Your customers will be loyal to AI, not your brand

Thanks for having me. I'm so excited.

Yeah, so I am a software engineer that got went from finance to marketing, where I met my co-founder 12 years ago. I left the marketing agency to actually go work as a trader for a little bit, did that for a minute, and then went back to consulting and then decided it was time to start a business when I missed having I missed having a team. I saw a need for it, and then founded FaZe Wheel. We're about 11 months in now. So Yeah, from an econometrics degree to a finance background, straight into marketing, straight into software engineering, then back to business. It's kind of the gist.

Yeah. So I actually started doing GEO as an independent consultant last April or two Aprils ago. So April 2025, which is very, very early. and that was because I was looking at server logs for a client and I saw a bunch of like crawl activity from Chat GPT. And I said, What is this? What's going on? And then I slept on it for a while and I thought this is gonna be, this is gonna happen. People are gonna start using Chat GBT. At that point, it was pretty much just Chat GBT people were using. They're gonna start using these things to shortcut their shopping experiences and shortcut their research. And I should get on this. So I started learning everything I could about GEO, AI SEO, whatever you want to call it, AEO, I whatever. started learning everything I could about it, and then started doing it for clients. I saw it return incredibly quickly. I saw it working really well, and then tools kept being adopted and adoption spread of those tools. So people were using. Claude came out. that was the biggest one. People are starting to use AI more and more throughout their daily experiences. Siri, even Google AI mode. And I realized this was a big thing. I then went and found Eric, who I worked with 12 years ago, and realized I knew Eric was the right person to do this with me because he has a background and everything I don't. He has a sales background, he has a company building background, he's very well connected, very well spoken, very patient, all the things that I am not. So I grabbed Eric and then I we kind of built the team around the culture that we wanted for phase wheel, which is a culture of people that are curious, people that are aggressive, aren't passive and we want the best for our clients because GEO is a f is a flywheel, you have to we get the win again and again and again. And so we kind of built the team from there. But it really just started to answer your actual question, really just started with me looking at server logs. And this really didn't have a name yet. it really wasn't a thing yet. And it was just being curious and digging into client strategy and understanding this was gonna be a a big deal and a big part of my job and my mindset is future proofing. So I think what's happening in six months and how can we get our clients ready for that now? And that was how it started.

Yes, I've gotten into a lot of LinkedIn fights about this. I've tried to be better about it lately, but I do get in arguments sometimes, or people say that GEO, A SEO, what I'm gonna call it, is just SEO in a better outfit. And that's just not true. I do think SEO is very important. I believe SEO is still important. It still should happen. It's a fantastic foundation for GEO. They're not the same thing. SEO gets you found, GEO gets you ranked, or gets you picked. So SEO ranked, GEO picked. It's SEO, you're one link on a page. GEO, it's being hand delivered to the person. So SEO is you're finding the internet. The internet is there. You're going out searching for it. GEO is the internet is being delivered to you from an LM, from a model. And so the difference there is just gigantic. SEO is written, SEO is generally not optimized towards integrity, not optimized towards not integrity. It's sorry, sorry, anyone that's SEO, I believe you're Ethical. it's just SEO is optimized towards almost not clarity, where geo models prefer clarity and a clear idea of credibility. So it's just like a different method where one, yeah, one is one is built, geo is built, SEO is paid for. I think is like kind of the difference. I hope that answered your question. I'm not actually sure. That felt like a bit of a ramble.

As like an provider or as a client, or both. As a client. So if I was starting a new company and I wanted a GEO, they value credibility and they value clarity. So I would first make sure that all the information I'm putting out there is accurate and consistent. So that means who you are as a company, what you sell, how you sell it, what it's priced at, who works for you. All those things should be completely consistent. So I shouldn't, you shouldn't have let's say you're A B testing pricing. Love that.

Great when you choose the price you're gonna go with, remove all the evidence and redirect it of the old pricing. So conflicting messages are bad. Like make sure it's very, very clear. I would both have a version of a website that's made for robots and one made for humans because those look very different. Humans want descriptions, robots want answers, question and answers. I would also get the founder ego out of it. Like founders, myself included, think that people care about. ethics of the comp they care about not the ethics like the ethos of the company. They think that found the people care about the story behind the company and they might, but GEO does not. And so it should be instead of like for example, instead of just if I was a jewelry brand, instead of saying, My great Aunt Elizabeth like wore this jewelry and I was inspired by it, da-da-da-da-da, like that's great. And humans want that. GEOs just want to know what's it made of, what's the stone, what how much is it selling for? And so there's a difference there. I just think that founders get romantic about it and I would make sure that's great, but make sure you have a version that's unromantic for GEO to capitalize on. Because the way the LLMs make money is they they people use them. They get your data. And so they want the answer as easily as possible, as quickly as possible. And if you're hiding that answer and making it hard to find, then the lazy robots aren't gonna find it. So robots are lazy, make the information as easy to find as possible and as clear as possible is like the first thing that I would do. And also like look your brand up. And look it up in incognito, ask your friends to do it and like see how your competitor's showing up and then verse engineer how they're showing up and try to attack that competitor. People miss that piece in GEO where it's both it's both for your own brand, but there's also a very competitive element for it where you can actually attack your competitor's market share through GEO. And that's missed like massively.

I would say we we have it before, but the the old content is just like clean it up. You're gonna have old blog posts with old pricing, you're gonna have old competitor knowledge and things like that, and like clean all of that up and then make sure that you have your schemas in place, your robot.txt file in place, all those basic things that LLMs are gonna look for as they're categorizing your site and then go after the competitive analysis. So creating blog content, looking at market share. But the biggest thing I see is just old content that they think doesn't matter because no one's looking at it, but LMs are still finding it. That and not Geo is it starts with their website, it starts with their own client branding website and then also goes into third party sites, PR, Reddit, YouTube, LinkedIn, like all of those things should be clear and very, very credible. So it should be very accurate across the board. It should say the same message across the board and it should be coherent. Or cohesive. Sorry.

Yeah, so if you're a founder or you work for a company, you've gotten those emails in your inbox, most likely, that are like, we can get you ranked on every LLM within a week. That's a giant red flag. Promises don't count. Like GEO is built. It's not bought. Paid search is renting attention. GEO is building attention and keeping it. So paid search, you rent their attention for the time at which they do the search query. GEO, you're actually building a credible brand that LM's gonna come back to you again and again and again. It's actually a flywheel that builds on itself. So if anyone's promising an instant result in the way that paid search would promise, that's a giant red flag. The difference there is that companies are unethical. And like I I think these companies think they're doing what's best for their clients. I don't think they're like garbage people. I just think that no, I do think they're garbage people. I think it's awful, actually. Like I don't think people like. So the difference between an ethical company and not ethical company is ethical companies focus on strategy, ethical companies focus on automation. Are they just spewing out blog posts? Are they just automating your schemas? Are they paying any attention to what's going on? And are they adjusting? So, for example, like Muse and Instinct were giant this month. everyone's using Instinct, everyone's using Muse. They're very different products. but they both have a very low barrier to entry, where before there was a higher one. With like so the example is that a claw power user and a muse user are gonna use the models very, very differently, which means we need to account for that. Which means as a company, we go back and we change things we did before to account for that. If your GEO partner is not going back and adjusting things as new models come out, that's not good. If they're just spewing out blog post after blog post, also not good. And if they're not asking for input, like we have we have very open lines communications with clients saying what's working for you right now, what's not, what's the new strategy, how are sales going? We always have a clear idea of the company strategy. So we can feed it in. So if anyone's promising low price point, making promises that are absolute, automating things instead of having discussions with you, they're not a partner, they're a tool. And I like tools. Tools are great. Problem with tools is it everybody has the same tool, they're getting the same recommendation. If your par if your competitor has the same tool as you or the same like unethical company as you, they're getting the exact same recommendations. And that's not great. You're you're not gonna outs like outpace your competitor if you're getting the same amount of knowledge. So that's why we hire we hire partners, we look at strategy, and then also at phase wheel, we only take on one client and one vertical and one region. Because you go after competitors, we can't ethically have two competitors and work with each of them.

Yeah, it's not good. Yeah, so we wanted to make it really clear that GEO is not like a one-time thing. It's a process and we go back and we kind of re-audit and relook at what's going on frequently and constantly because it's a moving, it's a moving industry. And it's a moving industry both because of the new models being released and both because we're getting feedback with every single piece of content we put out there, every single optimization that we make. So we're constantly going back and checking things are working. The Illuminate, we do an audit. So when we work together with clients, we start with an audit always. We get a baseline, we understand what's going on, and then And from there we optimize the website. And then from there we measure and then we go back and do it all again. So the measurement protocol that we do, we have three main pillars right now that could change at any point. but it's clicks. So actually analytic tool, how many clicks come through from G or from an LLM that are being tracked? It's crawl insights. So we have a tool that captures server logs and sees what what bots are crawling, when are they crawling, and how are they crawling? Is it a crawl for training? Is it a crawl for retrieval? Is it was it a call to like there's an actual query? Like what kind of crawl is happening there? And then the third thing that we do is the share of voice. So we use a third-party tool that runs those queries again and again and again. And it says, what percentage of time are you showing up versus your client? And then we kind of average those three together. it's not perfect, it's just what we have. So there's that, those are the three ways that we kind of put them all together. Because if you take it like a lot of companies take a screenshot from Chat DBT and say, like, this is it, that's not helpful. That's a random antidote. We don't we don't want that. We want like an actual signal that's going across several models for weeks and then we want to compare that. So that's how we do it. It works for us right now. Like if there's a better way later, it will change. but yeah.

Mm-hmm. This is a great question. I love talking about this. I also am willing to acknowledge that I could be very off-base, but I mean, last April I predicted this coming. Last November I filed for a patent for a tool we actually use at Phase Wheel, which is a machine layer infrastructure that translates your website into robot language. So it strips away everything that LMs don't want to see and feeds it to them in a structure they do want to see it. So you're much more likely to get cited. So that's the one thing that we use at phase wheel. We're hoping to eventually sell that as a standalone product, but that's a big tool that we use. As far as future proofing, I do think within like the next year or so, we're gonna have a world in which you get a new laptop. I mean, if you're an early adopter, you get a new laptop and that laptop builds itself for you. And so it builds itself depending on what you want. So I think that right now we work for software. Software is built, we use it, how it's built, whatever. I think in within a year, it'll be the world where we soft it's reversed. So software builds itself for us. And I think that's gonna change dramatically a couple different things. One is I'm so sorry, branding won't matter as much. And like I love branding. I love pretty things. I love website, like I love those things. But I think the thing about that is is right now, like for example, I was gonna go buy if I need like if I need a hammer for my house, great. I live in the middle of nowhere, there's no store, I'm gonna go on Amazon and buy a hammer. Instead, now I'm gonna go to Chat TBT or Claude or whatever and say, Hey, I need to buy a hammer for this. And it's gonna say, okay, here are your options. And it could be a website I've never heard of before and I'd still buy the hammer from them. Where before I automatically go to Amazon and buy the hammer. So the brand loyalty piece is gonna matter a lot less because now my loyalty is to my LM, not to a brand, not to an Amazon, not to whatever. and so it kind of short like the shortcut of going from Amazon. is gonna be a lot less helpful. So that's one thing to be aware of. I think that's gonna be important that your product is better than your brand. Or right now I think branding does a lot of work or the ease of going to the same website. the second thing I think is gonna happen is we're gonna have a network and it's happening now where just robot to robot, you're gonna speak, I'm not gonna be involved. So to set up this podcast, we went through a few C emails, we did a scheduling link, all those things I think within a year, your robot will talk to my robot, they will shake hands, do a little dance, whatever. And it will schedule the meeting for you and then it'll let me know. So right now I'm telling agents what to do. And I it could be because I'm not good enough with agents, but I do think within a year, people are gonna be very comfortable having agents act and then telling them what they did. And that's a massive difference. And that matters too with selling and with brand structure because that credibility and that clarity are gonna matter so much more than they do now because people aren't gonna automatically shortcut to Walmart, Target, Amazon. They're gonna go to whatever the LM tells them to do. so that's what I think is gonna happen. The way to prevent that or way to like future proof yourself for that is GEO. It's GEO and making sure your website has an MCP. So that's called it's a model context protocol. Shopify has it built in. WordPress, Webflow do not yet have it built in, but you can buy tools for it. And that's just like a layer. the the tool that we're building is called Mirror. Mirror also does it in a different way that's patented. but it's just a way of making sure your website is translated for those robots because when you start going agent to agent, handshake to handshake, API to API, whatever it is, your website front end's not

Yeah. Do you find that you like trust so I I trust my models too much. Like if they gave me recommendation for something, I just buy it without question. Do you find you have that same trust? Or what does that look like for you?

Yeah. So we've had a really interesting change. But when I first started doing this, it was brands that were entrepreneurial, they're well established, they were high growth phase, and they've kind of tried everything else first. The brands that we're working with now are this is new to them. This is new to them. They're not super experimental. They are a very high growth mindset. They are very solution-oriented. They're they generally have a product where it's a high consideration product, it's a high price point product, and the GEO effort is tied directly into revenue. So the brands I'm working with, we're working with now at Phase Wheel are client are brands that really have no baseline. like one of the brands that we come, we had a brand that did this the most interesting one we have, I think. We do a of HVAC and plumbing. That's a big one. We do a lot of automotive, those are big. but my favorite one so far from a vertical standpoint, not from a client standpoint, don't worry, I love all of you equally. is they they sold custom outdoor kitchens. And so it's a high consideration product, it's high research, and it's a high price point because it's a custom thing for you outdoor kitchen, whatever. And it's it was all over the country. And they had absolutely no visibility whatsoever. we got them up to the point where they were quite literally doing millions of dollars in revenue a month from GEO. And we could see it directly because it's click-through rate. And that was a click-through to their, they had like a video call where you could like walk through it. and that was my favorite result because they went from zero. And that was if you my thing is if you don't see some kind of movement within 10 days, you're doing it wrong. Like 10 days is a good enough time. I'm not expecting full results for like three months, but 10 days you should see movement in the right direction and you should be showing up more for share of voice. the issue there is our job is to get the client to your website. What happens after that is your business. If your website has a it's got horrible CRO, you've got bad conversions. That's not great. And that's something we can help you with and we'll look at. But our job is really to get you the website. And the cool part about this client with the outdoor kitchens was the leads were it was a very easy way to see, are we sending qualified leads? And their conversion rate was incredibly high. And so that was that was really, really cool to see because it was a one-to-one conversion. They weren't like car dealership where the lead it might take months for someone to buy a car. It would be generally someone to find them and they would book the call and they would schedule the installation within a month. So that was my favorite client because it was a really cool way to see the entire funnel. but I will say, yeah, if you're doing GEO and you're not seeing an increase in share of voice within 10 days, change something. And and you are like actually doing it. You're you're putting the blogs out, you're putting your quotes out, you're doing your thing, and you're not just like passively sitting by. you should see a substantial improvement in your share of voice within 10 days.

Yeah, so we do that part as far as making sure that agents can navigate your website and make it very easy for agents to navigate the website. The way that I put it is like agents are walking into the supermarket and if you're not on the shelf, then they cannot buy you. And if they're not, if it's not easy for them and if it's not easy for a bot to walk in the front door, pick your product off the shelf and check out with that product, they're not gonna do it. Bots are so, so lazy. LMs want to conserve resources like everyone else, make it as easy as possible for that bot to buy your product. And if you're not doing GEO, you're not on the shelf at all. So we go in and we do make sure that like agents can check out on your site. If it's humans, that's on you. Sorry, you're on your own. But for agents, we do optimize websites to make sure they can go and actually do what they need to do and find what they need find. And we do that. The first step I would do if I'm like, how do I do this is schemas. So schemas and sit maps and raw txty files, when you walk into supermarket, you see those things on the top of the aisle that say what's in each aisle. And that's what your robots TXT file intersight map does. So make sure both of those are in good shape and robots know where to go to find what they're looking for. is like the biggest thing that I I would say that people don't don't think about and it's not high priority. But if it if it's hard for them to get or like pop ups and things, like they don't want that. That's why the mirror software works so well, because it takes all those things away. and if you have like one of those spinning wheels that does like a discount, like forget that's not no one wants that. LMs don't want that. Humans might. and so it's just making it as easy as possible for them to do what they need to do because they're so lazy, is what I would like focus on.

Yeah. we have we're all on LinkedIn. the best way to do it though is probably phasewheel.com. We have a contact us link in there. Or you can email hello at phasewheel.com. Someone will get back to you, probably me. you can also always find us on LinkedIn and yeah, yeah, that's probably the easiest way is our website.

Thank you so much. It was so much fun.