How to do AEO - Answer Engine Optimization

Want to get people to come to your website, download your app, buy from your business? Well, here's a guy who runs an SEO company who tells you, "Don't start with SEO." Instead, start with AEO. Answer engine optimization. In fact, I spent about an hour with him where he's going to walk you through how you can get to be the first result when people ask about your business to chat GPT, perplexity, clod, what else is out there? Let's get into it. I feel like a lot of people think that this is just too premature to start thinking about how to get into AI answers. Give me an example of what you've been able to achieve for a real customer, one that we know. So the way that I think about timing is it's a good time when the channel is large enough to be impactful. So what is the quantity of conversions, the quantity of traffic that you can get? And since January, the channel has been growing dramatically. If you asked me a year ago, we started playing around with answer engine optimization in June of last year. And people were intellectually curious about it, but they weren't prioritizing or spending money on it because it wasn't profitable channel because it wasn't big enough. And so it's now, for sure, the size of the conversions is substantial. And therefore, I think the timing is good. So Webflow is one example that we worked with. And there are not that many examples because we're still pretty early. So Webflow is an example, 8% of their signups come from LMS. Got it. All right. And this is because of you. This is a team effort. And we were part of that team, I think. So it was because of us. Let's break that down. There's stuff that you do for SEO, and then there's stuff that you do for AEO that you weren't already doing for SEO. And those are two different things. And so a lot of what you do for SEO actually helps with AEO without trying to. So the stuff that works really well for SEO that we work with the Webflow team on is a bunch of use case landing pages and templates, like how to build a website, how to build a website for designers or restaurant websites. So templates, use cases, how to do X type pages. And this is all the same stuff that you would do if you wanted search engine optimization to be increased, but also it might do for your customers, right? You're trying to show them how Webflow is used by restaurants, etc. Okay. And so you're saying some of that just naturally will feed into traffic from chatbots. So I would say most of the signups from LMS were from SEO work that we had already done what we're doing. In addition to that, we did things specifically for a French engine optimization that we would not have otherwise done for SEO. And so those are mostly offsite optimizations. So if you search for best website designer and Webflow's URL ranks number one on Google, that's a big win. And if in chat, you say, what's the best website builder and Webflow's URL is the number one citation, and nobody else mentions Webflow, then they won't win. And so for these popular questions, the way to win is not just a page on your site, it's pages off of your site. And so offsite is that that's one of the there's two differences with answer to optimization and SEO is offsite and the long tail. And so for offsite, we the Webflow team and we worked to get them mentioned on other platforms. So there's earned and owned offsite. So owned offsite would be Webflow has a YouTube channel, and there's tons of videos about how to use Webflow. And then Webflow can also engage in Reddit threads. So there's a lot of the threads either asking about how to use Webflow or asking about how to do things that Webflow can do. And so we collaborated with the Webflow team, the Webflow team themselves. So the Reddit's an interesting thing. When you ask growth people, really any question about how do I optimize a channel, they immediately anchor to scaling things and to doing using products in ways that they're not supposed to use them. So as soon as you say, how do we optimize Reddit, the growth person says, Well, I need to do hundreds of threads, hundreds of comments and not use it as a real user, because I'm using it as a real user, not a growth person, that's just, you know, I'm just a real person. So I need, I need to do something that I'm not supposed to do. And so the immediate strategy would be, well, let's create a bunch of Reddit spam with hundreds of fake accounts where we're all liking each other and we're creating karma and we're uploading each other's comments. We say, Hey, you should really consider using Webflow. So instead, we helped identify citations on Reddit that were part of the questions that Webflow wanted to win. And then people at Webflow then said, Hey, I'm Vivian. I work at Webflow. I'm a real person. Here's, here's some useful information that answers your question. So that's what actually worked. And then similar on YouTube, how to use Webflow. So those are things that, you know, YouTube, they were already doing that, but you know, we could sort of configure the YouTube strategy to be a bit more specific for AEO. The Reddit stuff probably we would never have done for SEO. But then we did a bunch of stuff for SEO that just helped for free in answer to optimization. But that's generally what worked. So if I'm understanding, you're right, Ethan, you're saying you go into perplexity, chat, GPT, clod, et cetera, you do a search or you ask a question that your users might ask, you then find a Reddit, Reddit link as the source, you click over and you say, I'm going to go and answer too. Or one of the people at Webflow will go in and answer. And what they'll do is say, I work at Webflow. So they're fully open about it and give a detailed answer. And that's one thing that you, that you all did as a consistent rhythm to your day. And then the second thing, if I'm understanding, you're right, is just posting YouTube videos, which I wouldn't have thought would help. Does YouTube videos help with the Gemini? Does it help with clod? Does it help with chat, GPT? What does it work? It generally, so LMS, the classes of domain, types of domains that are cited frequently, video and UGC are huge. Instagram, TikTok, LinkedIn, I am not seeing X that much. And then video and YouTube shows up a lot. And the less entertaining your product is, the better, the more interesting video is. And the reason why I say that is because video is made to be entertaining and entertaining categories like food, travel, beauty are the most crowded. And the least entertaining things are the least crowded. So website builders is moderately entertaining. We work with a lot of SaaS companies like payroll management software is not entertaining at all. So these are not crowded. And so the less, less entertaining and more niche your, but high LTV, your product is, video is the most interesting. So, and it shows up a lot. And there are all these AI adjacent channels. So there's answer and change optimization. And then there's stuff that's next to answer and change optimization, which is Reddit, affiliates, video, LinkedIn, Instagram, and a lot of these have not been optimized. And actually each has its own unique special strategy that's completely unrelated to answer and change optimization, but it's adjacent to it. So actually to do answer and change optimization really well, I think it's not doing pure answer and change optimization. It's doing all of these other channels, but doing YouTube and Reddit with the intention of helping you on answer and change optimization. But as a result, you show up in LMS, you show up within YouTube, you show up within Google search because there's a YouTube video about payroll management software showing up in Google search. You get the video on the landing page and then you get free views there. And so it creates this, this loop. How important is it that the LinkedIn posts, the Instagram posts, et cetera, has a lot of views already? It seems like there's a correlation, but it's not just ranked by number of views. So if you ask a question about payroll management software website builders, the videos that are appearing are typically the most viewed, but you won't see it where it's just literally sorted by number of views. There's other signals. I don't know exactly what they are. It's probably traditional search authority, like number of backlinks and shares and things like that, but it definitely does matter. What I will say is that for many questions, the number of views is not millions or hundreds of thousands. It's hundreds or a few thousands. So it's still pretty, it's still something that you could win without being an influencer. You also told me before we got started that for Webflow, help center content was useful. And that just means that if they're using, I don't know what they're using intercom help scout, what are they using? Not actually sure what they're using, but yeah, help. So the two things that are different with answer and generalization and SEO are the head and the tail and the tail is larger and longer in chat. The average, I think the stat roughly was 20 average or 25 words per prompt and average of three to six for search something along those lines. So chat is significantly more specific. And part of that's because search is made. People know that they need to translate their intention into a short keyword because they know that the search engine won't understand their actual full intention. So they don't act, they don't speak in conversational ways and they don't be, they don't get so specific that they know that the search engine will fail. Whereas chat can, you can be very specific. So you could ask a question like what's the best website builder that has this integration and integrates with Shopify and also has this feature and is, you know, has support for Korean language. You would never ask, you never put that in search, but you know, but that's what you want. And you can put that in chat and the chat can interpret that and give you an answer and help center. And people have many questions about, I want a product that fulfills my use case and help center is actually a great place to do that because help center is you talking about how your product works and what it does, what it does not do. And so help center optimization is not something you would focus on for SEO, but you would for answer engine optimization because that's where you have all this information about what your product does. And so do I now need to create all these different answers? Isn't the AI smart enough to grab three different answers, combine them and say, here's, here's what my user needs? Yes, you do not need to create a page, millions of pages for every single question. You need to generally answer the question. So an example that would be which, which for rippling, I want to payroll management software that has these different features like attendance tracking and time tracking and offboarding and this. So rippling has a page for each persona, HR, IT, finance, and all of the use cases and features that they have, but there's one page for each. So it's not that you need one page for every single use case and every single question. It's grouping and theming them. And the AI is indeed smart enough to just find the answer. So you don't need it explicitly verbatim, exact match for every single question. So going back to that hypothetical example you gave earlier, a site builder that integrates with Shopify has a set of features that we didn't get into and also supports Korean language. One page could say here all the languages that we support. Another page can say here's a list of all the features. And then another could say this is a list of integrations and including Shopify, maybe even have a Shopify only page. And the AI is smart enough to grab all three connected together and give it to the user. Yes. Okay. All right. So help center. What about how two articles on the site? That's helpful. It is help center enough. The more the better. An interesting thing when you search for when you search for a rippling and pave and or sorry, when you search for which payroll management software integrates with pave, there's multiple articles that show up saying rippling. There's a rippling pave integration page. I believe there's a pave rippling integration page on pave.com. And I believe there's a help center article. Another example is a meeting note transcription that works with zoom and order shows up with a feature of zoom feature page, a help center article saying how to do this. And I believe there's a page on zoom, but they're basically there's multiple pages and all of these are getting order to show up. And so it's as I wouldn't say that, you know, the more the better forever, but I would do multiple versions help center article, how to do X article, ping the company that you're integrating with and ask them to do both of these things. I told you before we got started that I was a little concerned about us using Webflow as the first example because Webflow has a giant team. They have a lot of resources. What does this mean for somebody who doesn't have a giant team? Is it not possible? It's actually the most interesting, one of the most interesting channels for small companies with limited resources. And the reason why is because in search engine optimization when early stage companies ask me for help and what their strategy should be, I tell them to not work on SEO because they're not going to perform. You need enough authority to be able to perform an in search. And a lot of other channels are pretty crowded. So it's, it's, you know, especially hard to cold start problem with the early stage startups. And this is one of the most interesting channels. And the reason why is because, because you can do offsite optimization with no brand. So if I am a brand new Aero Management Software company and I want to compete with large incumbents, if I'm making viral YouTube videos and, you know, a TikTok video and Instagram story and a Reddit thread, I can be the best product tomorrow. So the time to impact is fast. And I can grow on these other channels without really having a substantial brand. It's actually one of the most interesting channels. One of my friends and listeners created your 360 AI. It just launched. And essentially what it does is it will ask you some questions using like a voice. It seems like it's an 11 labs voice. Understand what you're trying to learn about your leadership and management style. Then it'll go and contact the people who work with you, who you told it to, and have a conversation with them to understand how they're working with you and get the answers to the things you're wondering about. And basically it comes back to you and it summarizes what you could do to be a better leader, a better performer at your job. If he wanted to do this, to start to rank, this is not a very sexy topic, right? There are not a bunch of YouTube videos on how to do a 360 feedback. What would you recommend for someone like him? Yeah, that's exactly what I would do. And, and this is non entertaining. And so I would do exactly what I just mentioned. So how do you, so I would create, I would create a YouTube video about the various use cases of my products. So there's 360 reviews. There's like how to give difficult feedback, how to receive difficult feedback. So what are all the problems that your product is a solution for? And I would make videos on each of them. Then I would try to become a real Reddit user. And there's probably threads of people asking about management and how to, how to get, how to get 360 feedback or 360 feedback is helpful because it solves the problem of not getting feedback generally. Like 360 feedback is a particular design solution for the problem, which is I need feedback from my team about how I'm performing. So like, what are all those problems and solutions? Find Reddit threads about those? Be a real person, say who you are, give useful information. Don't say check out my product. Just say, Hey, I'm Ethan from the 360 review dot AI or whatever. Here, you know, here's how I would approach it. I would do that. And the time to impact might be very fast. And actually product time is probably another interesting channel where a brand new product can quickly go viral there. And that might actually get picked up in the alms as well. Okay, that is huge. All right. Now that I understand what's possible, let's talk about some of the myths. And the first thing that I was thinking is SEO is basically dead. In fact, here's here's why I was thinking that because I know that you disagree with it. It's not so much that people aren't using search engines, they're going to use it for a long time. There's some people using the white and yellow pages, I think it's not that it's dead, that it doesn't work. It's just that they're killers like you out there who are so good at it, I can't compete. What do you say about the SEO is death worry that I have? Yeah, so there's a few different, let's break it down into a few different arguments. So argument one is what you just said, which is it's too hard, it's crowded. And I mean, I do think that it's hard. I think it's hard for early stage companies, which is why I mentioned when I want to start up, ask me what they should do in SEO, I say, don't work on it and wait until you're larger. Wait until you're, you know, series A or later, maybe series B and later, because it's hard. Like maybe you could get a little bit of impact, but you're not going to get millions of visits if you're a small company. But if you're a big company, you can do really well. It's, you know, it's a massive channel. It's, I think it's the second largest channel after paid advertising. So I think it's 5x, at least 5x LMS, maybe 10x, depending on the category. So it's way bigger than, than AEO. So it's a massive channel, but yeah, I mean, it is hard. So depending on how big you are, you should decide whether or not it's too hard, but it's not that I am so good. It's that authority is, you need enough sufficient authority to compete on competitive terms. But if you have that, then, then you can be competitive. So it's not so crowded that no one can do it. It just depends on who, so that's argument one. Argument two is that SEO is, sorry, search is being used less and people are using chat instead of search. And that's simply not true. There's clear data showing that that's not true. Similar web, I have data showing that the overlap, basically everyone who uses chat GPD uses Google. Nearly all of them, I think it's like 90, 90% plus or something. So people are not using it instead of search. Just like they're not, a lot of people use TikTok to search and Instagram to search. Like if you're looking for travel ideas or restaurants, a lot of people use TikTok and Instagram, but that doesn't mean that they no longer use search. That's just another channel that gets added. Same with chat. It's another channel that gets added. The slice of the pie for Google search stays the same. The pie just gets bigger and there's new slices added and chat GPD is another slice to the pie. Okay. All right. Fair enough. Let's then talk about the other thing that you told me was a myth, which is that they're, they're different. You kind of dispelled some of that, right? You're saying a lot of the work that we do for SEO helps out with AEO answer engine optimization. The other thing that you told me before we got started is LLM.txt is not used. Are you saying that we're talking about that, that thing that you can put on a site that says, don't crawl my site and the LLMs are supposed to respect it and say, okay, I'm not going to take that data in. They don't, they're not paying attention to what you're saying. So there's robots.txt and then there's LLMs.txt. Robots.txt is to say, don't crawl my site. And as far as I know, everyone respects that, including LLMs. LLMs.txt is different. So LLMs.txt is an interesting case study in the how myths spread, how misinformation spreads. So LLMs.txt was an idea from a blogger, I forget their name, where they said, this might be a good idea. Like how do we should send structured data to LLMs. And that is a good idea. Actually, you know, there's all this stuff recently about MCP and MCP is basically how do you send structured information to LLMs. I think it's Jeremy Howard from Answer AI is what I just saw. I think so. I think so, yeah. And ironically, he does not have his own, there's not an LLMs.txt on the site. Okay. But so it was a, it was a good idea that he proposed. And then other people saw the idea. And then they repeat it to each other. And they say, Hey, have you heard of LLMs.txt? Like, no, I haven't heard of that. And then eventually, you're like, yeah, I've heard of that from five people. And then that I heard about it becomes, it must be true because I've heard about it. If you're an idea enough times, it becomes a fact. And then people start looking for evidence that it works, even though it doesn't work. And then you're like, well, it got crawled. My LLMs.txt got crawled, so it must be working. But of course, any page on your, your contact page gets crawled. The fact that it gets crawled doesn't tell you anything about whether or not. This is an accepted structured protocol for sending information to LLMs. Then Google, John Mueller, Google said, I don't know of anyone who uses this. But still, if you hear enough times about this idea, then it is a fact. And so that's why this is still a myth that most people believe. And if you ask ChatGPT if it uses LLMs.txt, it will tell you that it does not. At least the chat says it doesn't. And I get it. People are concerned that their stuff is going to be crawled, but I'm the opposite. I want more of my stuff to be crawled. I want more of my information to be out there. I want more traffic coming over to me because of it. And frankly, even if I don't get traffic, I just want people to be able to use and benefit from what I'm doing here. And I think most people are in that boat, but I can understand some companies saying no. So the intention of LLMs.txt is to get more traffic and to send the information about it's not to block. It's to send more information to the LLMs so that you get more traffic. But it doesn't work. People generally don't know how answer engine optimization works. And so that's why they're grasping at straws, trying to come up with ideas. And if those ideas sound like they might make sense and they get repeated enough times, those myths become facts. And LLMs.txt is one of many examples of false information and myths. Okay. So then how do you know if something works? They're not telling you what you should be doing, right? And you're trying to get an advantage over anyone else who might have any publicly available information. How do you, Ethan, know if you've done a thing that you got a result because of the thing that you do and how do you get that iterative process fast enough so you can keep iterating? The way that you know anything is true is with a scientific method. So you come up with a hypothesis, you test it and you reproduce it. So how do you come up with hypotheses? You come up with hypotheses by doing analysis. So you do analysis and you say something like best website builder. It seems like these sites are ranking. Maybe the reason why is for this, like it seems like YouTube is appearing. So hypothesis analysis, pattern match, YouTube videos are showing up, Reddit threads are showing up. Hypothesis, adding a YouTube video will show up in citations and will cause my answer to, will cause me to become an answer and to rank better within the answer. So then you set up an analysis, you take a control and a test group. So for Webflow, there's 100 topics or 100 questions we want to rank for. For 50 of those, we're not going to do anything. And for 50 of them, we're going to do something. We're going to add YouTube videos for those 50. I'm going to set up tracking for all 100 for, let's say for two weeks, four weeks, so that I have a baseline. And then for these 50, I'm going to add YouTube videos and for these other 50, I'm going to do nothing at all. And then I'm going to see if the ones with YouTube videos showed up. So did the YouTube video appear in the citations and did my answer go up? And then I look at that. The reason why the control matters is because LMs are going a lot and charts are just going to naturally go up because they're going a lot. So if you don't have a control, you don't know whether or not you caused an effect. But you know, the way that you know anything to be true is with the scientific method and AEO is no exception. And so we're looking at maybe four weeks to get a result. Do things, I'm right, right? I'm looking at your face to try to get a sense of it. Yeah. I would probably do two weeks pre and I would do four weeks post, so six weeks total. And then don't things change so fast that by the time you get your result, the whole thing could have shifted? No. No, they're not changing that much. No. And you know, there's, there's models and there's, I wouldn't say that they're dramatically changing. Maybe they're like, when is there a new model every six months or so? So you know, if there's like GPT five and four are different. And so I would say that you would get noisy data. It's the same with Google search, Google search changes their algorithms as well. But you know, again, if you have a control on a test group, then you'll know whether or not it was the new model or whether or not it was something that you did. I wouldn't say that the models are changing so much that, you know, a YouTube video worked before and it didn't work. I wouldn't expect that the underlying strategy to wildly change within six weeks. It's possible, but you know, the new models come out, I think every six months or so. I heard you say on Lenny's podcast that affiliate mentions help. Affiliate are basically people you pay. With Google, you have to disclose that and you have, and it's not as effective to get them to link. What is it like for LLMs? It does work in Google. If you search for the best credit card in Google, you'll get a bunch of affiliate sites. And that's been true for a long time. And it works in LLMs. Tech Radar is a good example. Tech Radar shows up a huge amount. It's one of the most cited domains in LLMs. And you'll see a snippet at the top saying we're an affiliate site and it does really well. There's no, there's no evidence that a paid for mention does worse than a non paid for mention. Okay. So then am I going after sites and offering to pay them affiliate commission just so I can get links? Yes. Mentions. Just get them to mention. Doesn't need to be the whole structure data thing that I would need otherwise. The more the better, but just a mention is sufficient. Okay. Cora, Stack Overflow, those communities all helpful. Go in and respond to as many questions as possible. Yes. It depends on your category, but yes. How can I tell which questions I should be targeting? Like what are the top ask questions? Yeah. There's, there's, what would be great as if chat GPT gave us a chat GPT search console like Google search console. I said, here's the search volume for these prompts, but they don't do that. I don't anticipate they'll do that soon. I think they'll eventually do that as they build out ads, but they don't, they don't give you that information. So there's a couple ways to do this. The first way is what I would suggest, which is to use search data. And so if I'm a brand new company, I would start with the keywords that I want to rank for and then make them into questions. How I would pick the keywords would probably be which keywords, what I want to pay the most for. So either I'm already doing paid search and I use my paid search keywords, or I look at my competitors paid search keywords. So those are the money terms. So if I'm, you know, if I'm a rippling competitor, I would go look at gusto deal and rippling and see what, what are they doing for paid search in the top paid search search keywords and then just make them into questions. The other option is to use, uh, subsets of real data, first party data through browser extensions or ISPs and things like that. Or like there's a, you know, a bunch of LM companies that might sell you actual real data. Like there's many, many chatbots you could buy for, you know, chat you between won't sell you their data, but maybe a less popular LM will sell you their data. And then you'd have a subset of data. It's kind of like doing a poll for who's going to, who's going to win the presidency. You ask a thousand people, hopefully your sample is representative and you say, this is the average. This is what I think 300 million people are going to do. The trick though, or the difficulty though is when you're polling for who's going to win to be president, that your sample needs to be representative and it's hard to get a representative sample. Same with alums. It's hard to get a truly representative example of the whole world or of the United States. Also the, I mentioned that the tail is larger and longer. So that means that having these subsets of data, the smaller the subset, the less it'll tell you about the tail. Now, if you had first party data from everyone, then, or for half of the world, let's say, that would be great. And then you would have very accurate information, but you know, it comes down to what's the size of the sample that you're going after and how representative it is. I'm skeptical that there's enough information to do that really well, which is why I say, I would just use search data for now and then wait for chat GBT to give you the real data. Are there any chatbots that are making their data available that are selling it? I think so. I think I talked to one that was considering doing that. So I think that there are, I mean, there's all these chatbots that are not monetizing well and they need to monetize. I wouldn't say that there's a known marketplace for chatbots selling their data, but I think that they would sell it to you. But, you know, if I'm working with a chatbot that is made for dating, let's say, my AI girlfriend, I don't think that that's going to tell you, I don't think that's going to be useful for Webflow to know what they should be doing. And I think a lot of these chatbots are very specific applications. So again, it goes back to, is this a representative sample? I imagine it might be one of these like iPhone apps that somebody built that people have on their phones. Is that right? That would be representative. But I guess at this point, everyone has chat GBT and Gemini. Why do they need a third one of these or perplexity? The majors are all over. Usually the other ones that are used a lot are application specific, like my AI girlfriend or my AI doctor or my AI partner or my AI business coach. Got it. And then if I find one that's specific to my industry, then I can go and buy their questions and I've got something that's related at least to me. All right. I got you on this. Okay. You said that there isn't a search console available. What tools are available right now? Could you give me a list of some that are useful? I have a list of 60 tools on our website. So if you search for AEO tools, you'll probably get our page with 60 plus of these tools. The one that's right at the top of your site. Yes. Okay. And they are all useful and they all do tracking. So some of my favorites are peak AI and Scrunch and Surfer and Athena. I actually know all of them. Personally, I have no financial relationship with any of them, but they're all really good tools. And then there's 56 other good tools. So I would pick one of those. These all do answer tracking. None of them do optimization yet. So there are tools that will start to do optimization. And I believe by the time this podcast gets published, ClearScope will show the first optimization tool. So ClearScope pioneered content scoring for search, where you take a keyword like website builder and then it gives you like a Grammarly report, but it's for SEO, like include these terms. This is how your page is going to perform in search. And you get a B minus. And if you add these words, then you'll get an A plus. The exact same thing can be true for LMS. We can have a content scoring tool. So then they're launching that imminently. And so that'll be the first content scoring optimization tool for AI. And then you're going to see that probably a bunch of people are going to copy them as they copied their SEO version of that. They'll copy the AI version of that. But so there's tracking tools. There's, there'll be content scoring tools. There will probably be other tools like there might be tools around finding questions. We actually have a tool to find questions. It's like medium quality because, you know, we're using search data. But those are the three classes of tools. But today there's tracking tools. And then soon there will be these find me questions and then optimize my content. All right. I see the tools. It's in a different site, section of the site. And it really is an overwhelming list of tools. And also the, the funding rounds. It's not the one that I thought earlier, this link to at the top of all of them. Is there anything that you're going to daily that like you can't live without and give, give people some recommendations? I think all of them, all their career use cases you can't live without. So, well, they'll have the same feature. I see you're just saying, look, pick these use cases and then hear the different tools that exist for them. Yes. There are 60 tracking tools. There are no tools that do a lot else. Soon there will be 60 content scoring tools. As soon as ClearScope launches their content scoring, everyone else will have that as well. And then maybe I'm not sure on the tell me which questions, maybe people will, will build that. But for sure, everyone's going to have content scoring. But I think the answer to your question is who's unique and special. And the answer is no one. If because it's impossible to be unique and special. And the reason it's impossible to be unique and special is because these are not hard to build. These are commodity, these are easy to build relatively commodity tools with no technical defensibility or algorithm defensibility or brand defense, maybe brand defensibility. But, but all of these things are easy to build. And so that's why use their everyone copy ClearScope on their SEO content scoring and why they will copy them on their LM content scoring. All right. Tell me if I'm wrong here. To me, it feels like this is a big opportunity for an agency to be built. SEO, mature industry, tough competitors like you, it's really hard for people to get in. AEO newer, you can go after smaller companies that don't have enough money and presence to go for SEO yet. You can deliver results though, it will take a couple of months for it to start to show up. So they have to, you have to maybe sign a three month agreement minimum with your client. What do you think of that as a business right now to launch? Great. That's, that's the business I'm building. And specifically because I think it's great. So the reason it's great is because of what you mentioned, which is that it's a new channel and there's no one's mastered. I mean, I wouldn't even say that I have mastered it. I'm, I'm, I'm in the early stages of mastery, but I have, I'm not a master yet. I have many good ideas and a few case studies. So eventually I have more case studies and then I will have mastered it, but I haven't, I don't think anyone has. And so I think it's a very interesting space for services. And again, not just answer and optimization, but all of those adjacent channels Reddit, LinkedIn, Cora, like there's no Cora agency. There's not many Reddit agencies. There's no one who's great at making use case B2B YouTube videos. So there's all this green field stuff that would be that companies need that are very impactful and no futon, no one are masters of that. So I think it's especially interesting, which is exactly why I'm building that. So wait, the vision that you're saying is, look, someone should create a Cora only agency and all they do is they sign a B2B client and they go to Cora and they answer all the questions in their name as possible with as thorough response as possible. Am I right? That's one. Not only Cora, I would do, I would do all of those things. Cora Reddit YouTube. Yes. Cora Reddit YouTube. If you just do Cora, that's extremely niche and probably, and companies don't want to work with 10 agencies. They want to work with one. So the interesting agency that I would build in, that I am building is there's ancestral optimization and these five other things. So we're going to do all of those things. And they all connect to each other. And the five other things are this user generated content. There is creating pages. What else is there? There's UGC. There's video affiliates. I would say outreach. Depending on the category, they're there might be some domain specific stuff like for engineers stack overflow would be interesting. But yeah, it's probably like four core ones. And then depending on the category, there would be a fifth one. There's a company called taco.co. I interviewed Chirag, the co the founder of the company. They're operating at sub 15,000 a month. Is there an opportunity for someone listening to us to say, I'm going to create a similar company. I'm going to charge 10,000 a month and I'm going to do all these, maybe not all the things that you just talked about. Reddit Cora YouTube has a way of getting into answers in LLMs. Is that is that a million dollar business idea? For sure. Yeah, you could definitely build an agency with with smaller deal sizes. And it could be bigger than us. The number one problem that we had when I was starting is it's hard to get things launched because every company is resource constrained. And the number one reason why why things wouldn't work was not that the idea was wrong, but it's that the idea did not get implemented. And that's just because companies are resource constrained, everyone's resource constrained. And so we solve that by we do everything. And so then we have a whole team where we can design, implement, write, do analytics, like everything from top to bottom. And that and then you have to have a whole team for that now that not every company needs to hire and rent an entire team, they have their own team. And then you could have a consultant giving the strategic advice and then you have to implement it. Or you would say, well, I don't need to do all five of these things. I could just do Reddit or I could just do YouTube. That's totally fine. There's tons of companies. Most companies don't work too expensive for. So the market potential is actually larger for someone who has a smaller offering than what we have. And and for sure, there's an opportunity there. It's just, you know, you run the risk of perhaps not everything gets implemented, but that that's totally fine. Why why you and not so many other people who tried to do this in SEO? I have asked a few people in your space, I said, why, Ethan? One thing that came up a lot was Ethan's so frickin good at getting media. He's good for some reason. People like talking and writing about you, right? That's a big thing. So I just started doing marketing for our company in January of this year to that of this year. Did you do Lenny's podcast like two years ago? I did Lenny's podcast. I think it was two or three years ago, but I didn't do that much. I and before Lenny's podcast, we built the agency before Lenny's podcast five years ago. So it's been around for five years. And so all of our early deal, we've never done any sales. So we we've never done any outbound sales. And all of our deals early on were referral based. So it's just people that I know are people asking was good at something. And so that the that was caused by getting doing really good work and getting great outcomes. And then people refer. And so then I finally did marketing in earnest, starting in January. It's a lot of work to do marketing, because there's traditional marketing, and then there's the marketing that I we do traditional marketing is mediocre content. And that does well, but I don't want to do mediocre content. And I cannot, I cannot hire someone to do my thought leadership for me. I have to personally do this. And it's very challenging to write a 5,000 word article about how to do answer engine optimization. And so then I just didn't do that until January. So then I said, I'm just going to spend my weekends writing out all of my ideas into content. And hopefully it does well. And then it did better than I expected. So the marketing is only like the marketing in earnest has only been around for 10 months. So it's phase one is just do really good work. And then people will tell others about it. And then phase two is describe the ways that you did that great work, and then market it. What's been the ways that you've been describing it? I see you on LinkedIn, you've got posts. A lot of the posts are kind of referring back to things that you've done in other places, like a blog post, like your most recent Lenny podcast. Where else are you writing and posting the stuff that's actually useful? I write playbooks and studies on our site. So we do how like how to do answer engine optimization or how to do internal links, how to do topical authority. So we have long posts on our site. And then we have studies. So the studies we have, there's almost very little studies in our category. I think ahrefs and SCM rush both do pretty good studies, but there's not a lot of original research. And so we just do original research. Like we just posted an original piece of original research showing that half of the content published on the internet is generated by AI. And no one had looked it up. So we post that on our site. And then I write about that in LinkedIn posts. And then I do webinars like this one. And that's it. And I just started using Reddit. So I'm starting to comment on Reddit threads, but I'm like two weeks in personally going in and doing that 100% of the time. I never, no one ever does that on my behalf because I just I could I could delegate mediocrity, but and that and mediocrity works, but I that's not what we do. And so I have to personally do it. In fact, I asked you if you can give people an artifact, something they can actually go and use afterwards. And you said, I want to create something original for you. And so you said, I want to create an AO roadmap for your listeners. What's the AO roadmap? I'll I'll I'll answer a different question, then I'll answer that questions. So the biggest problem in SEO is wasted time. Most things don't work. And so most of your time is on things that are true, but zero impact. And so we did a webinar with reforge about the 5%. And so there's a few things that are very impactful in SEO. And so we built a roadmap out of what are the what are the few things that are impactful for SEO, and only do those things. And then I think we want to build that for answer and optimization. Now, how do we know what's impactful is to do a bunch of experiments and have a bunch of case studies, and we only have a few we don't have 50 case studies. So I think I know what what's what are the few impactful things. And so those impactful things are just do SEO, like do all the things that you do for SEO, number one, number two is the head, which is owned off site. So Reddit, YouTube affiliates, it's like, do that. And then the tail, which is help center. And so what I think my roadmap would be is step one is what questions should I care about. So I start with my search or paid search keywords that I care the most about step one, step two, make them into questions. Three is track them, put them in any of the 60 tools that I mentioned, and track them, and then figure out what's being cited. And then when you figure out what's being cited, do citation optimization. Probably it's going to be some mix of Reddit, YouTube affiliates, maybe some other stuff for your category. And then I would build strategies for each of those. And then I would do the long tail, which is help center use, use case pages, integration pages. So that would that would be my roadmap. All right, two questions before we end. The first is, if I want to brag about this interview and I want to brag about you, but you've been very sedate here and very like humble. And that's not enough for me to get a good hook and good social pose for getting people to come back and read to this. Give me a couple of things that I could say that would tell people Ethan is a frickin baller. You got to come and listen to him. What would that be or graphite? Let's say are frickin ballers. Well, people will tell me that I make Mark Zuckerberg look animated and, you know, like a really cheerful person. So one could argue that the level of sedation makes me unique and therefore provocative and memorable, which I think is probably at least the memorable part, I think is true. But your question is what's something remarkable and memorable about me? Yeah, he was able to do this, then you better pay attention. Well, one of the projects that I'm most proud of is working with the masterclass team to build out their content and SEO strategy that did very well. I think a lot of the it definitely influenced many other SEO strategies. So I think that that's one of a few projects that I'm most proud of. I would say I also played a significant role in the 10 best the 10 best X for Y title tags, like 15 years ago. And I also played a substantial role in the download interstitials that said, sorry, you can't use the mobile website, you have to download the app. And I was not the only one, but I for either of those, but I definitely helped proliferate both of those, which I wouldn't necessarily say were good contributions for the world. But the masterclass SEO, I use my talents for good. I'm on your LinkedIn profile right now. I look at masterclass, it says you're an advisor, and then it says, basically, you helped grow organic growth, there's no numbers, two sentences. Like what were you able to take them from where to where? Well, they didn't have SEO before. So I write and then I worked with their team. I mean, their team is, you know, it's interesting that there's the LinkedIn mafia and the PayPal mafia and then there's the masterclass mafia, I think. So the people that I worked with are all doing amazing things like the CEO of AirOps, Alex, I worked with at masterclass, who's now building air ops, which is like a hugely successful company. So, so yeah, there was no SEO. We built the strategy from scratch. I think it, I don't know exactly what it went to is probably 10 million plus per month, but that did really well. I believe it was 10 million visits a month. I'm not certain, but it was, you know, it was a substantial amount of traffic. And then, you know, working with that team, working with Alex, working with Nairi, a few other people, Mallory and Hillary are all on doing amazing things. I was, I was not, you know, a full-time employee, but I would like to think of myself as sort of a peripheral member of the masterclass mafia. All right. Close it down on a personal note. You and I both added to our list of interests over COVID, the exact same thing. You showed me a photo, which maybe we can flash on the screen here. You got to tell people about it because I think this is so frickin' fun. When the quarantine started, I always liked to think of constraints as opportunities. And so I have a design background. And so adding constraints actually allows you to come up with interesting design solutions that you wouldn't have come up with otherwise. And so I cannot leave my house. How do I make my house the most amazing place possible? I like nature. How do I bring nature to my house? I have a place in San Francisco and I have a garden and I said, "I'm going to get into starting fires." Or, I mean starting fires, you know, hopefully contained fires. And so I got a fire pit and actually I had an old cast iron oven from 1880. And I, it was just sitting in the corner of the garden. I'm like, "I should use that." So then I ordered and I need really premium wood. So like, how do I make my, how do I bring nature? How do I make this fire amazing? So I ordered wood online that's delivered from Chico that's old walnut and almond trees that no longer bear nuts. And then that's all chopped up. And then I would chop wood in the garden to fit it in the small hole of the old cast iron oven. And then I would have fires out of my garden. And also an interesting thing was that during the quarantine, I would see all these birds that I had never seen before. And, you know, fire, or not fireflies, but like interesting insects because the, you know, nature could now roam free because there are humans destroying their habitat. So I just hang out in the garden and have a fire and look at the birds. And it was great. I did the same thing. It was so great to sit in the backyard staring at a fire. Even though I had my phone in my pocket in the beginning, I would find myself listening to music or I would find myself reading my Kindle. I then eventually took a tent back there and I would just go take some space. And now building a fire, sitting around it is just part of something that I do since I moved to Texas, especially even more. But it's such a good bonding experience with people. There's something about just staring in the fire and letting your conversation go that's been amazing and amazing. And I think that I used to expect it to be something that you could only do in places like Texas where you have a lot of land. San Francisco, you got a little backyard. You could make a wonderful fire. I never thought to go to Chico. This is Chico, California and get special wood for it. I totally get it. It creates the vibe. All right. Thank you so, yes. Buy a fire extinguisher. One suggestion. If you're in San Francisco and you're starting fires, make sure to buy a fire extinguisher. I did not do that. And then you know what? I was stupid too. I didn't know that the wood was different. I just thought wood is wood. And so I would see like scrap wood that people put out there and I would put it in the fire. That thing lights up like crazy because I don't know what they treat it with. And then eventually people told me you're inhaling it. That's how much of an amateur I was. And then I bought the logs that are meant to kind of the package logs. Absolutely terrible. You get really good wood and you learn how to light it. It takes a long time. It smells really good and it's a lot more contained. 100%. Yes. I do have a fire extinguisher now. Ethan, thank you so much for doing this. Thank you for having me. Hell yeah. Thanks. Bye, everyone.