SEO Week: The Inflection Point for AI Marketing
===

[00:00:00] 

Wil Reynolds: What's up everybody? Will Reynolds here. Thanks for having me back to SEO week and how many of you in the room know who these two guys are? Do you know who they are now? Yes, it is Black Sheep and the song is, you know that you can get with this or you can get with that. The Choice is yours. That's pretty much the theme of this entire freaking presentation.

I think this is gonna be one of those presentations that at the end of it, you have to make your own decision if you're gonna get with this or if you're gonna get with that. So let's start off talking about bad marketing. How many of us have gotten a message that looks like this? I thought I was the only one until one day somebody put on Twitter like I'm supposed of getting these messages.

And I was like, I thought it was just me and thought somebody was just trying to scam me. And it's like, no, they're trying to scam all of us. Anyway. Somewhere. Somebody's like, I'm an SMS marketer, and you're like, that's not fucking marketing. Right? We all know that. That's not marketing. Now when we do this is that marketing.

For a long [00:01:00] time, I think we called that marketing. And then if we were to think about the work that we've been incentivized for doing, and I'm so grateful that we're in this AI moment because it gives us an inflection point to choose if we want to be this or that. Imagine a world where I get off stage in 45 minutes and then the speaker after me repeats 90% of what I just said pretty much verbatim.

Would that be a good thing? I think you're thinking, no, that would fucking suck. Right? It would fucking suck. Because no one wants to hear somebody repeat the same thing that the person before them said, Ooh. So when we stop and think about SEO strategy, how much of our work has been looking at what other people have said and just barely tweaking our page just a little bit to beat them.

No original thinking. No original thoughts. Just a whole bunch of copycat shit. [00:02:00] So I'm not scared of none of you motherfuckers. I ain't scared of you motherfuckers. I'm gonna tell all of you guys today exactly what I am thinking. 'cause I am not scared of none of y'all. Shout out to Bernie Mack. I am not afraid to tell us 'cause I'm part of this, that it's time for us to level our game up and right now is the inflection moment for us to do it or not do it.

I've been saying this for a while. The job of marketing are these three things to be seen, then to be believed, and then to be chosen. Or as Kobe said, when the reporter said, Hey, Kobe, you've, you're up to nothing. Why aren't you happy? He said, job's not finished. The job is to actually win the championship.

The job is not to go up two and oh. But yet most of us are pulling up just a little bit early. Thank God for Kobe's focus. I wanna take it back to 1900. The Michelin company, they make tires. They decided to make a [00:03:00] guide and give it out to chauffeur so they would drive all over the French countryside to wear down those tires.

The more you drive, the more you wear down your tires, the more you wear down your tires. You gotta think, well, who's the right company to replace these tires? And the brand has been in your mind so much that maybe just, maybe they might build a company that would last till today. And now when we think about Michelin Star restaurants, we're here in New York.

You know, you gotta think this is a, this is a over a hundred year marketing experiment. Michelin started off saying, Hey, we're just gonna put these things out there and have people see us. And then eventually the stars became a thing that people believed if something got a Michelin star, that it was gonna be worth.

Eating at, and then after a while, the whole world knew. And now if you're in New York and you wanna book a Michelin Star restaurant, good freaking luck. You're probably waiting three weeks. I believe those restaurants have been chosen for sure. So then I was on another path and I'm looking at RAMP [00:04:00] and I'm like, man, they have taken their internal data and have made some very useful.

Um, information. This guy Aura is just, I love his Substack. It's amazing because he's an economist putting out real high quality content using ramps data. Thank you, aura. And then one day I saw their CEO, I forget his name on Cheeky Pint, um, which is a very popular podcast. I think that's. The CEO of Stripe, I try to not hang out or watch too much of this success porn bro shit.

But somewhere I, I know these guys are popular in the bro circles and I saw the, I saw him talk about the product and I was like, wow, there are some really good features in there. And then I went to go talk to my CFO about maybe changing over to ramp, but before I get there, ramp at the same time was doing something that I saw too.

They saw what Michelin did with restaurants and were like. We're gonna do restaurant content too. And I looked and I said, top five restaurants in Minnesota. [00:05:00] Minnesota's fucking big dog. You might have to drive seven hours to get from one part of Minnesota to the other. I don't think people search that way.

Right. So then I went, oh no, we've got zombie content. It's that soulless, lifeless, ignorable, repeatable crap. So then ramp pretty much said, we're gonna create all this business dining pages, we're gonna use our internal data, we're gonna make 'em all in one day. How many of you, after knowing how that content was built and that it could recommend things for you in an LLM, wanna go there?

Over going to the place that wins the Michelin star. Ah, well, why not? I think we all know why not. So I'm gonna ask you something that I, you know, I just believe in my soul. If you came to this conference today and woke up early to learn how to be deeply mediocre, you're in the wrong freaking place. I think [00:06:00] none of us woke up today saying, you know what?

I wanna be fucking super mediocre today. But yet, a lot of us. Are going to hear a lot of things that are gonna lead us down a path of doing mediocre work. And I want to cut that from the jump. Something that I believe, and I'll have to include the person who who sent this to me, but. It's funny that the more AI is super helpful.

Yes, it helps us to create all kinds of awesome stuff at scale, but the more you lean on it, I think the less qualified you become to judge what it puts out because you just get used to it and you just let it go, and you let it go and you've got a workflow and it does this and it does that, and eventually you're barely even looking at the shit that you built.

And now you're building out content for states that it takes 20 hours to drive across. I don't want the best dining restaurants in California. Nobody does. It's a huge fucking state dog. Use your brain, but this is what I mean. You turn your brain off and then you start producing slop. Oh, and to bring it back to ramp.

This is what happened. [00:07:00] I was like, Hey, man, this guy said this thing. It's super cool. My CFO is like, their salespeople turned me off. They have blanketed emails that are the same email over and over again. I don't like 'em. They're pushy, right? Within four minutes. All that goodwill that the Economist built, that the CEO being on that podcast built was killed in an instant.

'cause when I talked about all the things that I saw that I liked, my CFO was like, yeah, no, their people are pushy. So for so long, guys, this is the inflection point that I was hoping we would get to for so long, bigger always won in our space. If you had more domain authority, it was easier for you to win as an SEO.

The real job was how do I get things done in my company? Not, how do I come up with a strategy? You're like, no, I got a big domain. I just need to put the content on the page and convince somebody to let me do it. So for years, I've been showing Nudie jeans against Banana Republic, and I'm going to use it again because they think it's the best [00:08:00] way to show you that one brand leveraged their domain authority.

That would be Banana Republic to win. The other one became an authority in their domain. Those are two different things, guys. So what happened is, is Banana Republic said, Hey, if we wanna rank for ethical Genes, if we just put some content at the bottom of our page, we're gonna show up in the top of the results.

And they did. And they beat out people who actually were about this life and became an authority on the domain with something that was written by an AI at the bottom of a page. Yeah. It was that simple when all we had to do was get the right domain authority. But AI forces us in many ways to become an authority in our domain.

So when I typed in ethical Genes companies, banana Republic has not showed up on one of those queries in years. Now, Lily Ray said it best, A loophole isn't a strategy, and many of us who have become strategists are actually [00:09:00] loopholes. We find these little things and we exploit them for the short term, but no one ever wants to talk strategy with us because we're literally automating so much stuff that we're building content that's like, that's restaurant in California can't bring that person into a strategic meeting, can you?

So my focus as it has always been, will be on what would happen if we built things that won for actual humans. Alright, my listicle people bring it the fuck on. Let's go. So you have to decide if you're gonna get with this or that. Are you in the visibility game or the credibility game? Today is your day.

Now you might be in the, I'm just trying to keep my job because my boss wants me to do this shit game. Great. Then you're in the visibility game. As long as you get visibility, you're fine. And if it destroys the company's credibility, maybe there'll be another company to hire you in a few years. I play the credibility game.

Let me show you how bad these listicles have gotten. I recently did a search for time boxing app, saw that one of the recommendations was Reddit. So I went, okay. I had been noticing that [00:10:00] Clickup had built a listicle listing themselves at the top on the best time boxing apps. Great. Alright. Clickup, you're the number one one and you listed it on your own site.

Awesome. But if I got all those Reddit threads, why don't I just scrape all of them and find out what brands were also mentioned? Huh? Isn't it interesting that of all these threads about time boxing it. Ups. No one ever said fucking clickup when humans had to talk to other humans and give recommendations, but that never stops.

A freaking overly AI reliant workflow reliant person from kicking out tons and tons and tons of low quality, shitty content that will never win for humans. Look at all the brands that weren't even on click ups list. It's interesting that Clickup was not mentioned at all, but they're the number one on their list.

But Flo Savvy, who was mentioned by 10 different Redditors, oh, we're not gonna mention them. Or Ample Note, or Morgan or Task Kitchen. Come on, [00:11:00] dog. And the thing that's also working against us is the people who buy the surface from us are used to this being like SEO. Clut, your pearls all you want. I'm fucking saying it.

GEO is not a performance channel. I'm sorry. It is not. And I'm gonna make a case for that. One of the things that, Kirk Williams said is that when you start to think you can track everything, you kind of gravitate to the easy KPIs. And I think the hard part is, is the reality is the KPIs around.

LLM optimization are all hard KPIs and we don't know what to fucking do with it 'cause we've all shut our brains off and I'm here to wake 'em up today because the reality is the way that decisions are made are mostly in places that you have a very hard time tracking full stop. So visibility is just an opportunity.

You don't think you won 'cause you showed up at the game. It's just the face off. And then you get shots on goal. And that's also more [00:12:00] opportunity to win. Isn't it interesting how athletes never get it fucking twisted as to what the actual job to be done is, and then that's how you win the Olympics against Canada.

Go girls. So what we need to start to do is overlaying our scene metrics, our AI sessions, our visibility metrics against something that holds us accountable. Seeing is like the face off. That doesn't mean shit. Believed is when you start saying like, okay, are people doing something? Am I getting shots on goal?

Am I scoring goals? You know, being believed is like scoring a goal. It doesn't mean that game's over, it doesn't mean you won, but you scored a goal. And then I think we've gotta overlay that with our pipeline. If your visibility or AI traffic is growing like this, but your pipeline is flat, then maybe you're showing up for the wrong shed.

Maybe when people see you all the time, they don't believe you, so therefore it's not having the same impact. The days of if I get higher [00:13:00] rankings, I will then get more business are over. You have to look at these things together because again, this is the full job of marketing. Now this is my problem with visibility.

My friends look at this report. Air byte tripled AI brand visibility in one week. Who would not raise their hands and say, why would I not pick the company that did that? Well, I'm gonna give you a few reasons why you might not, because the way they did it was to launch a bunch of pages on one day that were all the exact fucking, same with just small little changes, and I'm gonna show you later how to hold people accountable.

This case study, even though it says, Hey, we closed a $100,000 deal from ChatGPT. Okay, but was it from this shit? What human read this shit? And went, I'm hiring those motherfuckers, no one. [00:14:00] And if they did, how are they gonna be a client for long? So I'm here to say that if that is now my job, if what I have to do is build this to win for my clients, fuck this battle.

I don't wanna win. I'm Audi. I would put the locks on the door today and go do something different because that's mindless work. It's shit work. And we all know that's true. But the hard part is, and I get it, is the pressure of the people who buy geo services. They're so used to buying SEO where you've got monthly search volume, you've got a number of rankings.

You can tell how much visits you're getting, you can tell how much conversions you're getting, and then you can put an ROI to that. The problem is none of that works in geo. People will tell you that they have search volumes. I got stuff to show you that are gonna make you think, really, doc, how good could they really be?

And then we've got personalization, which I've ripped all my slides out about because I don't have time to go through all of them. But personalization alone changes the answers [00:15:00] so much. You get different answers. If you are searching for things about fixing your home and ask for a credit card someday than if the AI knows you're going on a trip to Japan soon.

The answers aren't static, and that's been beat to death. So I'm gonna let that go. But let's have some real talk, real talk to real humans. So we talk to real humans and we have them do tasks and we understand in different verticals, what are the things that they're actually willing to do? What do they do the most?

What do they do the least? So we did, um, a research study recently in banking. We've open sourced that It's public. You can read it all. But the way that we did it is we took 30 real people, real, real people, and what we did is we looked at them, do a bunch of different prompts. We ended up getting 114 prompts and it cost $800 to do it.

The tool that we used was outset.ai. It's an AI tool don't have time to get all the details, but pretty much. It's an AI that walks [00:16:00] the real people through the questions they want answered like a UX person would without actually being present to do it, which allowed me to drop the cost to make this possible for my clients.

I recommend you do something like that yourselves. Now you're gonna talk to real humans, and once you talk to real humans, you start seeing crazy shit like this that makes you question everything you've been doing before. We took the same task across 30 people and looked at the lowest, the min and max of the number of words in a prompt.

Literally one task, one person put in four words to complete the prompt. Another person put in 119 words to complete the exact same task. Ugh, what a tough job right now. Right? And we gotta have the freaking fortitude to sit with hard charging leaders asking us very tough questions and showing them like, yo, the listicle thing ain't gonna win long term.

This is what the humans are really doing. And then show them, how am I supposed to track what we're visible for? If all my prompts are freaking six word prompts, it might be okay. But you [00:17:00] have to talk to leaders, do the work and show them, Hey, here's some nuance alright, so let me share with you another thing related to our brand.

So these are real conversations, real people I know. You want a tool to put a URL in and have it do it all for you? Or a prompt? I'm fucking sorry. We're not getting that today. 'cause everybody else behind me is probably gonna bring that. I'm trying to give you a frame to think about this presentation so that you can hold all of us as speakers accountable to helping you to not just wake up and be mediocre.

This is a direct quote from a lead, to be honest. I like learning from people in the weeds. Where do you think that opinion came from about the SR brand? I don't know. 20 years of blogging, watching us try and experiment and do stuff. That's where that came from. Lead, right? This is a client choosing us money, not visibility.[00:18:00] 

The other thing is, he said this, I'm a tinkerer. I'm happy to show you that the agents that I built. So he's like, I wanna be with the nerds. I'm a little bit of a nerd. And I wanna show you what we built. Maybe we can build some things together. 'cause I see that you, I see you guys as builders. And then the last thing that he said when I was younger, I ran away and stayed in a shelter and I end up turning my life around shortly thereafter.

You see, I sleep out every year raising money for homeless youth in the Covenant House and many of you have probably donated to it, and I appreciate that there's a link on my resources where you can actually donate if this is valuable to you. That's 15 years of me freezing my freaking ass off in the cold.

And people seeing that that affected who they chose to help them with their project. That's a multimillion dollar project.

Streets are freaking watching. [00:19:00] Founder led content is winning and I've got some, some resources for you later on that, but it's not just founders, people are seeing. A bunch of people at Seer writing blogs, getting their blogs mentioned in publications and things like that. Because I am saying company, if you write shit that I think is worth all of you guys reading, I will push and promote it.

So it's not just founder-led content. I think it's also how do you bring others with you? We'll talk about that at the end because what I'm trying to articulate to you guys is people will always find reasons to not buy from people they don't like, and they will invent reasons to buy from the people that they do.

And you don't have to look any further than this motherfucker, right? Funny. All of a sudden when you see Kyle during a presentation, people in Europe are like, fuck you. We're not buying your shit. People will make their own reasons to not support people they don't like. I have a Tesla. It is better than every other car I've ever driven, and I will never fucking buy one [00:20:00] again.

Ever. So. If you think the job to be done is to fight over best GEO agency, I'm gonna let these people win and I'm gonna keep winning for humans because as this world pivots and changes and we don't know where it's gonna go, ain't nobody gonna pay me a hundred thousand dollars a year to be a listicle motherfucker.

I gotta be able to use my brain. I gotta think and I gotta create. Here's another client. My current firm is not keeping up with a EO. I wanna find a new company. What do I need to be thinking about? It gave him back like a thousand word answer to, to which he said, gimme 10 consultants who would fit the bill?

You see what I'm doing here guys? When a lead comes in, I'm not running and writing the case study saying, this is how fast we showed up. I'm saying, let me talk to you and can you send me your prompt so I can better understand how people make decisions? And it gave us these three brands and seven listicle people.

Right. That was the 10. 'cause the [00:21:00] listicles work, but then the AI nudged this. Hey, if you wanna avoid wasting time, here's the questions I would ask these people and pressure test on what I love about this prompt. See, when you look at just, am I visible? You only think yes. No. When you actually read the answers, you start being like, Ooh, maybe if this is a consistent recommendation on how to pressure test an agency.

I need to write that myself and write that content myself because I'm already seeing that they are leaning to saying these kinds of things, three outta 10 of the prompts. Here's how to pressure test. He didn't ask how to pressure test. The AI assumed that that would be valuable to him. So where can I find the assumptions the AI is making on my prompts that are consistent across prompts and say, am I writing content on how to pressure test your a EO agency?

I'm not. Are they recommending that to a lot of people? Probably that's the content we should build. And please don't build it using freaking ai. Use your [00:22:00] brain. Alright, so then he started taking the, the different answers and pacing them across different ones. And this was Claude and it compared one of the agencies to Seer shit.

Who else was worth considering and who he should proceed with? Caution. Ultimately, we're in a race with the last one. Do you see how now that I look at human beings do things, I start having to question what prompts am I tracking and are they the right ones? And Elisa put together this great framework.

Also, John, um, who's also linked to my resources. You know, he really started nailing us on this concept of brand canon and this idea of like, you gotta define your brand attributes and then see what is the likelihood of when you define your brand attributes, that certain prompts trigger the things that you want said about your brand.

That's the first thing to do, because then what you start off with is, I think the first step isn't how do I show up for GEO Agency? It's what happens when people ask questions about my brand, [00:23:00] especially if they're comparing 'em to other agencies and things like that. And then once I feel good about being about defending my turf, which I think is a year long process in and of itself, and I'll share my examples with you later, then you start looking at the accuracy, the attributes you want.

Then you start moving up to the offensive. Long tail non-brand and then maybe someday you go, I wanna rank for GEO Agency. I'll tell you, as a person that ranked for SEO agency, SEO consultant and SEO company, in the top three for two or three years, all it brought me was low quality traffic. The word enterprise actually matters a lot.

Search volume was low, so I got a lot of traffic for that visibility. Didn't get a lot of conversions. And actually, if anything, it was a negative because it tied up my sales team with leads that we could not ever work with. I probably should have just ranked for enterprise SEO, which would've been a thousand times easier.

I've made the mistake. Let them go, make the same mistake. So you really gotta go through an exercise where you extract your brand beliefs, why you exist, who you [00:24:00] are, and who you're built for. And then go check the different ais. The other thing I think you should do is run a prompt like this, like start taking your, you want, you want to be more human focused in your ai.

That's all we got left. The content can be produced. You wanna be more human focused. So what you wanna do is run prompts like these against your prompts, take all the prompts you got, and then take the calls that you're getting and then say, based on these five or 10 calls, how well do my prompts match what my actual customer's concerns are when they're talking to people on the phone?

Humans told you I was gonna be about humans. Now we've automated this. So what we have is we have, a tool that takes consistent challenges that we see every two weeks. It takes the challenges, gives us the quotes from those different prospects, and runs them and calls them emerging trends and persistent challenges.

Then what it does is it will actually give me specifics for each one of my divisions and gives me a feel for what are my client's problems. Now [00:25:00] I'm doing this. Automatically in real time, which then allows me to go write the content that's gonna show up for these problems because it's so long tail, I'm never gonna know.

So I need to take my calls, figure out the trends, and then from those trends, go write the content. That's really what people are putting into AI that I don't get to see anymore because I don't get keyword MSV anymore. And the dope thing that you should be doing, if you ever build something like this, it's taken us three or four or five or six months and it's a constant improvement.

But now also my sales team instantly gets any blog post that match what the client's concerns are. Anything around my thought leadership, I'm recording this right now. I will send a file to the team and now the team will put that into our, uh, our tools. So then if a client is ever asking a question that connect on a sales call that I'm not on, that connects to something I said.

This will show up for the team, and a lot of times I'll just pick up the phone, make a Loom video and say, Hey, I heard this Babababa, human to Human [00:26:00] Harpreet did an amazing job of coming up with how to own your branded surges and to get you started with some things. Gotta shout my boy out. He's another great person to follow.

If you're picking up what I'm putting down. So when we were doing our banking research, we realized, you know, people care about overdraft fees, right? And these are the kind of things that we also have to do. So many of us are like, how can I show up for best checking account? It's like, no. How about looking at your overdraft fees?

Because we know people care about that. It's a branded prompt. And for Wells Fargo, seven outta nine, they owned. That's a good thing. I guess. Now if they wanna be nine outta nine, that's up to them, but seven outta nine, they owned. If you do the same thing for Bank of America, they own none of them. That's the bigger problem.

But they'll say, Hey, we gotta rank for checking. Slow down. Once somebody already knows they need a checking account, a lot of them worry about fees and when they ask about fees, you don't have any of the answers for your own brand. So now you've done all the work [00:27:00] for them to know your brand. They got down to the end, they wanna know your overdraft fees, and they get this.

That my friends is a problem that should be fixed first before you go after some unbranded keywords. So to really help you do this, I built a tool. It's on GitHub that you can run your branded prompts, run a branded prompt head to head. This is what real humans do. They compare companies to other companies and stuff, and when you do that, it'll give you a bunch of check boxes for the persona so you can check off whatever you want.

And then it'll give you a spider graph and you can do it for, um, web versus training, which to me, if I'm outperforming on web versus training data in the areas that I care about, that's a good thing. 'cause it means hopefully that data will get into the training data and make me a, um, a more resilient answer.

But then you also can get the different answers down here and you can do what you want with those answers. But I wanted to try to give you all something that you can install and go do to get started on this process. Because the last thing we want you to do is to stay in the current state [00:28:00] of the build trap because when you're in the build trap, it leads to this, look at this.

Every one of us in an agency here, search Bloom as build a page about them versus us. And for a lot of the answers for your brand, if you're not checking, are being owned by Search Bloom who build a low quality AI generated automated way to take control of your brand. And I'll tell you that it works and I'll show you how it works.

Ramps, it's got all these pages. Guess who is excited to click on this content? Guess what? Human's gonna share it with another human. Guess who's gonna pick up a phone call and go, you know what, man? Your top restaurants in Louisiana Page was awesome. Only this motherfucker. That's it. Nobody else wants that shit.

So why are we producing it again? Harpreet. You guys gotta follow this guy. He's dropping science. He's looking at his client questions and saying, wait, the answers are saying that we're slow, that we've got long implementation times, that we need [00:29:00] a lot of consultant hours. And then he's literally sharing with all of us, all the things that he did to turn those answers around for his brand Work on that first, because if you've got trust as your KPI.

Somewhere in your KPIs you can do big things like Michelin, no trust KPI. You're fucking Madoff. Now. This is what it also kinda looks and feels like. My friends, both of these are chicken. Now if you come over and I serve you, the one on the left versus the one on the right, how you going to feel about me?

Exactly. A lot of us are making chicken nuggets, but think we're freaking chefs. No, you're a high school kid dropping low quality chicken in a hot grease. You do not make high quality chicken today. I think a lot of people are gonna show you how to do things at scale, and I'm telling you, you making chicken nuggets dog.

And right now it's good until somebody tastes some real fucking chicken. [00:30:00] Alright, real chicken. That's the job of marketing. Want you guys to remember both of those things are chicken, but we believe something about both of them, don't we? One of 'em is from a Michelin restaurant, one of 'em is from a McDonald's.

So now let's start to talk more about, I'm showing you the negative. Let's move to positive, right? Let's do some human stuff. Some more. Good marketing. How do we do this? Well, first thing I'm gonna remind you, people are already decreasing their trust in AI answers. There is a distrust right now about them.

If you don't trust AI to answer your question, how else are you gonna get your question answered? Are you just gonna give up on it? No. You're probably gonna talk to a freaking human. And the more people talk to humans, the more we have to understand them. So if you can start to understand and invest time and understanding why people buy what they buy all that time, listening to people, all that time of me asking for their prompts, that is me trying to understand why people buy, which I can apply anywhere in marketing once the listicle stop working, what you're gonna move on to next [00:31:00] to me.

This is the best marketing ever. When you choose to do better marketing, you start to know people. When you know people, you can come up with ads like this, which to me is fucking amazing, one of the best ads I've ever seen, ever. I want you all to do the least scalable thing. Once in a while, ask yourself what percentage of my work is ultimately not scalable at all?

Me recording this video. And as Kara said, it's one of my favorites. She's the coach of, uh, she's coach at Duke. She said, it never gets easier. The journey never gets easier. Right now you're thinking, oh my God. Well this sounds hard. Yeah, it does. It's hard. Stop asking for it to get easier. Turn yourself into a person that handles hard shit better, that you can control.

We can't control when these models launch. We can't control what these answers are, but we can control turning ourselves into somebody that handles hardship better. [00:32:00] And what does that look like? Will? Well, it looks like this brave, literally posted this and answered every freaking person. Every person that had a question, Hey, what's the feature that you want?

This is what we've got. This is what we don't have. And it wasn't like a sales pitch. It was just like, yo, we're gonna talk to real people that are, that are installing browser and have multiple browsers. This is what it looks like. This doesn't scale. Although you might be able to use an AI for this, it doesn't easily scale.

And I really believe there was probably a human there that gave me a boost on their brand. Alright, for my MOP fans. You know the song Anti Up is about robbing people. Right? Good. Because a lot of people are just like Anti Up. I love that song. It's like that whole song is about robbing people. So this is what you're gonna do.

You're gonna tell all these people to prove it to you. I want you to stick 'em up and be like, nah, I heard about your visibility shit. What about the impact on the business? Shit. So when people start talking about their visibility. [00:33:00] This is how you find their visibility. Run some site colon searches. What are they doing to be, get all that visibility and then tell them to open up Google Analytics.

I want you to start telling people who are selling you on how great they are at AI search and how great their visibility is, and be like, can you open Google? I'd love to see you open GA and show me some of the impact. And is it getting any referrals from LinkedIn? No. Well, why not? 'cause don't nobody wants to post a listicle to their people that they're trying to gain trust with. Funny how we'll produce content for a bot that we don't gain trust with, but when it comes to humans, we're like, well, I would never post that. Then what are you doing making chicken nuggets?

And then ask 'em to show you their pipeline and explain it to you. That's mine. I got no problem going into GA and showing you the work that we do and how it's ultimately performed for my business. And I got no problem showing anybody my pipeline to show whether or not that's working. 'cause I'm telling you, the world's best listicle never wins on social ever.[00:34:00] 

But what's crazy is people have been doing listicles for years. Best places to work most, most place for this. Philly, that Inc. 1000 Inc. 5,000. We've been on all these, they're just lists. They're just lists, but they put brand on the list so it doesn't have the same stench as like you putting yourself at the top of your own list on your own website.

Can you imagine if we did the best places to work and put ourselves at the top? How many of you would take that job for how many of you would it be slightly shady for a company to be like, we won best places to work, and who gave you that award? Oh, we did. But then you leave here today and go, I'm gonna do just that.

Come on guys. Come on man. It's time for us to stand for something. I have decided, hey, I'm just not gonna play that game. I have decided that I want my company to be seen as a curator because we're now living in a world of infinite content. Infinite content, which is [00:35:00] all you are gonna get a bunch of ideas and you're gonna go produce after this.

Great. Thanks for adding the infinite content. I just wanna win here y'all where humans are talking to other humans. And I'm gonna show you my numbers. So my number of people signing up for my newsletter, growing number of people coming in, direct, growing people that know my brand, people that are sharing links with other people that I can't track their referring source, that's direct growing, referral traffic, growing sites, linking to my site, and real humans probably clicking on those links and coming to my website.

Growing some of the most important publishers in our space. We went from six page views from search engine land in 2023 to 4,000 page views. In 2025. We created as a company content that editors. Would be like, this is good enough to put in [00:36:00] my, in my newsletter. And then they kept doing it and kept doing it and kept doing it.

That's a good human signal. And guess what? People love human signals. Oh great. Why? Because they're freaking curators. Search engine land is a curator of a world of infinite content. That's why it's important to win there. And I showed you my pipeline already, so you know, it seems to be working. The trust channels are so much more efficient, guys.

I'm telling you when you win in Slack, can't track it, but it's better. It comes in direct teams, you can track better. The problem is, is they're really hard to track, so therefore we like the easy to track shit. So we don't look at this. Well, let me give you a little bit of dose of inspiration. Direct converts at one and a half times more than organic social, five times better than organic, and referral two times better than organic.

Direct social and referral are all more human than ai. So an [00:37:00] organic search. You know, when I think when I wake up in the morning, I'm definitely not thinking, how can I be really mediocre today? I'm thinking, how do I earn the right to be in somebody's custom instructions to the point where they will put my name in with a bunch of other people that I recommend to get great answers about GEO and ai?

That's the ultimate win. And if I can get in a couple of those and I think I'm on my path, I think that's the real win. 'cause what somebody's saying is, I trust you enough to have you influence my answers when I talk to AI about GEO. Anybody in here, um, doing that for one of the listicle companies. When you got that listicle, did you say, man, I really wanna follow the person that built this.

I'll tell you who it is. It's an intern with ai, so I also want you to interview your SMEs. I'm telling you guys, this is a massive unlock. We have been able to, like we did here, talk to people that have been doing [00:38:00] their jobs for 20 plus years. We've done this for so many clients, especially the ones that are in technical spaces, and they go, this actually felt really good.

It's an AI interviewing them. See how, I see how we're using the AI through outset to interview real people? And now we're using AI tools that we built internally to interview real people that know their jobs. What's really great about it is this more impressions, more sessions, more keywords, more citations, right?

This was an early effort for us. Thank God my client, let me share this with you guys. Um, but what we did is we interviewed them for 10 minutes. That's it, a 10 minute interview by an ai, and then the humans took the answers to those questions, turned it into high quality content because it was the stuff that somebody that's been doing this for 25 years said.

I'm sorry. I know the AI is good. It's rarely as good as somebody that's been doing the job for 25 years. So why don't you just go interview them. Why don't you use AI to go interview them? And I'm gonna give you a [00:39:00] start because this is how we did it. We took old content on our website. I had a developer build me an interface with, uh, you could probably do it yourself now with ai.

I had a developer build me something that takes the old content, runs deep research on it, and compares. What's the old thing we've been saying? What's happened in this industry that's changed? And then it would drop 10 questions into a Google Doc or a Google Sheet, Google Doc. And then I would walk around, open the Google Doc, hit the voice transcription, just walk around, answer, answer, answer, answer, answer, answer, answer, done.

Next thing you know, he had content that was easy to work with and approve. Because for too long our blocks have become the land of misfit toys where we just throw shit in there that we don't want to be on the major parts of the site. It's telling when the pages that you optimize, the content that they let you optimize is at the bottom.

And it's interesting when they usually give you areas that they really don't care that much about, and you're calling yourself a strategist. They're literally looking at you like you make chicken nuggets, dog. That's why I put your [00:40:00] stuff at the bottom and your stuff in this section that I don't want anybody to see, but I gotta placate the bots that's over.

Back when I did my Seer of sameness post, it was, it's over two years old. I did the analysis recently and I'm like, oh my God. Three of the top 10 results still don't mention AI at all on how to build an S-E-O-R-F-P 'cause it was built by people who did the checklist, did the thing they had to do, and never thought about it again.

Whereas people in this room, we think about this stuff all the time. So if you had one of those RFP pages, you would probably want to update it, right? Not these people that are just working on workflows and they're like, well, I'll build a workflow. Oh God. Or you could just write the content. This thing is two years old.

It's due for an update. I gotta update it. Hopefully one of my marketing people will punch me and be like, dude, you gotta update this post. We can help more people. Alright, next thing I want you to do is review your reviews. Amanda TDA said it best. We gotta really watch [00:41:00] these third party content sites.

Also taking a step back here. No, we're gonna keep going. So a client, they became a client, left the name of a competitor in one of their RFP docs. And then I went, Ooh, let's put that in deep research and see what they would pick based on all the transcripts. And they said they'll pick them. Well, they didn't pick them.

They picked us. Woo. But in that process, I saw this and I'm like, wait, people, client reviews that complain about account management turnover, agency spotter. I'm like, who's ever been to Agency Spotter? I'm like, uh, I know what this is. After all these years of being in business, the AI is not smart enough to realize that if you were founded in 2002 and you only got one client review and it's negative, that's probably a good thing.

We've, there's a lot of clients we have not pissed off. I didn't realize that. So it started putting this in answers about my brand. So what do we do? Could we wait to get case studies approved by clients? [00:42:00] No. Our clients send us a lot of shout outs, but we've never done anything with them.

Well, in an AI world, we gotta go in the offense. So now we gotta do something with those quotes. So now we dump 'em all on our page. To give the AI something to fight back with. It was happening in Slack. We just never made it public. Now we are, we're also asking clients to review us once in a while, and I hate that.

It's so cheesy asking people, can you review? Don't you hate when you're at a restaurant and somebody's, like, before you've even done eating the meal, they're like, can you review on Google? And they come over to the little QR code. You're like, uh, nobody's like sitting. You know what? Nobody's sitting there waiting for that.

Like, I can't wait for them to come over and ask me to do a review on Google in front of them so they can actually see. I did it. But yet here I am asking for reviews once in a while from clients that are already happy. Now we did a blog post with this stuff in it. Brave What Company do you know that takes somebody saying, worst agency I've worked with and puts it on their own website, us.[00:43:00] 

That's how you earn trust. People are like, damn, everyone talk about that. Yeah. They're probably gonna be honest with me if my numbers aren't good. Yep. That's better than beating up an agency where they're trying to bullshit you into thinking that their numbers are good when they're not. Yep. Maybe I'll give Seer a chance.

Yep. But what we did is we used this as an experiment and we shared it again and it was like, how do LMS Lyer brand our brand? Misconceptions? We posted it when we talked about having a negative review. We posted it when Lauren Boyd, one of our recruiters was here. She gave. A long blog post about all these data points around our brand that were like, yo, this is what happens at Seer.

We had a really bad review one year from a bunch of team members, internally we opened that said no, last year sucked. We had to get better and here are 10 different things we did to get better. We allow people to write that shit to the blog, which then gives the AI something to battle back on on our branded searches, which I think are more important than anything when we changed our footer.[00:44:00] 

And it worked within 36 hours. We said 97% client retention rate. Boom, it showed up in the answers within 36 hours, winning. Guess what? Retention went down by 5% this year and we updated the footer. And you can go to my website right now and you can see it in the footer. That's how I think I'm earning trust.

People would be like, well, why would you, why would you put it? Why would you, why would you just leave it at 97% because it's not true anymore and we know it's 92 now we gotta get better. We also have shown how putting this stuff out there is literally, this is a horrible, visualization. So I'm gonna build the new one.

But we've even built, we've even shared how when we wrote articles about our own brand, how that affected the negative stuff showing up. You can see it, I mean, it's really hard to see here. I'll do a better version of it later. But now when you type in Seer Interactive account turnover, you get an AI overview.

It's using our employee retention number because I posted it. Also it's getting, um, third party endorsement [00:45:00] because Spark Toro talked about the fact that we are experimenting with this. Anybody here know Spark Toro, right? Anybody here know these listicle companies you've never heard of? Right? And then what we did is we did research to look at the overall industry to say, well, what's the norm?

And then where are we? We're actually lower than the norm. Sweet, good for us. And look at that. We own these answers about our brand. That's where you want to be. And when, no matter what study I look at. It looks more and more like people know brands before they do their Google searches. They know brands before they do their AI searches, and they rely more on their brands and their networks than ever.

And when I think of how do you build a brand, I've never seen a brand built off of listicles ever. Challenge me on that. Can you name one? Are you willing to sacrifice visibility for believability? Are you willing to say, you know what? Will I need to stop just being somebody's like, tap dance. Oh, visibility.

Visibility. And be like, what are we doing to be believed? 'cause that's where the money happens. [00:46:00] We gotta build better moats. Here are things that have never built moats. White text, white background link exchanges, bottom of bottom of page text and listicles or listicles, listicles. I disrespect listicles so much, I misspelled it.

These things don't build moats. Now we've got people out there like Ham. He's saying like, look, lead everything to AI and just build distribution. How do you build distribution

trust? When people choose to follow you, when people choose to invest their time with you, sign up for your newsletter. That is part of distribution. That's why I'm saying you've gotta write high quality stuff that people share with other people because that's how you're gonna build distribution.

Build an internal influencer program. As John said, a lot of people are like, what happens if I build an influencer program internally and they leave? What happens if they don't join you in the first place? [00:47:00] That's a bar. And then when you look at some of the content from, uh, this is the LinkedIn study that was done on founder, founder-led content that's a little self-serving.

They want you to make content. They want more traffic. I get it, but I didn't run the study. The study's the study. Right. Deals close faster when buyers feel like they know you. Deal sizes are larger. When prospects follow the director and the executives on LinkedIn, these things matter, and they're human driven.

Brand awareness increases performance marketing conversion rates according to track suite and inbound converts better. Of course, these are people that know you, so as I wrap up, I wanna leave you with this. Trust will not be lost because brands use ai. It's 'cause we just use it in ways that take away our trust.

We build content like do you wanna go to a restaurant in California? Said by zero people. But we've turned our brains off and just used this thing indiscriminately. And now it's just, [00:48:00] now it's hurting our brands. So there's a bunch of speakers are gonna come up after me. I want you to rob them. Rob them of the anonymity of showing visibility only to you.

Ask 'em. Ask me. Open up your Google Analytics. Open up your Bing webmaster tools. Show me that this stuff is actually leading to people being believed or chosen. Otherwise, you just got me being nothing more than a mediocre marketer doing mediocre marketing, and I think that AI gives us a chance to finally end that shit.

Thanks so much. Appreciate you guys having me here at SEO Week. Hopefully I'll be back next year. Peace.