, there's some like pretty off the wall things you could try. And then maybe you get out, you know, wildly rewarded by doing it. It's not going to try it. I'm going to put in the effort, put in the structure data, make the content and see if we can increase that Google organic search results again. You've inspired me. Thank you, Patrick. Yeah, thank you, Kerr. I appreciate it. Love to be on again. So, I appreciate your time. Hey, before you go, I was hoping you would check out our new app promo party pro. It is what I want to be the single best, easiest way to run a free gift with purchase promo on Shopify. We just put it live in the app store. We've got less than 50 users. We want your feedback. So if you need to run a free gift with purchase promo in the near future, install it, try it. There's a live chat. I check that all the time. And so if you have any issues at all, or any suggestions on how we can make it even easier to use, let us know. We're happy to help. If you want to try it, search promo party in the app store. promo party pros the app. Give it a shot. It's got a free trial. Thanks.