New SEO and AI Strategies You Should NOT Miss!

All right, so this next episode is a collaboration with HubSpot's Marketing Against the Grain podcast. So we got the CMO, Kip Bodnar, and they also have Ciarán Flanagan, who's their VP of Marketing on the pod. And then it's also Neil and myself, and we talk about the future of marketing. That's where the talk is focused on. We talk about the future of SEO, the future of CRO, what we're excited about. We talk about what's old that is carrying over into the future. It's a great episode. We think you're gonna enjoy it, and we'll see you inside.

Big fan of the Marketing School pod. I know you both have avid followings because you're deep practitioners of marketing and you're kind of deep at the intersection of both results and kind of emerging trends. And what I'm hoping to do today is take us through a tour of some of the most important aspects of marketing and talk about them and give people a new blueprint moving forward. And one of the places that I was hoping we could start was SEO. SEO, if you go read the SEO subreddit, it's a very depressing place. There's a lot of people talking about all the traffic they've lost, the websites they're shutting down. It's very, very doom and gloom. I think you all are probably more positive about the future of SEO and leveraging AI for SEO and those things than most people I talk to. And so I guess I'd love to kick it off and hear from both of you around What's happening with SEO? What's the state of the state and what should people who are trying to get traffic from search do in this kind of post-AI era?

Yeah, I'll let Neil go first because he has— he's been saying this quite a bit. He likes— he has a new definition of SEO that I think is catching on well. So Neil, why don't you kick it off first?

Yeah. So first off, we look at SEO as not search engine optimization. We look at a search everywhere optimization. You know, as a Google VP said, I won't read the quote word for word, but he was talking about how when young people are doing a search, they're now turning to TikTok and Instagram before Google. People are now asking ChatGPT for questions and hiring companies based on the results that they're getting on ChatGPT. Like, I was talking to this lady and I was like, how'd you find out about us? She's like, ChatGPT told me to hire your marketing firm, NP Digital. I was like, all right, sounds great.

Sweet.

So you can optimize for so many channels, right? YouTube has search, it's the 2nd most popular search engine outside of Google, which funny enough is owned by them. I don't have great stats on Baidu. It's a little bit tough to get stats on Baidu comparing to other channels because it's in China, but platforms are everywhere that have search now. The 2nd thing to keep in mind is when you see the doom and gloom on the internet, like Twitter and Reddit, It's usually from people who are very technical, and that's one end of the market. Danny Sullivan, who's the search liaison of Google, has stated every single year Google has driven more organic traffic to websites than the previous year. It doesn't matter that there's more zero clicks or any of that. Overall, net-net, they're driving more organic traffic to websites than the previous year. And even with the AI Overviews release, we saw an increase in traffic to our client sites, assuming you're doing SEO well, because you got another place to get cited. Not by much, a little bit more than a percent, but that's better than a drop. And the real point that I'm getting at is when you think of SEO, when people talk about, oh my God, look how much traffic I lost, my rankings went down. Well, if you lost rankings, someone else climbed rankings. It just doesn't happen where everyone loses rankings, right? Google has roughly 10 results on the first page. And I say roughly because it really does vary. But if you've dropped, someone has taken those spots and is getting traffic. So the concept that organic traffic or SEO is dead, not really. It's changing though, and you got to adapt with it, not just focus on Google, but there's still a ton of traffic to be had.

Yeah. And just to add some context to that. So the sales intelligence tool we use is Gong. And so I was telling Neil and I were talking about this about a week ago and I was just like, if I type in podcasts, okay, so 10% of the sales conversations involve podcasts in there. And then about 9 to 10% involve the keyword YouTube. The fact that we're here, Kip, we're on like, we're, this is a collaboration with the Marketing School Podcast and the Marketing Against the Grain, right? Yeah. We're both pushing this to YouTube. We're both going hard on podcasts, it's because the attention is being spread everywhere else. I think we tend to see the people that complain the most on, especially on Twitter or X, whatever you want to call it. It's what Neil mentioned. It's the SEOs that complain about it. And the SEOs that complain about it are typically, they put all their eggs in one basket and once they start to lose that traffic, they start to panic. And, you know, Neil and I, we've gotten a couple texts from friends this year talking about, oh my God, we've lost all our traffic, right? But some of these friends, they've bet most of their, most of their traffic into SEO. And it's no wonder why they're panicking, right? And so attention has changed up quite a bit. I mentioned that I would travel with friends this year and I look at their behavior and it's like they're not even using Google, they're just using TikTok to search for things, right? So we have to adapt with the changes and we have to roll with the punches. I think the people that complain the most are the ones that are unwilling to adapt.

Yeah, well, there's a few things I would bring up. The people who are losing the most search, to your point, are not— they're not diversified. They're all in on search, which means they're deeply into informational search. Like very how-to-based search versus transactional search queries. And those how-to-based search queries are getting more disrupted by AI overviews, by ChatGPT, by YouTube, by TikTok, all of those things. And that's, I think that's what's challenging. I think the advice we're giving people is like, look, you have to be multimedia by default. Like you have, like people are using video and podcasts more than ever before. And you're going to have to be there as search. Kind of basically fragments and dis— gosh, I can't even talk right now. It unbundles and it comes across all the different platforms. The second thing I think we're talking about today is that AI is a new force in recommendations. And it like, I call it like we're going from the word of mouth to word of machine era, which is like a hyperbole, but like it's, you know, you will go and just get an answer from ChatGPT that is far more in-depth and action-oriented than previous search queries would be. And like, I use the Replit agent to like build lightweight apps, and like that agent is telling me what APIs to use, what web infrastructure to use. It's making the buying choices for me. And so if we don't show up in those platforms, you're not gonna be set up for success. I think my follow-up question to both of you is, what do you think are the advantages in SEO right now? Or in terms of like, how can you use AI to actually get more traffic to your website versus worrying about the traffic disruption?

Yeah, I can go first. So what we do at my ad agency, Single Grain, so I'll use a case study here. So, you know, some people poo-poo on programmatic SEO. And the way you want to think about that is you're just creating, you know, location-based or templated So a TripAdvisor would be a good example. Fun things to do in Rome, fun things to do in Japan. And that's why they rank for millions and millions of keywords, right? Or like a Yelp, for example, or an Expedia. So you can generate these pages programmatically, but I think some people have taken that too aggressively. And so if you think about SEO 10, 15 years ago or so, there's a lot of these black hats that were just pushing out spam and they would rank for whatever. This programmatic content, to me, is still a way of spamming, but it's a way of spamming where you're adding value, I believe, to the web. And as long as you're doing it in a tasteful way, and you're putting out quality content, Google has said themselves that they don't care about it, right? Like, as long as it's helpful to humans, that's great. And so we have one client, they rank for about, they were ranking for maybe like 500, you know, bottom of the funnel keywords, transactional keywords. And then now they're ranking for about 7,000 or so. So it's not like we went hardcore for them. Like, we went, we produced some programmatic content for them, but not millions and millions of pages, right? And so I think some people tend to over-index or just not do it at all. It's like, oh, I'm not going to do AI content. or they do too much of it and it's low quality. I think it's a happy medium that works. And the one final thing I'll add before I flip it over to Neil, actually 2 things, is if you're creating maybe hundreds or thousands of these pages, and again, you're QAing them, you know, pretty closely and making sure the quality is good, that actually enables you to do programmatic conversion rate optimization where you can change the call to actions on these pages. And we're seeing about a 50% lift on these pages. I'll give you an example. On our website for Single Grain, and I'll slow down a little bit. On our website, Single Grain, our typical call to action says work with us. Now, maybe we have a page that talks about conversion rate optimization services for e-commerce. Well, in that case, we'll change the call to action on that page to optimize your conversions, which seems trivial, but it actually increases the conversions on those pages by about 50% or so. And so what we do there is we leverage OpenAI's API And then we just scan the context of that page and we'll just, we'll create a prompt for it and we'll make it like a, you know, 6-character or sorry, we try to make it like less than 6 words, the call to action. And it's worked out really well for us. And then, you know, I'm happy to talk about how we QA things, but that actually leads me into the third and final point before I flip it over to Neil is where we're producing agentic workflows, right? And I know you and Kieran have talked about this, but for us it's like, okay, SEO is a very thankless job. I don't want to spend time updating content. I don't want to spend time looking for internal links. I don't want to spend time doing a lot of this stuff at the end of the day. And so what we've created is, okay, we think about all the jobs to be done for SEO. You have to update your content. You have to index your content quickly. Okay. So we create all these agents that are able to do the work for us now. And I've asked my developers, I was like, how much time are we saving from this? Like 10x more time and we're way more efficient, right? And so it's like people talk about agents, but how do you actually implement it? You think about the jobs to be done first, then you map out the workflow for it, and then you can figure out how you wanna daisy chain them together. Okay.

Can you give us like a specific example of an agent, what it does, like what it was built on? I, I just want everybody watching to understand like what is possible and what something that they could go and do on their side.

Yeah, let me, um, let me do a little screen share here. So, okay, this should, this should work out. So let me share my screen. All right.

Does—

can you see this?

Yeah.

And if I make it bigger, can you see this?

Yeah.

Okay, cool.

All right.

So here's an AI consolidation agent. So in SEO, we know that if you update your content, you're going to gain more traffic. If you delete content, you help free up crawl budget. You might gain more content from it, but also you have to consolidate content that might be very similar, right? You guys do a very good job of this at HubSpot. And so, okay, what if we created an AI consolidation agent where what we do in the very beginning is we simply look at Google Search Console and then we fetch Google Analytics metrics for URLs. And then as we continue down the workflow over here, we'll look for, you know, the closest vector matches, right? What are the closest types of content that should probably be combined? And then we'll filter Google Search Console data and then we'll fetch the URL. And then at the very end of it, we'll provide all the URLs that we think should be consolidated. And then my belief right now, at the very least for the next 2 to 3 years, is you probably need a human in the loop for this stuff. can't just let it roll on its own. And then that's what we basically do. So this is one example.

So is that all a series of agentic prompts for like the OpenAI API? Or like, tell us about the technology behind it? Because everybody's gonna drop comments on YouTube and be like, all right, well, I like this. How do I go and like actually do that?

Yeah, it's exactly that. So it's leveraging OpenAI's API. I believe we're using some, some newer products as well. I think we just started using something else. Our devs started using something else maybe like 2 weeks ago or so. I don't know what it is right now, so unfortunately all I can say is OpenAI's API.

Yeah. But, but basically what you're saying is you can build basically a series of small web apps on the OpenAI API to basically automate different parts of SEO. In this example, it's like consolidating similar content to basically have fewer, more authoritative pages on your website instead of having 10 different pages that are kind of all saying the same thing.

Yep.

And diluting your, your search, uh, power.

Yeah. And Kip, imagine this, right? Like you guys You guys are producing a lot of podcast content. So what we're doing for a lot of our pages to keep them updated is we're linking to the most recent podcast episode or the most recent YouTube video that's related to that. So it keeps refreshing the page. And then we also are keeping— if we have an internal linking agent that's constantly working, then we're probably— we're trying to put in the most relevant or the most recent internal links as well, just to help give love to those pages. So a lot of this stuff, I think humans, we're very limited, but This stuff to me seems like it should be done programmatically.

Totally agree with that.

All right.

So I wanted to take a second to tell you about the Agency Owners Association. So this is an agency peer group ran by my podcast co-host Neil Patel and myself, Eric Siu. And these were 9 and 8-figure agency owners respectively. And what I will say is, at the end of the day, you want to learn from people that have been there, done that. If you want to grow your agency to 7 figures, 8 figures and beyond, this is the group for you. And we bring in other agency owners to speak as well. And it's a great peer group where you can help each other out and you're going to learn how to acquire more clients. You're going to learn how to take yourself out of the day-to-day and to build the business of your dreams. And the cool thing, till November 30th, we are holding a special price and you can apply at marketingschool.io/agency. And the great thing about it as well, if you try to get into group, there's a 7-day free trial, right? So once November 30th hits, we're going to increase the price by $200 per month. So right now, You're getting it at the lowest price possible. So go to marketingschool.io/agency to learn more and we'll see you inside.

The way we use AI, you can end up using it in many different ways, but when we look at results for customers, we really see 2 avenues producing the most results. The first being research, because if you look at the content that Google tends to rank, it's the content that's the most thorough. So the more thorough you are, typically the better you're going to end up ranking, assuming you answer everyone's objections that they have around a specific topic. If not, they click the back button, and sends a signal to Google. They keep searching, looking at the next webpage and the next webpage until they get their answer. So a great example of this is we work with a company called University of Technical Institute. We were able to grow their search traffic by more than 200% by using AI. So they have campuses all around the United States. We used AI to do research for all the locations and cities and all the information about those cities that we know interest students that go there around those campuses. So then when we sent that, and through AI you can also have it organized in a specific format and a layout that you want, you then have your team, or we had our team go and create content based on all the research that OpenAI or ChatGPT ended up providing. And then from there we ended up going and modifying all the local pages and that saw massive boosts in traffic. The second way is if you do a Google search for anything, you see Wikipedia. The reason you see Wikipedia ranking for most things is their content is fresh and up to date. Most people have this wrong that are SEOs though. They believe if you just quote unquote update a page and modify a line or two or add a video, add a link, it actually helps. We actually see if you update a page too frequently with very little value that you're adding, it does not help your rankings. We're actually seeing it hurt rankings. But if you go and you update a page and make it more thorough, answer objections, improve the user experience, right? All the things that are causing people issues, we're seeing rankings go up. But the key here is you can go into Search Console, you can use the agent for this. You can see what pages are decreasing in traffic year over year or quarter over quarter, however you want to end up basing it. From there, Those are typically the pages that need updating. You can then end up running a search for all the keywords that have dropped in rankings, see who's ranking above you, what are they doing that your page is not doing, and then you'll have better ideas of what you need to do to update the content and make it better. You can have AI help you write some of that content, but you need a human in the loop. And the big thing that Eric and I believe in is you can't eliminate humans. You know, maybe in the future, and that's a real maybe, And the reason being is with AI, there's a lot of edge cases. For example, if you're HubSpot and your terms of service page gets little to no traffic, you can't have an AI just say, hey, this page is getting little to no traffic, where your SEO, you should only have pages that are getting traffic, so let's eliminate it. You're going to be in legal trouble, right? And there's a lot of edge cases like that in all forms of marketing. For example, in paid marketing, you know better than anyone else, If you're running a new paid campaign, whether it's Google or Facebook, and your AI's analyzing your data in real time, in most cases, during the first day, first week, first month, that campaign is going to be negative. And AI would tell you—

Yep.

Hey, you're losing money on this campaign. Shift the budget to these campaigns that are more profitable. Well, it takes time for, you know, Google system to learn. It takes time for, you know, Facebook system to learn and for you to fine-tune and tweak the campaigns. And then as you're scaling things up, it's not like they perform the best instantly, right? But all these things will give a computer a signal that, hey, eliminate, stop, pause. I'm not saying the computer AI can't get smarter in the long run. It's just you really need to have a human in the loop. And the best feedback I got was from a Google engineer who actually works on their AI piece or their AI algorithm or AI overviews, whatever you want to end up calling it. Another example with AI is If you're running paid campaigns, let's say it's with Facebook or Google, as marketers, we know that you can't just say, hey, your campaigns are good or bad in the first day or week or month, because it takes time to scale up and it takes time to learn. Like Google's Performance Max isn't going to be optimized and ready to go with your campaigns from day one. But if you just relied on the AI, it may end up telling you, Hey, you should shut this off and you should stop tomorrow and reallocate the budget. But as marketers, Eric and I know that we need to be patient and give it time. Look, the big thing with AI is you really need a human in the loop. I was talking to a Google engineer and this engineer works on Google's algorithms specifically with things like the AI overviews. And their biggest problem, and they've been fighting this problem since Google ended up coming out back in the day, is there's misinformation on the web. But with AI, it's pulling from everywhere on the web and every day there's new misinformation. How do they know what to eliminate and what not to when it's giving you answers? These are reasons why they cite sources and why ChatGPT is doing this and other players are copying because it's too hard for them to eliminate misinformation. If this was possible, Google would've done this years ago. And this is why AI is never going to be perfect.

Yep.

I think maybe we can work towards talking about the old stuff that still works really well today, even though we're talking about the future of marketing, right? So what's old kind of becomes new, and then I think we can end it with talking about the new stuff that we're excited about. So maybe I'll kick it off with some of the old stuff that Neil and I have been doing here. So what we really like, so we really like events. Neil and I are both going to Germany to speak at a conference called This Is Marketing, you know, leaving this Saturday, or Neil, you're leaving Sunday, I guess. But within that event, what are we doing? So we're speaking at the event. The event organizers were gracious enough to host a private dinner for us as well with with other entrepreneurs in the area. That's super high value as well.

Yeah, so one of the big things that we do with events is if you're going to an event, there's usually a lot of people that are there from corporations of all sizes. So one thing that we tend to do is we tend to invite people for either private events or we tend to invite people for our own quote unquote session that may be hosted in a room at the event or a separate place. And we'll give a talk and a speech. And we found that to generate a lot of revenue. And separately, we'll do our own events, excluding, you know, events like HubSpot or whatever it may be where we're running our own thing. We're reaching out to our network, to people on LinkedIn, inviting them. And these are ideal customers, and they're in roles where they would be decision makers. We found this to help us generate quite a bit of ROI over time because you're closing quite a few deals from it. We don't close deals right away. It takes us a year to 2 years to see ROI from any events that we host on our own.

I want to take a second to tell you about my podcast co-host's agency. So NP Digital, so Neil Patel Digital. What they do is they do a whole host of marketing services and they are global. They're worldwide. They have SMB services as well. They have mid-market to enterprise services as well. They cover the entire gamut. So you can just go to mpdigital.com to learn more about it. And now back to the episode. And I think the other stuff too that we still— what's old is still new today is, you know, Reddit has been around for a while. Quora has been around for a while, but communities have gotten a bump and Google's kind of adjusted a little bit. But still, you cannot doubt that Reddit's added about 700 million in monthly organic visitors. And that's a lot, right? So in a world where we're going to see a lot more AI content production, that this is why Quora and Reddit are taking off. But, you know, there's also other communities out there that you can make yourself. You can use apps out there called Skool, S-K-O-O-L, or Circle would be another one. And then for Circle, for example, I know you can put it on a subdomain so you can get some SEO benefit. And so I think communities to me are interesting because not only are you going to gain SEO traffic from it, but you're also going to build a retention mechanism in for your business, right? Because people ultimately, they might buy from you, but they stay for the community, right? And so community could be in person, like the events that we just talked about, or it could be online too, or ideally, You're doing online first and then you're bringing them to in-person, right? So Neil and I, we have this agency group called the Agency Owners Association, and there's an online component, but we're adding an in-person component because Neil and I basically live half a mile away. And when we're both home, it makes sense to have a workshop that we can do in person, right? So Sam Altman himself has even said that he believes that in-person communities will be even more valuable in the next 5 to 10 years or so because The fact of the matter is, when you look at charts, humans are getting lonelier and lonelier. They're not getting out there, not talking to people. We're kind of not doing well with our social skills. And so I'm excited and I'm pretty bullish about that. Anything else missing there, Neil, before we talk about other stuff, future? Yeah.

When you look at community, I think the biggest issue that we're facing right now is with technology and AI, it's becoming less of a person-to-person world in which people are interacting online more with robots, technology agents, whatever you want to call them versus other humans. And people are actually craving those in-person interactions. When Eric and I started in marketing, you know, 10, 20 years ago, I don't know what timeframe, it was more than 10 years ago. In-person events were really huge. People loved them. Social wasn't as big as it is today. And then over time, people started going to leveraging a lot of these social platforms, a lot of events started closing down and weren't doing well. New ones of course popped up. But now as Eric and I go to events, we go to a lot of the same ones because we live so close to each other. What we're finding is people are loving these in-person events.

They're craving them. They're wanting more of them.

And as technology keeps improving and evolving, it seems like people want even more and more of a localized community. But not just a community of just meeting up with other random people. It has to be like-minded people that are trying to solve similar problems or have similar goals and that can help each other.

Thank you.