Episode 2 – Interview with SageSEO CEO Peter Yeargin | The Boost

Pablo Calvo (00:00):

Welcome to the Boost. I’m Pablo Calvo and I’m sitting here with Peter Yeargen, CEO of Sage SEO. And we’re at Geekdom. We have a great interview in store and I hope you stay tuned. Thank you very much for joining us today. Thanks 

Peter Yeargin  (00:20):

For having me. 

Pablo Calvo (00:22):

I’m actually pretty excited from the time that we first met to learn about your product and had a chance to play around with it yesterday, and it’s amazing what the product actually can do as far as helping content creators and content writers just facilitate the process of developing new content. And I was just wondering if you could tell us a little bit about the history of the product and essentially how it came to be. 

Peter Yeargin  (00:48):

Yeah, I’m happy to. So basically, we started out as a completely different startup. We were a Q & A platform. We were kind of building Quora, but if you know what that is, like a Q & A expert sort of answer type of thing. And we realized that for that space, people needed to trust the experts when they answered their questions. And so we figured, hey, there’s a market for this where they’ll pay the experts for their expertise. It was a classic, if you’ve been in the startup space, it was a classic, classic two-sided marketplace where you needed people that had questions, you needed experts that had answers. And so we built the platform and we were iterating on it, and then we realized, Hey, we’ve got to bring users in and how do we get users? And so we started running a bunch of Google and Facebook ads, and I quickly realized, Hey, we’re going to go broke doing this. 

Peter Yeargin  (01:39):

I’m technical. What do we know about SEO? How can we get some organic traffic to our website? And so we started experimenting with a bunch of different things, mostly around technical SEO, so we can make our question and answer pages really friendly for Google when they crawled them. So they were looking for information on there, and we got pretty good at technical SEO, but it wasn’t kind of moving the needle yet. And so we started doing keyword research and trying to figure out, Hey, can we answer some of our own questions and start to generate some traffic based on using keywords in the answers? I know it’s cheating, but people like Reddit did it all the time when you kind of look at their history. And so after a little while we were like, well, we’re getting really good at this, but I wonder how good we are at this. So we launched a blog just to test how good we’d gotten at SEO, and the idea was to connect the blog to the q and a platform and use that as a funnel to bring users into the q and a platform. 

Peter Yeargin  (02:33):

It was a small business tech kind of blog. We had articles like how to troubleshoot Google Nest wifi or How to Connect Zoom to your G Suite account. And that blog, within a few months with our efforts around SEO started doing very well, and we were like, wow, we know this is working pretty well. And then at the same time, we ended up in a startup accelerator and within the accelerator we were like, you know what? There’s a lot of people telling us they’re having the same problems with SEO. They don’t know how to do it. It’s really complicated. It’s confusing the ways to do SEO are live in a bunch of different tools in a bunch of different applications, whether it’s Google applications, whether it’s SEO applications, keyword research, things like that. And so we started talking to some of our other fellow founders in the cohort and said, Hey, are you having the same problem? 

Peter Yeargin  (03:19):

And I remember this vividly, it was in August of 2021, and over a two week period, we ended up talking to 14 different startups, some of which were in our cohort, some of which were just friends that I’d met through being in the startup community. And after two weeks, we went from, I think I asked the first person, Hey, would you pay me to do your SEO monthly? Would you pay me 10 bucks? Because they always say, in the startup world, will someone pay you to do what you help them with? And of course, the first person said, yeah, of course, $10 a month. By the end of two weeks, I was at a hundred dollars a month and we were at $1,200 in monthly recurring revenue. And after a year and a half of doing the q and a platform, we had zero in recurring revenue. And I was like, I think this is our business. So we just pivoted hard and became, say, Jesse o, we didn’t have an application, we didn’t have anything other than notion and email. So the first nine months we just worked with customers using email and notion as our sort of communication mechanisms and while we were learning and while we were figuring out, Hey, how do we productize this? And then over those six to nine months, we started building out the application. 

Pablo Calvo (04:21):

And what were the, I guess, first steps for the application? What did you find were the things that you needed to get out to get people to really get excited about the product? Right. 

Peter Yeargin  (04:30):

Well, when you look at SEO, it’s all about content marketing for your website. It’s how do I create content that both my target audience is going to find interesting and that Google is going to be able to get good metrics around people coming and sticking around because what’s your average session time? How long does someone stick around your website? Once they come to the first page, what’s their bounce rate, so to speak? If they come to one page, do they go and open up another page on your website or do they just leave? That’s right. And those are the kinds of metrics that Google thinks are interesting. And so we realized in order to get people to click on the second page, they needed to get a good experience from the first page. And so how to write the content, how to make sure you integrate good keyword research into the content, and how do you build a platform to simplify the whole workflow was really my focus because SEO, like I said, is very hard. And so we wanted to simplify the very basics first and then build upon sort of a good foundation. Well, 

Pablo Calvo (05:23):

From personal experience, I mean with Digi Boost, we have SEO clients. And let me tell you, if you don’t already know, it’s highly, highly manual. So it takes a long time to do it, and you don’t necessarily know what all of the steps that you’re taking, what impact they’ll have right away. You go ahead and optimize a page, you go ahead and start going through your checklist of everything that you need to do for proper content optimization, and you realize, oh, I skipped a step. I should have done this, I should have done. So you have your best practices, you line them out as best as you can, but there are so many different steps. And what I realized is that when you ask a client, in my case, I’ll ask a client to say, oh, well, can you give us some ideas on content? And there’s a big gap there, right? It’s not just about the treatment of the content, it’s also coming up with the subject matter on many times subject and content that you’re asking a copywriter to really write something that they’re not really 100% clear on. So guess what? These days you’re using AI, right? You are using that to be able to identify that content and then come up with a list of things that the client would like to write about and then give them ideas to give you some feedback. But Sage, SEO does that for you, right? 

Peter Yeargin  (06:40):

Yeah. And I mean, I think if you take a step back and think about what was happening in August of 2021, OpenAI wasn’t a thing yet, right? People knew about large language models, they knew that AI, the people that were in the technology space knew it was coming, but most people didn’t. The public didn’t know about it. And so when we first started, we were just an SEO company that was connecting to places like Google Analytics and Google Search Console and saying, Hey, where’s the data for us to inform how well we’re doing from an SEO perspective and from a content perspective, and where does it live and how do we incorporate it into the platform? And so we started out using a bunch of APIs to do that. I think my first exposure to OpenAI was in April of 2022, I was at a startup conference in California in Northern California and Open OpenAI had a booth there, and it was about a week after the announcement of Microsoft investing a billion dollars in them. And so I was like, what the heck is this? Because I’d looked at ai, AI was out there, and I’d looked at what’s the tool that was the main AI writing tool, 

Peter Yeargin  (07:48):

Jasper. Jasper, thank you. I looked at Jasper and spent a lot of time, Hey, do we want to incorporate Jasper? And I was like, I don’t like the output yet. It’s not good enough. And when I saw open AI in April, I was like, wow, this is light years ahead of where Jasper was. And I come to find out Jasper had been using open AI in the backend, but they hadn’t used the most recent Chat GPT version three. They were using much older models. And when three came out and then 3.5, it was exponential changes in quality of content coming out of ai. And so we started looking at it like everyone else, Hey, do we want the AI to generate content for our customers and make it easy? And so like everybody else, we incorporated OpenAI into the backend into our content writer. So if you got stuck, you could have OpenAI ask you and suggest a paragraph or suggest an outline or whatever it happened to be, suggest an article topic, for example. 

Peter Yeargin  (08:37):

But what we quickly realized was OpenAI, the real power in AI in large language models in general, is their ability to process a whole bunch of information, whether it be text, whether it be statistics, whether it be large amounts of content, and then make very proactive, succinct suggestions about what to do next. And so we were like, well, AI doesn’t need to necessarily be the writer for us. It needs to be the engine underneath the covers that’s simplifying each step along the way of the content marketing process. And we started envisioning this idea of a content marketing flywheel, and the idea was, Hey, understand your audience. Plan your content based on your understanding of your audience. Once you’ve done that, do great keyword research. Understand what people are also asking on Google, some of the free tools, some of the paid tools, those are the types of things that can help you write the article and then publish the content. 

Peter Yeargin  (09:34):

And I think a lot of tools stop there, but from my perspective, the real power of automation was being able to automate the measurement process of SEO. How is this working? Which content’s resonating, which keywords are showing up that are pointing back to our articles? And then how do we make it easier for people using our software to incorporate the metrics into their next round of content? That’s right. And if you could use the data and proactively help them process it, it would make it much easier to have a more efficient set of content the next round, and you end up back to the content flywheel where it gets more and more efficient at creating content that resonates with your target customers. 

Pablo Calvo (10:14):

And what’s interesting about the tool is that when you’re working with so many different platforms, you have to jump from platform to platform to platform. You write the content, you have to then go into Google Analytics to see what kind of engagement numbers you have. You have to obviously review how long people are on the page, are they engaging with it? Are they clicking anything? Are they just dropping in and leaving? But you have to go to three or four different platforms to just come up with one focal point on that you want to target in on what is your learning on that article. And Sage SEO gives you the ability to look at those metrics all in one place, which is very, very, I mean, that facilitates a lot. I mean, it makes the process significantly easier to gather data quickly. And if you’re talking about serving people that are more like call them novices in the SEO space, that gap is sometimes very time consuming. So I feel like at a minimum, there’s a significant amount of time savings that you’re also introducing into the process by which people learn the game of SEO, right? Because it is a game, right? I mean, you do certainly have to master all of those. I 

Peter Yeargin  (11:25):

Played a lot of games as a kid too, so that’s why I think it was really fun for me too. So yes. 

Pablo Calvo (11:30):

Well, you mentioned meeting with OpenAI in California, right? But as of recently, here we are December 1st, 2023, and just in the past two weeks there was huge, huge news with Sam Altman and the board at OpenAI and where he was kicked out, and then he was asked to come back and I mean, it was a major tug of war. I mean, it made international news. Now obviously your product is based on that platform as a foundation. Are there any concerns from you or do you have any comments on what’s happening there with that? 

Peter Yeargin  (12:07):

Yeah, I mean, I’m not going to lie to you. I think when all of that was happening, I was keeping a close eye on that because like you said, we’ve used OpenAI for a lot of our backend, primarily because the APIs are so easy to use, they make it really easy for developers and software companies to access their software programmatically. And since we’re programmatic application, OpenAI was very attractive to me because of that. I think there are a lot of solutions out there that are open source that are AWS, obviously Google Cloud platform. They’ve all got various components of that, but you’ve got to stitch it together in a much more complex fashion. So as OpenAI was going through that, I was concerned that it was just was going to be much harder for us to do what we do now with different platforms. And open AI is an easy button, and for me as a startup, it’s similar to AWS. When AWS came out, it made it very easy for startups to start small and scale up with cloud instead of having to buy their own infrastructure. And open AI is kind of the same way when it comes to ai. They make it really easy for someone to start out quickly and then scale up as they need, and it’s very, very inexpensive at the beginning. And even now, I’m sure they’ll raise their prices, but that’s why OpenAI was so attractive to me was how easy it was for my developers to build to it. So 

Pablo Calvo (13:22):

Do you think the dust has settled with that or do you think it’s still an ongoing conversation? 

Peter Yeargin  (13:27):

I think the way that all went down the dust is pretty settled. I mean, when you oust the CEO and then 700 of seven 70 employees say they’re going to follow him to Microsoft, and Microsoft has already invested 10 billion in your company, I think you realize quickly, Hey, that was dumb. And I think that on the flip side of that, I think it was as tumultuous of an exercise as it was for open AI to go through. I do think they’re going to be better coming out of it because I believe that accountability around AI safety was a big concern of the board when they went through that and it became very clear that, Hey, this is so important. We’re willing to kick Sam Altman out of the company. And it wasn’t the right move, but it was a move that I think put a lot of focus on AI safety because I think as someone who understands AI quite well, I also understand how little, even the people that do AI and build AI systems understand about how it works. And I think understanding what’s going on under the covers with large language models, with how AI operates is really paramount to making sure that we can continue to deploy AI safely. 

Pablo Calvo (14:40):

Those are all very, very good points. It’s interesting is that everyone, my experience personally is leveraging ai, albeit a manual basis, not using great tools like what you’re developing with Sage, SEO, but as you start learning how to piece these products together in the end, so not just produce content but good content, content that not only helps you from an optimization standpoint, but at the end of the day, an engagement standpoint, I mean, you do need to take that content and you do need to edit it. You do need to work with it. So there is still high touch required between the production teams that are putting this content together, editing it, reviewing it, and making sure that it actually speaks truthfully about the points that you’re trying to make, whatever that topic matter actually is. So in that process, Sage, SEO is learning as you mentioned. So as it’s learning, as you’re a business, let’s just say content creator selling widgets, and now you’ve written a number of articles to be able to sell widgets better and talk about more diverse content matter related to those widgets, how does that process improve over time? Does it give you better suggestions? Does it remember things that you’ve written in the past to feed off of that to give you more streamlined information based on previous learning? How does that process work? 

Peter Yeargin  (16:02):

Yeah, I think the answer is all of the above. As AI has continued to evolve, and as open AI has continued to evolve, they’re exposing a lot of functionality that allows us to go deeper and deeper into the content personalization around who a company is and what their brand is and what their messaging is. It’s allowing us to train customized models around a company and then use that customized model for doing just about anything. So you can imagine a model that’s an AI model that’s trained on all of the company’s website content, all their product content that’s in their website. OpenAI is making this very easy to do, and it was always very easy to download a website’s content just by going to their site map file and downloading everything. But what OpenAI has done is make it very easy to ingest that content into a custom trained model that knows who you are and knows what you do. 

Peter Yeargin  (16:56):

And from that perspective, you can then take that model and use it to do just about anything. You can use it to write new content. You can use it as a classifier for relevancy. Is X, Y, Z relevant to this company? How relevant is it to it? This company, you can imagine the plays around that for keyword research. You can imagine the plays around that for coming up with article suggestions and ideas, right? That’s one of the first tools that we rolled out is using OpenAI underneath the covers to analyze a content piece, understand how it’s performing in Google search, and then suggest three more articles to write about around that article topic. And not only suggest articles to write about based on the content that was already existing, but suggest articles that target keywords that your content is already ranking for. But maybe it’s on page two or page five where you’re not getting clicks yet. 

Peter Yeargin  (17:45):

The idea is, hey, Google is starting to look at you as an authority on a topic, but not enough of an authority to rank on page one where you’re going to get the clicks. So we use AI instead of a person having to go to a Google search console data and analyze all that data themselves, which most of our customers are not capable of doing that, right? They don’t understand the tools. Most of ours don’t even know they exist. I think that’s why marketing is such a difficult job, is the data is living in so many different places. So anything we can do with open AI to simplify that process and automate it, so we obfuscate the complexity of it from the customer, that’s what we’re doing. Well, 

Pablo Calvo (18:23):

The interesting thing as an agency, from my perspective, we see AI as it’s either the harbinger of bad things to come where the agency model’s dead. You see the articles come up all the time. I take the other approach, which is you can do much more with fewer resources and you can serve a much larger client base. So from my standpoint, I see it as a multiplier if used properly, but tools like yours are critical because at the same time, if you’re doing everything separately, it still takes a lot of time regardless of how quickly you can produce the content. Everything else still takes a lot. And there’s no, what’s the word? Almost like there’s no record button in a manual process where a process that’s housed like it is on Sage, SEO, you have the ability to learn, and that’s something that you can now task the AI with as opposed to tasking it to a person, which sometimes that expertise level, sometimes we’ll walk out the door when they take another job. So I think from that standpoint, as an agency owner myself, I see that as a significant positive. But outside of agencies, what types of other clients do you have and what industries would you say are the ones that are adopting most quickly? 

Peter Yeargin  (19:37):

I think you and I talked about this a little bit before we started recording, was kind of who our target customers have been, and we both started in the same space. We both started targeting small businesses. We both started targeting startups. And I think that both of us got to the same conclusion in the sense that, hey, what we’ve been building is relevant to more up market as well, so we can serve the small business and the startup to simplify SEO and simplify content marketing. But from our perspective specifically, we’ve been building what we’ve been thinking about as an AI content engine. It’s a content engine that incorporates ai, but not in the way you think about it as a content writer. It incorporates AI into every step of the content marketing workflow. And because of that, that is a valuable tool to agency owners. It’s a valuable tool to, especially when simplified to small business and startups, it’s a valuable to tool for marketing teams that work for large and mid-size enterprises. So that’s kind of where we think we’re going to continue to grow is we’re going to have a combination of being able to serve the individual or the small business with three to five people, people, but we’ll also be able to serve a marketing team that has 20 people working for a large enterprise. 

Pablo Calvo (20:53):

So it’s a product that’s scalable to the size of the team. Absolutely. Now, as far as the customers that you have that sign up for your services, how does it work with multiple seats? I mean, is there a pricing model for additional seats within the team? 

Peter Yeargin  (21:07):

Yeah, so far what we’ve done is we specific to agencies, the way our agencies are OEM’ing, our software is they will have one account per customer. As we’ve been growing, once we get up to, I’ve sort of got a critical threshold in mind as far as number of agencies that are working with us where we’re going to build some custom overlay software that will aggregate all of those individual accounts and report up at the agency level as well. So you’ll be able to go into each individual account, do everything you need within a customer, and then you’ll also be able to pull up your agency level data and dashboards and statistics. So that’s one area that I think is really interesting. 

Pablo Calvo (21:44):

Wonderful. No, as far as my review of the product and working on actually with the import of the articles that we had on digiboost.com, one of the things that I noted was grading your own work, even previous work, I mean the part of the SEO, the one that’s doing the work is that you don’t just sort of leave it alone. You have to revisit some of the content that you’ve produced. Part of that is also reviewing the quality level, whether or not it needs to be updated, but when you run it through a tool that actually graze it for you, completely independent of other tools like hrefs or what’s the other one? Moz. SEMrush. Yeah, SEMrush, exactly. I mean, as you’re running it through those products, people don’t really look at their previous content. So if you’re looking at it in real time, it gives you an idea to like, okay, well I need to improve these things even though I might’ve written them two years ago. It’s still that process and it reminds you to do it, otherwise it’s kind of in this empty kind of vacuous hole that you forget all about and then you realize, why aren’t I producing as well as I used to? And you realize that all of your older content is losing traction. So I think that’s an incredible, I’d say benefit from the product 

Peter Yeargin  (23:04):

Itself. Well even imagine once we’ve got all the data for your content in an engine that’s for your company, imagine AI and changes in how an article is performing in Google search and understanding when those dip below a certain level or when those go up above a certain level and those indicators being watched automatically for you by the engine and then proactively telling you this article is starting to drop, or this article’s doing really well, write more about it. That’s the kind of stuff that we can do with the articles being in our platform and with the fact that we’re proactively monitoring Google itself, which is Google Search console and the stats coming in from that. So I think we haven’t built that yet, but that’s on our roadmap for 2024, like simplifying things. 

Pablo Calvo (23:56):

That was exactly my follow-up question with regard to your roadmap, what types of exciting news features are you looking to develop in 2024, 

Peter Yeargin  (24:05):

Right? Yeah. So everything around simplifying each step of that content workflow, especially around understanding what’s working and what’s not working and being able to use that data to make smarter, faster, better decisions. And open AI as a solution is so good at doing that with some prompt engineering, right? I mean, we spend a lot of time testing, and for those of you who don’t know what prompt engineering is, it’s basically what do you tell open AI to do and what data do you give it to do it? And then based on the results you get back, sometimes it’s going to suck and sometimes it’s going to be pretty good quality from open AI and iterating on how do you prompt it to get better output. So that’s prompt engineering and the 

Pablo Calvo (24:46):

Key and the quality of that. And that prompting obviously is the other piece that people still have yet to really learn, right? They’re still figuring that out. 

Peter Yeargin  (24:53):

There’s a lot of intellectual property around how do you create better prompts and do better prompt engineering. I think that’s where a lot of our secret sauce sauces is. We’ve gotten very good at the prompt engineering aspect to get the output we need. So I don’t know, is it a good time to take a quick break for me to do a demo for us on one of our new features called the AI Smart article suggestion feature? We can take a quick break and do that and then we’ll come right back. 

Pablo Calvo (25:16):

Okay, great. 

Peter Yeargin  (25:18):

Hey, everybody just want to do this quick demo. Like I mentioned in the interview with Pablo and the Digi Boost team. The idea and the concept of what we do here is we use OpenAI to analyze analytics coming from Google search data. And one of the great things about Google that they do is they provide you with a bunch of rich data that happens to be in Google Search Console. For those who do SEO, we all know that Google Search Console is the place where you go to determine how well you’re doing from a search perspective. How often is your content showing up in a Google search, how many impressions it’s getting, how many clicks it’s getting once it does show up in someone’s search, and then which search terms led to you getting those clicks. On the Sage app side of things, we try to simplify a lot of that work. 

Peter Yeargin  (26:04):

So you as a user or you as a marketing team or you as an employee of an agency, don’t have to do a lot of that analysis yourself because if we can save you a few minutes here and there, it’s worth it. It’s weight and gold. So what we’re looking at now is our app interface in the center here. We’ve got all of the articles that this particular customer has written, and this customer does color analysis to help match people’s clothing with their eye color and hair color so they pick the right color for the clothes. Let’s take a quick look at one of these articles. Are you a winter type? This article is about the winter seasonal colors and which ones are most important to you to match up with your hair and skin color? As you can see, we’ve got a bunch of data coming from Google Search Console over the last three months. 

Peter Yeargin  (26:52):

For example, they got 16,000 clicks on just this one article of Are You a Winter Type? If I were to open that article up and take a look at it, you could see, hey, this kind of takes you right to the article. So it makes it easy for you to see which article is being analyzed and which article you’re getting stats from. The power of this is when we look at the search query data and which search terms have led to people clicking on their website. So for example, winter color Palette led to 1200 clicks out of 61,000 search impressions for this customer. And if you go back and you look up top, you can see that that was just over the last three months. So with OpenAI, what we do over here on the left-hand side is we look at your top search queries, your top 50 search queries to be specific, and we look for opportunities to write similar content to the article that you’re looking at. 

Peter Yeargin  (27:41):

And the reason and the power of this is to say, Hey, I’ve got a really high performing article. I want to write more content like this. Sometimes it’s hard to figure out which topic to write about or what are some good topics to write about. And we use OpenAI to not only suggest topics to write about, but it’s suggesting topics based on the search terms you’re already performing well for and or the ones that you should be targeting because you’re starting to perform well, but you’re not quite on page one where we all know you get all the clicks. So if I were to click on one of these three suggested articles, let’s say this middle one, they’re suggesting writing an article called Dress for Success Outfit Ideas for Winter Skin Tone. You can see that it targets a particular keyword, winter skin tone, and it also targets three more additional keywords. 

Peter Yeargin  (28:23):

These keywords are directly from Google Search Console data, and it even tells you why it’s suggesting to write this article. So this is the power of open ai because I can feed it and we can feed it a whole bunch of information. It could parse it quickly, and it can suggest really powerful ideas that are going to target article ideas and article suggestions based on how you’re already performing. So we’re building lots of tools like this where OpenAI operates under the covers to really speed up the analysis process. And hopefully you found this really interesting. As always, you can reach out to me at [email protected] or feel free to come to our [email protected] and all the best. 

Pablo Calvo (29:05):

That’s an incredible overview of the product. I’m sure that people are going to be just as excited as I am right now when they get a chance to actually try it. And I want to thank you again for joining us. So thank you to CEO Peter Yeargen of Sage, SEO. I’m Pablo Calvo with The Boost, and I hope to see you again soon.