How Autonomous SEO Agents Are Replacing Traditional SEO Work

0:00Thenuka Karunaratne: The Voices of Search podcast is a proud member of the I Hear Everything podcast network. Looking to launch or scale your podcast? I Hear Everything delivers podcast production, growth, and monetization solutions that transform your words into profit. Ready to give your brand a voice? Then visit IHearEverything.com. Welcome to the Voices of Search podcast, a member of the I Hear Everything podcast network. Ready to expedite your company's organic growth efforts? Sit back, relax, and get ready for your daily dose of search engine optimization wisdom. Here's today's host of the Voices of Search podcast, Jordan Cooney.

0:49150%. That's how much ChatGPT referral traffic to websites increased as OpenAI started showing more links in its responses. And whenever a new source of traffic starts growing, marketers do what marketers do. They chase it. New tactics, new frameworks, new acronyms. Suddenly, everyone's talking about GEO. AI is changing the way teams research, create content, and execute campaigns. And it forces SEO professionals to rethink what skills actually matter. I'm Jordan Cooney, and joining me today is Sanuka, CEO and co-founder at Sunbeam and Daydream. Today, we're going to talk about autonomous SEO, the GEO hype cycle, and what the future of SEO teams looks like when AI becomes part of the workflow.

1:37Tanuka, welcome to the Voice of the Search podcast. Thank you, sir. Good to see you again. Likewise. Tanuka, we've known each other for a few years now. And when we first met, you're building up, Daydream, in maybe what we can call AI 0.1, which is a lot of SEOs and organizations were thinking about AI in this utility way. How do we grow and how do we build content systems, tools with AI. And now you're taking that to an entirely new frontier. Tell us about your journey. Tell us about how you've gone through this experience of building in the SEO industry, in the search industry, from Daydream to where you are now with your latest startup.

2:36yeah so when i started daydream in 2023 the premise was that we wanted to initially um use the large language models that were available at the time which is mainly gpt3 because we started right before chat gpt was released and then gp3.5 and so on to automate a lot of programmatic seo so in the first call like era of the AI plus SEO, you could use LLMs to automate certain workflows, but in very like small chunks. They weren't good at handling really complex reasoning tasks. Reasoning models weren't available at that time. So we built a platform drawing on my background doing programmatic SEO.

3:24And the idea was, hey, before programmatic SEO was a niche strategy used by Zapier and Zillow. But now if you can use large language models, you can automate long tail content creation for a much broader set of terms. So we built that SaaS product, took it to heads of marketing, heads of growth, and very quickly realized the problem was that they didn't know how to do the strategic part of SEO. So you could give them a platform that would actually automate a lot of the content creation, but they didn't know how to actually think about how to do this or how to put together the strategy above it.

4:04And from there, we unintentionally decided that, okay, well, not unintentionally, but because of that feedback, we realized, okay, well, this is an interesting problem. they don't know they have the strategy problem. So one way you could solve this is we could just go do it for them and make it a fully managed service. And they'll pay a good amount more for the fully managed service versus just the software. So we did that, started as a managed service for just programmatic SEO, then expanded into all the other aspects of SEO, editorial SEO, keyword strategy, eventually AI visibility. So morphed into a fully AI native SEO and AI search agency, raised$20 million plus in venture capital, served a lot of great companies, including Notion, Replik, Clay, lots of great companies.

4:59Tell me about this daydream evolution because I think this is a really interesting segue of, this is not something that's often super understood by a lot of SEOs, right? And you've got a lot of great brands that you've helped build this programmatic approach with. And I think that that's a huge unlock. It's a huge unlock for you. It's a huge unlock in terms of where we are with SEO today, which is there are scalable things that you can do to achieve reaching your poor audience, traffic goals, what have you. So just for a second, before you go into where you took that, how you took that learning to where you are today, what was the crowning achievement of Daydream?

5:42Like what was the biggest unlock that you guys were able to generate? I mean, I think I'm mostly just happy with the quality of work. Like for the Shorty Awards, we won like best SEO SCM campaign for one of our customers. And it was a finalist for like the AI search category for them as well. So I think like my perspective was that all hype aside, What AI or any technology should do is allow you to materialize a great outcome with more efficiency or maybe with more enjoyment. And I would say probably the thing I'm most proud of is just the group of people that we assemble, the kinds of customers we work with.

6:27a lot of fun experiences also working with companies ranging from really small pre-seed startups that were just trying to figure out SEO all the way to companies that were already public so for me I would say the craft of working across dozens and dozens of customer accounts seeing not in theory but in practice how every small aspect of SEO could or could not be automated with AI was probably the most valuable part of that for me. Because it's easy to think, oh, you just throw AI at anything and then boom, it's more efficient. This is the narrative you get. But there's a lot of nuance, right? How would you apply AI to automate the right aspects of technical SEO versus keyword strategy versus content?

7:10Those things are very different. And it's not as easy as just spitting some prompt and then boom, you have a great outcome. There's an incredible amount of nuance based on every case. And I don't think you get a full appreciation for that until you have to put your money where your mouth is, sign a customer, and then you have to deliver for them. That was a really challenging experience. I mean, I love this story because I think, Tameka, when I look at your journey with Daydreaming and what you guys built, you got 15 years of SEO knowledge condensed into 2 years of just running hard and building a product that helps clients and customers figure out how of scale with SEO, which just isn't easy to do.

7:53And it's not, it doesn't work every single time when you're trying to move that fast to scale. But you genuinely were building that vision and that capability with Daydream. From Daydream to like where you are today, like what motivated you to think about pivoting? What motivated you to think about like, hey, there's a next evolution of what we need to build in the SEO industry? yeah so while we were working at uh like on daydream um if you'll recall the the thing that got us to build an agency and a fully managed service provider was that there was no way to do the strategic parts of seo and at the time the only way that we could do this at a quality we were happy with was uh with a human in the loop in a fully managed way so um around like the towards the middle half of last year, obviously things have come a long way from where they were in 2023.

8:50Reasoning capabilities of models are much better amongst a number of things. So we started to experiment with the possibility of, is it possible now to automate not just content creation like we had in the past, but the entire chain of SEO work from technical SEO to strategy to keyword research, search, competitive analysis, AI search optimization. We look at all of this stuff. And so we built an agent in-house. And initially, we tasked it with automating a lot of the work of our internal team members.

9:25And obviously, our vision was that this would be the best possible product if it was possible to build, which is a pure software approach to handling many of the core aspects of SEO. And for probably like, you know, eight months or so, we were just hitting dead ends. But towards the beginning of this year, there was a number of inflection points. I think one was definitely in the quality of the models, but also a few breakthroughs we had internally that convinced us that a fully autonomous SEO agent was possible to build. Now is the time to do it. And, you know, that became its own entity, which is Sunbeam now, the autonomous SEO agent.

10:03So let's get into this. I mean, I have like 100 questions already. Super excited about this because functionally speaking, there's one piece of this, which is, you know, it's kind of the evolution that got you to something that works. You mentioned the models, the model evolution. I got some definite questions about that. Like what is working in these models and that's why do these models help you get to a place where the product and the delivery of what you're prompting or doing is working. But before we get to that, when you think about having a system that can help SEOs do this, where do you see the value here for the user, the person behind the screen?

10:44Why is this such a critical need for organic discovery to have better platforms and capabilities Yeah, well, I think you could think about this in the context of two different types of users. So for us, we work with a lot of companies that are very early. So a consultant might be able to give them some strategy, but they don't have the executional resources. Or maybe they want to work with an agency, but a lot of the best agencies are priced out for them. So for those users, the process of navigating the strategic considerations around how do I think about my SEO strategy? How do I learn from what my competitors have done?

11:29How do I formulate like some scaffolding of, hey, this is the direction I should go in and then actually take all the steps to materialize that strategy from creating the content, optimizing technical SEO, like all of this kind of stuff. So for a certain subset of our people that we work with, we are giving them access to something they never had before, which is just very solid SEO strategic guidance and execution that was previously priced out. There's a second group of people, which are people that already know how to do SEO, that are SEO specialists. And for them, I would say their job is just evolving, I think, in a really interesting way right now, where to me, it actually mirrors what you might have seen with software engineers over the last couple of years.

12:14So rather than writing every line of code by hand, a lot of the role of a software engineer today is to architect the system and essentially leverage agents to do a lot of the work while taking care of a lot of the strategic considerations around that. So I think for SEO practitioners, you could say that a very similar thing is going to happen. So instead of doing every keyword research report pull yourself and looking through a technical SEO report and trying to figure out what to solve and then giving a ticket to someone on the engineering team. I think we're approaching a world where, hey, a lot of the research will be automated.

12:53The agents will not only identify your technical SEO issues, but they'll fix them. It'll generate large parts of the content that you're going to generate for you, or at least a lot of source material that then you can use and format yourself. So I think the role of an SEO person is also evolving to be more of a similar thing to what you see with software engineers, which is the architect strategy, They reconcile the strategy with the business customer product context, but their role shifts towards coordinating agents to do a lot of the work that they may have previously done. So I think this is more of the world that we're coming to.

13:33So you're telling me, Tanouk, is that you are the Henry Ford of SEO automation. You are assembly lining this and we're going to bring out cars every day. right and like I mean I'm being a little bit of a jokester on this but like the reality is that like I think there's something super valuable around the creation of a technology like this one and one of them you alluded to which is there's a couple of different core audiences that you need to meet in terms of where they're at in their journey of learning and growing in organic, right? One of those areas is people who are early just don't know a lot or don't have the time or capacity to learn this vertical, and the cost is too high for them to acquire it, right?

14:25The other is this, like, the industry is just changing, and there's a tremendous need for individual SEO operators to have better capabilities around them to execute the work that they want to accomplish. When you think about solving for these audiences, when you think about building a platform and technology that's going to help them solve the SEO journey, a lot of people go to this place, which is, hey, it's really hard to determine what performance is. It's really hard to determine what is useful or valuable. How are you solving that question of value in this product?

15:07I mean, I think ultimately value, you know, even on the daydream side, when we were running these engagements, like this is the question every customer has about SEO, which is like, hey, what is my payoff? When is it going to happen? What can I expect at this time? I'm sure as you know, like SEO is one of the most difficult channels to forecast performance for in the early, early days. I think as you get bigger, it becomes more predictable. but the the thing that I would say there is it doesn't change that much in the way that we thought about it on the agency side and the way we think about it with the agent which is that ultimately what the agent is responsible for is creating performance so you know when you first publish your first set of pages the initial leading indicator might be are they getting index and crawl then it's are they getting clicks and impressions then what is the trajectory of their ranking performance.

16:01Like to us, what we really felt like we want the quality bar to measure up to is what you would expect if you were working with someone that was just skilled. Now there are certain limitations and certain strengths that LLMs have versus what you might expect working with like a consultant or an agency. And I think that's something we should talk about at some point as well, because I guess I have the experience of running both the agency and like building an autonomous agent. Right. Don't worry. Don't worry, Timo. We're going to get to talking about autopilot and full autonomy of agents. But like with Sunbeam, let's talk about Sunbeam for just a second here.

16:44That, and like, let's get to why you decided to launch the product and have it focus on where marketers need help in organic. What is the type of customer that you think is in the most critical need today? And then where is the customer, you know, 6 to 12 months from now? Because one of the things that's changing right now is this entire industry. The entire SEO industry has changed more in the last three years than it did in the 20 years prior. No joke. And I think that you're one of the shining examples of how we should be innovating in this space. And so what I want you to center on is the customer.

17:38Who's the person that we're using here? And if you told me like, hey, three months from now, this is who we've captured. This is the audience that we're winning. and then 6 to 12 months from now. So I'm even giving you short... I mean, for most people, those are short windows. But I think I'm giving you short windows because that customer is going to grow and expand to a very different customer in that short of a period of time. Yeah, it's a good question. For us right now, we wanted to serve the audience that we felt like does not have access to a better alternative right now. And that's why we focus on founders and early stage teams.

18:16So these are teams that don't want to pay a consultant 15 grand for a report and then have to do everything themselves. They can't work often with the really good agencies because they just are, you know, take a good amount of capital to invest in as well. And for these folks, this is a novel thing. They're like, OK, wow, there's finally something for me that I can use that handles everything from start to finish. And I think that's something that they were not able to do before. So I think historically, if you look at even a company like Canva and how Canva grew, Canva grew because it gave the capabilities around editing, creating images, assets to an entire class of people that didn't have it before.

18:57And so I think when a new technology diffuses, typically who adopts it are the adopts it earlier are the people that just couldn't do that before. In the longer term, like six to 12 months and down the line, we feel like, especially with the pace of improvement of some of the models and also the product, we fully expect that we'll start working with larger companies and start embedding into organizations that already have an SEO strategy. But obviously, there's a different challenge around that from working with teams that have a green field. But we found certainly founders right now are very open and receptive towards using autonomous agents, which is something I think that larger companies still have some like brand risk sensitivity around.

19:43And I think that's why we started with some of these earlier stage companies for now. And so as you think about autonomy and this core customer that you're reaching out to today, you mentioned earlier consultants and agencies and where their role fits within this ecosystem. But is this an autoplay? Is this Elon telling everybody that Tesla should just roam three through the world and just drive themselves without any limits? Like, where is that limit? And like, you know, how should autonomy in the way we work in SEO be? That's my first question. My second question is, how does Sunbeam help with automated work and workflows?

20:32And then my last question is, what does success look like? What's working? Yeah, I'm really glad you asked this question, actually. So to me, autopilot and autonomy are two fairly different things. And I'll tell you why. So autopilot has this connotation of I download this thing. I let it do whatever it wants. I give it no input. I spend no time trying to work on it. And if I buy this, it's meant to go solve SEO for me end to end. I don't think this works. I'm sure you'd agree, right? There are products that claim this autopilot thing. And what you end up with is mostly a lot of garbage. And so there's a mirror example to if you hired someone and you're like, okay, just go do stuff.

21:16You never talk to them. You lock them in a room. Like, how is that going to create like a great outcome? Autonomy to me means that the agent is proactive and does not wait for a response. But there's still a sizable component here of you give it feedback on what it creates. You give it the context it needs around your business. if it you know like there's still a component of working with it it is proactive it's not a it's not a you know it's not a workflow it's not just an agent that you tell something to do and then it stops it is a proactive agent that operates without you prompting it to but it doesn't mean that it's on autopilot yeah so I think that's a pretty critical thing that even in like our messaging and everything I'm trying to like really convey because I have the same like disdain for this autopilot approach to SEO.

22:07I don't think it works at all. If you think about like the eras of SEO tooling we've gone through, first you had like AI assisted tools. So, you know, they might scan your, even in a pre-LM world, like they might scan your piece of content. They may give you some suggestions, pull out certain paragraphs that need adjustments, so on and so forth. Then you had large language models come and then people started using entirely kind of like, like we started building AI workflows. So, you know, write this article end to end, create this keyword research report, so on and so forth. But now I think, but I think this to me was still a transitionary stage.

22:47To me, the best possible version of the, at the intersection of SEO and applied AI would be that you have a teammate who behaves and can act similarly to what you would expect a great team member to react like. And when you think about a great team member, a great team member is not just doing what you say. It goes out, it figures out what your competitors are doing. It looks at something you just pushed and tells you if it's having crawlability issues. It looks at your content and tells you, hey, is this performing or not? It does a lot of work for you without you asking it to, which I think is a fundamentally different paradigm from many AI tools in the space, which are trying to build workflows.

23:30And it's also fundamentally different from an approach where you talk to an agent, and it's a purely reactive experience. So like Cloud Code, and the way a lot of people use it right now is a purely reactive experience. There's a lot of tools that you're probably familiar with that do a lot of AI workflows. But this new segment of building teammates that are proactive is untouched. But I think that is close, gets much closer to what you expect when you hire a great person than anything before it. So that's kind of the positioning with Sunbeam. And in terms of your last question of, it was how you measure success, right?

24:09Yeah, yeah. How does this work? Like, how do you know that this is working? Yeah. Like, for us, when we work with customers, typically, there's an initial onboarding phase where the agent essentially gets context on everything the company does. After that, the way you measure performance is very similar to the way that you probably measure success with your clients, which is from the time they started and they onboarded to time milestones A, B, and C, you know, you could say a month, two months, three months. What is the velocity and quality of the work that is being produced? How much content has it produced?

24:43What are the results it's producing? in the beginning, I think the leading indicators are softer. So are the pages being created, crawled, indexed? Then very quickly, it goes to, are they ranking? Then from there, it goes to, what is the compounding rate of growth on the traffic? And then ultimately, it's, does the traffic convert? Time for a one minute break to hear from our sponsor, Previsible. So you're looking for SEO help and you got a couple of options. You could start replying to spam from agencies that claim they can get you to rank number one on Google. You can pay an hourly rate for a consultant who will inevitably nickel and dime you with hourly charges.

25:18Or you can work with a cookie cutter agency to quickly launch a strategy-less project with low success rate. None of those sound very good, now do they? Well, that's where Previsible's integrated consulting model comes in. Previsible draws from a collective 40 years of SEO and digital marketing experience to unlock your organic growth opportunities. They build custom solutions that combine strategy, technical expertise, content, and reporting to effectively operationalize SEO for your business. Pre-Fizzable's four-stage approach ensures that your SEO programs thrive by starting off with a strategy-first approach.

25:54Then they support you in your efforts to create quality content, help you identify technical issues, and most importantly, they'll work with your cross-functional teams to integrate your SEO strategies to make sure that your SEO budget actually drives results, not just your agency's bottom line. So join brands like Yelp, eBay, Canva, Atlassian, Square, all who rely on the SEO consultants at Previsible. For more information, go to previsible.io. That's previsible, P-R-E-V-I-S-I-B-L-E dot I-O.

26:28So for us, with each of the customers, that's what we're looking for. If they're not getting those, if they're not able to ladder through those markers of success, obviously it's not successful. So our sole focus is building a product that helps customers get through each layer of running their SEO strategy to a place where, hey, it is very obvious that the agent is providing value, creating performance, and therefore you want to continue working with it. So that's kind of how we think about it. So I got a couple of follow-on questions to this. And I really want to get your perspective on how Sunbeam and the way that you've built this product is different from what we're seeing today in the market.

27:11You brought up these kind of AI visibility tracking or these AI tooling that exists in the market. Many of them are chasing what they think is the next goldmine, which is task execution. Right? You do some sort of visibility tracking, then I'll go create new content. You do some sort of visibility tracking, I'll go monitor the technical health. You do some sort of visibility tracking, and I'll go and give you a signal when you're no longer showing up for something and have you re-optimize your content. These task flows is where a lot of these software companies are going to tokenize and build what they think is value and wealth behind their software and technology.

28:01How is Sunbeam different? How are you actually competing in this race against some of these now, I mean, arguably billion dollar valuation companies? Yeah, it's a great question, Jordan. I think if you, so the first thing is Sunbeam, Sunbeam has no workflows. So when you work with a, that's probably the easiest way to say it. Like to me, the idea that you outline step A, B, C, and then you run it, and then maybe you run it on a cron job. That cannot be the best experience and most effective experience for using agents in this category. That seems too narrow, limited, and relatively rigid. That is very specific intelligence, you could say, around a very specific workflow.

28:59But to me, when I think of you or when I think of any of the great SEO people that I work with, they are not running a workflow over and over again. They are looking at the environment. They learn from what the best companies are doing. They look holistically at how the current SEO strategy of the company reconciles with all of that. They look at the quality of the content. Are we using all the best assets we have? Once one layer of things is done, they are already on to the next wave of thinking about how do I take the SEO program to the next level? So these are things you don't get by following a workflow-based approach.

29:45A workflow-based approach is as good as you architecting a workflow and then telling an agent to run it. The more interesting way to handle this is the agent is constantly giving you new things that maybe you didn't even think about and then going and executing all the work that needs to be done, but it's not a workflow. It behaves much more similarly to the way that you'd expect a great team member to work, which is they have a problem set area and they are responsible for handling the work around that area and doing what they need to do. If it's strategic research, it's strategic research. If it's fixing technical SEO stuff, it's that.

30:18If it's updating content, it's that. If it's creating new stuff, it's that. Fundamentally different from a workflow. To me, I really believe that the workflow-based approach is an intermediary step in the AI adoption cycle towards something that... I think what we're approaching is something that's closer to fully autonomous agents. Right. A much more agentic approach, right? Because in this agentic approach, the theory that everyone has about AI, not just SEO and AI, right? Like all AI, right? Like in every way. Is that like the AI can actually make decisions about what to do next in a more efficient and effective way than a human can, right?

30:59And in that case, then technology is going to help us do SEO or do organic in a much more efficient way than any great SEO, myself or others, yourself included. Like it will beat us, right? AI is beating people in ping pong and chess for a long time, right? Like it's going to beat us in SEO at some point, right? Let's talk about that then. You know, here at Previsible, we've been tracking for almost a year now the LLM models and their model evolution against 50 SEO questions. These 50 SEO questions haven't changed. They're the same SEO questions we ask our own strategists and our own talent. And most people who know SEO real well can score at an 80 % right.

31:48But most models are not scoring at that level. Some of the lesser known or used models are scoring at less than 50 % accurate. Maybe Gronk, Llama, or DeepSeek. And then some of the more visible models that we see in the market, like Claude, ChatTBT, and Copilot, will score anywhere between 60 % to 70 % accurate. One of the interesting things we've seen, though, as this model evolution occurs, and we go from 5.4 to 5.5 and so forth and so on in terms of any of these models, is that there's at times regression. right like the model just doesn't keep getting better every single time there's moments in time where dips or declines and doesn't get a couple of the answers right how are you managing that expectation in a technology like sunbeam how are you thinking about the integrity of model evolution and you mentioned even earlier in our show here that because some of the improvements the model you've actually seen better results, which I agree with.

32:56Our benchmark study also proves that. But there are moments in time where there's regression. So how do you ensure that the integrity of the product is still going to be there for your clients? So we definitely have this problem as well. And that's why we will thoroughly test any new model on the same task before we just automatically upgrade to it. so I'd say that that would be the the the practical answer to that question I think you raised a very interesting point though which is you have I'm not sure exactly how you constructed those questions but you have the set of questions and you know the agents are scoring in the in the ranges that you mentioned and even though there's temporary regression it does seem like there's just the inevitable arc of how things are going is over the span of years like it's Think about this, how if someone had told you five years ago, 95%, 99 % of code will all be written by AI.

34:00That's crazy, right? Correct. That's basically arrived already. Same goes with AI answering SEO questions. If you would have asked me five years ago that AI is going to help answer 50 % of SEO questions, I'd be like, you're an idiot. You're crazy. There's no way that's going to happen, right? But no, they do. Even the bad models answer usually 50 % of the questions correctly. So that's a pretty fast evolution in a somewhat complex and subjective industry. Our industry is way more subjective than writing code. Writing code has to be able to fulfill a certain task. That is one plus one. But SEO is sometimes not one plus one.

34:38Sorry, I didn't mean to cut you off. Go ahead, Timothy. Yeah, it's interesting. So the question that I think about is actually this one. Right. So to me, it's very clear that, yeah, in the longer term, on any like discrete tasks, like doing research reports, doing technical SEO fixes, like these things are very clear to me. Agents are just going to be very good at doing. But the dilemma is this, right, which is the models will keep getting better. The agents will keep getting better. More and more companies will have access to all of these agents. So if you and your 10 next competitors all use the same agents, which are all riding the same improvements in the foundational models, what are the parameters for what determines who wins in SEO and who doesn't?

35:23That question is very interesting to me because that points to something else, which is what is the role of the human being as models continue to get better at an exponential rate? And I think that question, SEO, has not been explored very well. And it's actually a personal interest of mine. Because what I've observed is that the best SEO people, as agents come out, are only going more in demand, exponentially. They're getting more business than ever. And so it is very interesting to think about what this pace of advancement means for the career of the SEO. And even more importantly, the skill sets, right?

36:02Like, what are the skills that actually make you good at SEO or AI discovery? And like, why should those skills be sought after? You know, I mean, historically, I'll tell you, you know, being an SEO now and being in-house for a lot of my career, like people really value technical capabilities, right? but technical capabilities are now becoming maybe a little less important in an ecosystem where a lot of the technical tasks that an SEO can do are very AI automated, right? Like, if I prompt an AI model correctly, I can pretty much get the schema output I need. I don't need to sit there and write that out anymore.

36:45Those days are gone, right? um let's talk about some being though in the sense that there are these tasks that you guys do what what is the what's the landscape that you can cover and you touched on an area briefly in your previous answer that i think is is interesting which is there are other influences at play when it comes to success it's not just the technical foundation it's not just the content sometimes it's about the authority and the brand that the company has. And so how is Sunbeam helping in this landscape of solutions that exist in organic? Which ones do you guys solve for? And which ones do you not?

37:30And how should potential suitors of the software think about those different solutions?

37:40so um yeah as you know i think domain authority is probably the hardest thing to move um and i've long been a believer that domain authority is largely the byproduct of just product market fit if you have product market fit you're probably going to be successful you're going to get press it's going to raise the authority of the domain there are some um good techniques for building domain authority in an ethical good way. So, you know, digital PR is a good one. And a couple other things like that. But Sunbeam right now focuses mainly on the on-site aspects of SEO. So it will do, it will look at the context of your company and your competitors and formulate what your strategy roadmap should be in terms of the topics you should write about, what clusters they're in, all of that.

38:36It'll write the content. It will analyze your website for technical SEO issues and then write the code to go solve those technical SEO issues from end to end. So all the on-site aspects are what we're focused on. Domain authority is an interesting one because I think there are a lot of cool things you can do there. But there's a big constraint, which is I think that domain authority is more influenced by like non-SEO factors, you could call it, than not. Like I think it's like a lot of the best companies with the best domain authority that I've met have put zero effort into SEO. Like they are just products that have become super hot.

39:15And then obviously there are other companies that have just gotten there like incrementally by really grinding it out. But I think a lot of it also depends on the category. Like if you're an insurance company, how hard is it to get press and differentiate against every other insurance company? Maybe if you're an AI native SEO insurance company, you may get a lot of backlinks very quickly. But for that reason, I think off-page SEO is definitely a trickier domain. But I'm actually curious what you think about that. Yeah, I mean, I think, you know, we're in the midst of probably one of the biggest moments where that off-page factor is sought after.

39:55But to your point, it is a very complicated journey to seek it, right? If you don't have product market fit, if you don't have good press, you don't have a lot of these foundational off-page signals, it's really hard. It's really hard even if you buy as many placements as you can to generate the exposure you want from organic. And so I think that it's so important to build a good product. It's so important to be a good service provider. It's so important to establish your reputation. And establishing that reputation will then lead to many of those great off-page signals. Yeah. I couldn't agree more, Jordan.

40:37I think also, if you think about something like Reddit, Reddit has become very important in AI citations. But okay, so how are you going to go affect Reddit? The right way to do this would be that you automate maybe the component of it that is finding good opportunities to post on. And then you actually send in someone to give a really great, thoughtful, non-promotional answer. Or you build your own subreddit, which is maybe to answer customer support questions from your customers that use Reddit. And it works out that way, right? The bad way to do this, which you could automate, and I'm like air quoting that, is you could go spam a bunch of comments into a bunch of random stuff using AI.

41:15Is that a good approach? No. So this is where for us, like we're trying to be thoughtful about the way that we think about what the agent does. We don't want it to just do bad SEO. And there's a lot of bad SEO out there. So those are some of the nuances when we even think about which areas to cover, because if we cover an area, we want to do it in a great way, in a way that you would be proud to do. But this could change very quickly. So the next time we chat, I will probably tell you that we have built a bunch of stuff in this space as well. No, I love that. And I love your answer because I think this is one of the things that as marketers, especially growth marketers, there's so many growth marketers and startups.

41:57There's a lot of desire to latch on to some of the trends or some of the ideas that fundamentally just don't align with what success looks like. And I think that that's one of the dangerous places that we're in right now where there aren't great playbooks, there aren't great workflows for some of the activities that we're trying to do. And because of that, there are no real good guidelines or regulations to protect marketing teams and SEOs from trying to invest in them. And so like there is no real answer to your Reddit question that you just stipulated, right? Which is like, how should you go about Reddit?

42:39You could try all three of those tactics, but there's not really like some sort of Reddit guidelines. I mean, Reddit has some, you know, use policies that they have, but it doesn't necessarily mean that any of those three tactics are out of bounds. And I'm sure that if you talk to a dozen people that have tried Reddit marketing, each of of them have tried one or all of those. And so that's, I think, the place that we're in right now when it comes to a lot of the SEO strategies that are out there with AI is that you're getting very conflicting emotions and expectations. But we all know that the use of AI is going to be at the backbone of everything that we do.

43:19And so I want to kind of close out this section by asking you this really important question, which is, where is the marketer, where is the SEO, the individual that's responsible for organic, going to be a year from now? What tools are they going to have that they don't have today? What capabilities do they need to be thinking about that they don't have today? You have been one of the individuals in this industry who's been And thinking ahead on how to build not just the technology, but the practice of SEO and materializing that for brands and companies in Daydream and now in Sunbeam, the operators and individuals that are helping scale their companies.

44:06So what should these individuals be thinking about in terms of where this industry is going to be and what skills they need? I would say my overall prediction for how the SEO career profession evolves is that SEOs, now that much of what they previously had to spend their time on, like a lot of the tactical work gets automated. I think one pivotal piece of the SEO career path, which I think was already moving this way, but it's going to accelerate is SEO was starting to get classified as a product role at more and more companies in the last few years. I know Pinterest has done that. I think Coinbase has done that.

44:49But great SEO, I think, will end up requiring a lot more understanding of the product, the business, the customer, and how to expose the SEO promise from that in a more effective way. so this is very different i think from the very narrow tactical kind of like in my mind like low grade version of thinking about what an seo person can do which is oh they just come up with topics and they delegate it to a bunch of ai or contractors they publish it and then they keep updating it and do more right yeah i think it becomes much more strategic so if agents can do a lot of the work a lot of the ground level tactical work i think you end up becoming more strategic i think the other thing is, as this is a prediction that I've been thinking about more recently, but I am starting to also feel that many of the best SEO people will likely expand into other channels in the near-ish future, largely because if a lot of the tactical work of this one channel has been automated, and now you're more focused on strategy, my overall sense is that this frees up enough headspace and bandwidth to be able to handle other aspects of a company's marketing and distribution strategy outside of what was just narrowly called SEO.

46:15So to me, some of the promising areas that start to get encompassed in this could be product marketing, like even versus pages, right? Or alternatives to your competitors' pages. These are things that an SEO person's zone, but they're effectively product marketing. I think it starts expanding to that. It may also expand into more content marketing that it's not SEO content. SEO becomes one piece within content, but you have a lot of content that's great that people love that's just not super SEO optimized because it might be distributed on LinkedIn or some other channel versus search. So my overall sense is that SEO should one, try to create as much value at the strategic layer of reconciling business context with the SEO strategy.

47:00And the second thing I would say is that I'm noticing this morphing of not just SEO people, but a lot of marketers that have been channel specialists to becoming generalists as they get more leverage on their time. And I think that is my prediction for how the profession evolves. Love that. But now I have to move us on to our next segment, which is our lightning round. I'm going to ask you five questions from our topics today. and you're going to give me hopefully a quick 30 to 60 second answer on each of those. You ready? Great. Yep. Okay. What is the most overrated AI, GEO, SEO tactic right now?

47:41I would say it would be llms.pxt because this has been confirmed that it does not help with search and visibility. There's a separate argument to be made for authentic browsing use cases, but for search and visibility, Google made that pretty clear. And I think multiple other organizations have looked at websites with LMS.txt files to see if they have any positive influence on ranking and searchability and discoverability. And the answer is overwhelmingly no. Yeah, absolutely. What is the number one skill that SEOs and organic marketers must have? I would say it's product sense. If you don't understand the product well, how will you understand what to cover about it?

48:33In what detail? What makes that article better versus just, okay, there's some fundamental product sense you need to have to be a good SEO. That's so true. And like on that product piece, like connect that to the business. Tell me why that's so valuable to the core ethos of any company and what they're trying to accomplish. I think there's something like whether you're founding a company or you're working on a company or you're involved in some capacity in a company, if you are very disconnected from the holistic experience of that product. So what it does at a deep level, how customers receive it, what they have trouble with, what they don't, what customers complain about, when it's being evaluated, what questions come up from a prospect.

49:25These are all very important inputs in being able to understand, hey, from a search standpoint, what should I convey to the world about this product? If you don't have that, But how do you even do that job? At that point, you just do the worst version of this, which is you just purely mine keyword research tools and then you just do it high to low volume. I love how you connected that back to SEO, but it's so spot on. It's exactly the core of why that product skill is so necessary, not just for SEOs and marketers today, but just everybody who works in an organization. what's one thing ai still can't do well i think the macro strategy around really empathizing with the customer really understanding what the mission and focus of the team building the product is and understanding core product capabilities and tying all that together in a really smooth way i think is something that requires a lot of high-bent frequency like communication between human beings.

50:33And you can distill some of that into an agent through connecting with call recordings and your CRM and your help desk. But there's some element of what does the company do? What are our priorities? A lot of this is implicit, happens at water cooler conversations, on walks with your boss. These things are very hard to replicate. And I think that is very useful and important in today's world. What's one of the biggest misconceptions today around automating SEO? Autopilot doesn't work. The idea that you can autopilot your SEO, that is the biggest misconception of all time in this space right now.

51:12So let's take a look at like two to three years from now. What does the SEO team now start doing that they didn't do in the past? I'm biased on this answer, obviously, But I think the use of autonomous agents to screen for a lot of things that otherwise a person would spend their time doing, like tracking what every competitor is doing, creating a report, comparing, like tasks like that, to me, it's inevitable that people use agents to do that because synthesizing many, many data points very quickly into something that is clean and summarized is one of the use cases that agents excel at. And I think that's definitely coming.

51:55Okay, that wraps up this episode of the Voice of Search podcast. A huge thank you to Tanuka, CEO and co-founder at Sunbeam and Daydream for joining us. If you'd like to contact Tanuka, you can find a link to his LinkedIn profile in our show notes or on thevoicesofsearch.com. You can also visit his company website, sunbeam.ai. If you haven't subscribed yet and want a daily stream of SEO and content marketing knowledge in your podcast feed, hit the subscribe button in your podcast app or on YouTube and we'll be in your feed next week. Okay, that's all for today, but until next time, remember the answers are always in the data.