210: AI Agents Are Here (And They Hate Your Website) with Jes Scholz

There is a new user on the internet. It doesn't have eyes, it reads code, and it has zero tolerance for bad UX.

In this episode, I sit down with Jes Scholz (SEO Futurist & Marketing Consultant) to discuss the "Great Correction" coming to our industry. We talk about why a decade of obsessing over short-term metrics has "corrupted" brand marketing, and why AI is finally forcing us to fix the fundamentals we’ve ignored for too long.

Jes explains why AI agents are abandoning websites with messy HTML, why "entity optimization" is the real key to future visibility, and why the popular tactic of spamming Reddit to influence LLMs is a ticking time bomb for your brand.

Topics Covered:

• Why short-term metrics broke modern marketing (and how to fix it).
• The "New User": How to optimize for AI agents that read code, not screens.
• Why "trash" code and interstitial popups are fatal for AI discovery.
• The "Reddit Spam" Rant: Why trying to hack AI will destroy your brand reputation.
• The #1 skill AI cannot replace in 2026.

Timestamps: 

00:00 - Intro 

00:02:07 - How Short-Term Metrics "Corrupted" Brand Marketing 

00:03:59 - Why Marketers Must Relearn to Work Without Perfect Metrics 

00:08:00 - AI Agents: Why They Fail on Bad UX & Messy HTML 

00:17:52 - Beyond the Website: Why Entity Optimization is Key for AI 

00:21:14 - Why Spamming Reddit to Influence AI Will Destroy Your Brand 

00:27:23 - Final Advice: The #1 Skill to Keep & The #1 Habit to Drop

About the Guest: Jes Scholz is a global digital strategist and SEO futurist. Formerly the International Digital Director for Ringier, she has led digital transformation across 140+ media and e-commerce brands in Europe, Africa, and Asia. She is now an independent consultant helping enterprises adapt to the AI era.

Connect with Jes:

• Website: jesscholz.com
• Newsletter: SEO Brief on Substack
• LinkedIn: Jes Scholz

Connect with Sani:

• LinkedIn: Slobodan (Sani) Manić
• Website: nohackspod.com
• Newsletter: No Hacks on Substack

---
If you enjoyed the episode, please share it with a friend!

Transcript

[00:00:00] Sani: Jess, welcome to the podcast. Uh, great to have you. You are a self-described SEO futurist. There was never a better time to be an SEO futurist in 2025, if you ask me. So what does that mean to you when you say SEO futurist?

[00:00:15] Jes: It means looking forward to the tactics that are going to make sure that your brand is still relevant in three years time, five years time, 10 years time. It's about looking for what's coming. Tomorrow rather than relying on the tactics that worked yesterday, which I find a lot of the industry is looking back to, someone else has done it.

[00:00:39] Jes: So now I'll start doing it, and if that's your approach, you're always gonna be one step behind the times. So I rather look forward and try the new things, then repeat what everybody else has already done.

[00:00:53] Sani: It feels like the very early days of SEO when you just had to kick the box and see what happens and, and are we doing the same cycle with AI mixing in with the search, or is this something completely different?

[00:01:08] Jes: I think the, the level of strategy is a little bit different because we are dealing with something more complex than what we were dealing with, with the SEO algorithms back when you could put. A keyword in a keyword tag, and then you immediately shot to number one of the SERPs. But the principles are the same.

[00:01:24] Jes: It's still, we don't have this confirmed concrete playbook and we're trying to figure out, well, what does this look like? But I find that whenever we're looking forward being anchored in what's worked in the past, and that's not the past of the last 10 years of SEO, that's the past of since. Mad Men invented marketing back in the 1930s.

[00:01:49] Jes: Um, there's this commonality, a common thread that never changes, and that is a strong brand and how you are representing that brand online and now how we're representing that brand with the minds of these machines. It's still got that core essence of marketing to it.

[00:02:07] Sani: Would you agree that brand was neglected over the last 10, maybe 15 years when meta ads could get you to, to. Your brands to thrive even so would you agree that branding was, I dunno, because of the distractions of ads and pay traffic was kind of neglected compared to the madman era and compared to where we're heading now.

[00:02:29] Jes: I wouldn't say neglected. I would say corrupted. It's something

[00:02:35] Sani: worse.

[00:02:36] Jes: well, previously brand is something you had to invest in and you had to to grow, and everybody understood that that took time and commitment and consistency. And so we would invest in brand and and be measuring it on a yearly basis or maybe a six month basis.

[00:02:54] Jes: And then when we've had this, this CPC and and short term metrics, and I can see every single clicker, at least I have the delusion. That I can see every single click. We've come into this very short term performance mindset, not only in the performance channels, but also in channels like organic and channels like email and channels like social, where everything was expected to drive to the destination of the website.

[00:03:25] Jes: Without that clear strategic understanding that the vast majority of your users aren't in market. And if I'm not in market, I don't really want to come to your website. That doesn't mean that you can ignore me and not reach me because everybody will eventually come into market. And if you hadn't had those consistent brand touchpoint, then you're gonna struggle to have that brand salience for that customer when they're ready to buy.

[00:03:54] Jes: And then you have to buy it. And this is the rise of CPC.

[00:03:59] Sani: So short term metrics corrupted the brand, a version of that, do you think getting the brands to think, uh, in terms of metrics that cannot be tracked as. Maybe not accurately, but as detailed in a, in a way that's as detailed as the CPC and the short term metrics. Do you think it will be difficult for brands moving forward to go back to that reality of we cannot track it anymore

[00:04:25] Jes: I think it's gonna be uncomfortable because there's many marketers who have never had that experience. Especially the, the younger ones have always had everything measurable, everything trackable. They're looking for every 24 hours. You know, I need my data from those 24 hours. And there are some things that you do need to measure in the short term.

[00:04:48] Jes: Don't get me wrong, like short term metrics can be valuable, but there needs to be a balance of, you know, the long and the short of it. Paper written, what, 15 years ago. Still highly relevant today of.

[00:05:00] Sani: Right.

[00:05:01] Jes: It is, yes, you have short term performance, you have your campaigning. You can measure that, that's good, but it needs to be balanced to this long term brand building market coverage, revenue generation metrics.

[00:05:16] Sani: What you said there that there's generations of marketers who have never tried it any other way, is something that hasn't crossed my mind, but that's going to be very exciting and very interesting to see that transition to we, we just. We don't know, and we have to be realistic with not knowing. I mean, we didn't know before because our meta metrics accurate is what they're telling us accurate.

[00:05:36] Sani: I mean, they did, there was just a, a news story yesterday that they're 10% of their ad revenue is the scam for scam ads, and they know it. They knowingly let that happen. So I think going from corrupted and broken metrics to not really having metrics is not a

[00:05:55] Jes: have metrics, but

[00:05:57] Sani: Not, not as detailed. Yep.

[00:05:59] Jes: it won't be handed to you on a silver platter of Google Analytics. You need to work for your metrics. You need to. Do the focus groups. You need to get the market data. You need to understand and have agreement, how you're measuring that and make sure that those measurements are meaningful.

[00:06:17] Jes: And we did get spoiled for a long time, and so it's easy to see how we slid into this focus on the short term because Google Analytics or Adobe back in the day did hand it to you and say, here is everything you want to know about your website. And that made us focus on the website. But a brand is so much more than a website.

[00:06:40] Jes: And this led to this separation of here is the SEO team working on the website or the CPC team driving traffic to the website. And here is the branding team that does everything else. And then SEOs, bemoaned, you know, why did they get million dollar budgets? And we get a, a couple of thousand and you're like.

[00:07:02] Jes: Well, because they're, they're working on like what's really going to maintain the market position and you are focused on these. Yeah. Short term performance goals, you get boxed into not living up to your full potential.

[00:07:19] Sani: It never really made complete sense, and I wonder if a hundred years from now when a marketing historian looks back and will they think of this as, what the hell were they thinking? Why were they operate? I think so too, but. It would be interesting to just look the same way. Uh, doctors prescribed cigarettes to children and that is now insane.

[00:07:41] Sani: Will, is this the cigarettes to children of, of our profession? I, I don't know, less harmful, hopefully, but it never really made sense. We just learned, uh, to be comfortable with something that doesn't make sense. I, I think that that's a 2000 5, 2 2 2025 or whatever the era is. Uh. You do a lot of work and a lot of research around AI agents, the new user of the internet, the non-human user of the internet, operating on behalf of a human and trying to either do some research, complete a goal on website.

[00:08:15] Sani: I do that as well. Uh, I'm very interested in this topic. So what is your number one advice to someone who, who knows nothing about this new breed of users? What, how do you prepare for them?

[00:08:28] Jes: The best way to prepare for them is by using them. So go into church pt, buy a pro account, and put in whatever it is you say, Hey, go and book me a hotel if you are. Hotel website, go and order me food. If you are a supermarket, like whatever it is your conversion goal is, tell the agent to go and do that and see how it operates.

[00:08:51] Jes: See if it chooses your website. If it does land on your website, where does it get stuck? 'cause I'm guaranteed it will get stuck. And then think about how can you remove those friction points. And it's not enough to do it once because it's very similar to an LLM. It's never going to do the exact same path twice.

[00:09:09] Jes: And so if you do this 4, 5, 10, however might many times on your website, you can find, okay, these are the priority pain points, priority friction points to be able to then address with better UX or address through other access methods like MCP.

[00:09:27] Sani: Uh. I don't think we'll have time to talk about MCP, but if there's the future of agents browsing the web that is bright to some extent, at least it, it, it's MCP and not sending them to the human UI or UI built for humans, because to me that is, that doesn't make a lot of sense. It, it's a human interpretation of machine readable code for humans that now they get to decode into machine readable code again.

[00:09:52] Sani: It's like trying to guess who's behind a face tune, filter it. It just doesn't work. It, it shouldn't be like this, but yes, MCP if you're listening to this, read upon MCP, uh, it, it, it's, it's a new protocol model, context protocol that's hopefully going to be widely accepted and, and, and part of the future.

[00:10:12] Sani: Now, when, tell me about the experiments you did. Do you have a search engine land article that you published about a month ago now? What do you try to do there and what were, what were your key findings with the agents? It was chat, GPT agent that you use, right.

[00:10:26] Jes: I use church PT agent. I think that's, at least in the short term until Gemini is opened up publicly, which it's not right now, you have to have like the super high paid plan to be able to use it, which is limits its growth. Um, so I was using Chet PT agent, um, and it was really looking at across a wide variety of sectors.

[00:10:48] Jes: So doing hundreds of prompts. Where do they get stuck? And I think a lot of that comes down to aspects we've been talking about from a usability perspective for a very long time, but have never really had the drive to take to make a priority. Everyone was like, yes, you know, we know we shouldn't put interstitial popups in the user's face.

[00:11:13] Jes: We know this is bad for users, but we make revenue, but we want users, subscribers. But you know, there was always a but. And now, yes, you can put that interstitial up. The agent is not going to fill it in. The agent is going to immediately abandon the site while a human would be like, oh, annoying. And maybe they close or maybe they abandon.

[00:11:33] Jes: An agent will not tolerate, uh, an interstitial unless it's a legal one, like a cookie banner. Um, and so these factors of user experience where previously people will be like, oh, but we can get the human to do this now. There is no human in that loop. The human is coming in at the point of conversion, so that's at login or to put in a credit card like at, at order, and everything up to that point, a human's not going to see.

[00:12:02] Jes: So I think this will actually make websites better because all of these things you did to manipulate the humans, a bot doesn't tolerate that because they have thousands of choices and they know it. They're not loyal to your website the same way a human would push through.

[00:12:22] Sani: every, I know it's wrong, but tactic that has been used in the past is just not going to work. I mean, even if, even if. The goal, we don't know. Number one, we don't know what the goal of the agent is because that's between the user, the human user, and the agent. It could be go specifically to that website and buy something.

[00:12:39] Sani: It could be research this product and buy it from wherever we, we will never be able to notice. So it's a discoverability and then conversion. Those are to separate things that the agent has to do. Even if it, it wants to buy from you specifically if it fails because your buttons are not labeled as buttons.

[00:12:59] Sani: You use a span tag with some cosmetic tailwind classes to make it look like a button. It may figure it out like three times out of 10 it might, but seven times out of 10 it will not. And, and that is the beautiful mess of having an LLM make decisions based on a messy website. So, uh. If you don't, if your website is not accessible, that's, I guess that's my point.

[00:13:22] Sani: And you're making it accessible because of bots, and you didn't because of humans. Shame on you, but glad you're finally doing it. That, that is the only way to look at it. So, um, some things that you found is that 46% of the times, and you did a hundred plus tests, uh, with chat, with, that's, I did 10 and it took 30 minutes to do 10 tests.

[00:13:43] Sani: They're slow. That's another thing. They're super slow. 46% of the time it went straight to the reading mode, which means there's nothing visual. There's no CSS, there's no JavaScript semantic. HTML is now a thing again.

[00:13:59] Jes: It should have never not been a thing. But yes, it becomes much more important. And I think what's very interesting, if you look at your website in a text-based browser, so I used links back in 1995. Of course it. We didn't have Chrome back then, right? It was, it was the browser. Um, and I'm using links again so I can see what these agency and the vast majority of these, these visits, when they're deciding will I stay on this website or not.

[00:14:27] Jes: And a lot of the time these. This header structure that, again, has been designed for SEO, not even designed for humans, these huge, verbose, they put all of the categories in there, so there's an internal link from every page to every page that's taking up thousands of rows of code. And the agent on a text-based browser starts at the top.

[00:14:49] Jes: And so if they're going through thousands and thousands and they're not finding what they're looking for, they do get to a point where they're like, I can't find anything relevant. Here, it's just an entire category tree. I will leave and so we, we start thinking about real usability and how that's reflected in the code.

[00:15:09] Jes: And looking at that through a text-based browser is an exceptional way of getting a solid understanding of what can I understand when I look at this website and how much trash is in this code.

[00:15:25] Sani: What. Wasn't HTML to text or HTML to context to content and SEO metric, quote unquote metric. A while ago, people cared about how much text or real content you have compared to HTML and it was always 10, 20, 30% at best, which is horrible. A lot of junk on the page, basically. Um. I don't know what to say about, first of all, the fact that we are both using chat, GPT, uh, agent to test is because it's there, it's accessible to most people.

[00:15:54] Sani: I don't think the future is chat, GPT agent, anyone will be able to build an agent and send it to do different tasks. If you, I mean, already with NA 10, with Lindy, with all those tools, you can build an agent and it can do whatever you want it to do. Google will have its own version, like you said, at some point.

[00:16:11] Sani: But for most people, this will be the entry point to using an agent to do something. This or, uh, the browser that, that the spyware browser they released, or perplexity comment or those dangerous tools that most people probably should not be using with their sensitive data. Even when you install, uh, Atlas, it tells you don't log in.

[00:16:30] Sani: And if any data, it is something we care about essentially. I, I just don't see a way how this can work. I'll be very honest with you. I just don't see a way how, based on how they fail, and based on the numbers that you shared and, and the test I did, and I'll share some details. I, I picked a skincare website and I told Chad J PT agent to summarize the same product 10 times with the same prompt, and asked for adjacent as response description, uh, price, five different fields.

[00:17:02] Sani: It was wildly inconsistent, incredibly wild. Four times. It said, this is a refillable packaging product, six times. He didn't mention that. I think that's important. One time it said it's vegan, nine times. It didn't say that the product is vegan. Nothing made sense. And then I asked, uh, I asked, uh, Chad GPTN five nano to summarize a long JSO that has fields that map to what I, not the same keys, but the same values to the last character.

[00:17:31] Sani: It was consistent every single time, which is why I think MCP and, and, and anything that has structured data rather than HTML and, and CSS and JavaScript is, is just going to be, uh, a better way to handle this in the future. I could talk about agents and agents failing all day because that, that's really what I optimizing for agents is going to be a big thing.

[00:17:52] Sani: S-E-O-C-R-O, whatever it is, I, I think in the next few years until we switch to MCP, optimizing websites for agent interactions is going to be a big thing. One more thing that you talk about frequently is entity optimization, and that kind of has to do with being discovered by lms, being discovered by ai.

[00:18:10] Sani: So can you tell me more about. Why entity optimization and being an authentic, being authentic online is, is so important in this era.

[00:18:20] Jes: I think there's a few reasons. The first is that we, we know. Many of these platforms, whether that is a, a Google surface, whether that is an open AI surface, whether that's a Microsoft Surface, they're all using knowledge graph. And so making sure not only that you have your brand entity in the knowledge graph, but that that brand entity is logically connected to other important entities.

[00:18:46] Jes: So the, the topics that you write about, the products that you sell, whatever is going to help it understand the relationships of the web. And when we think of entity, I think too often people think of. Them in a single silo rather than thinking about establishing those relationships. And you can do that in many ways.

[00:19:08] Jes: You can establish relationships through the way that you do your internal linking. So another reason not to have everything in the header, because not everything has the same level of relationship. You can do it through structured data where you declare the relationships between things. Um, especially if you're doing that in a graph based approach to structured data rather than an individual script based.

[00:19:30] Jes: Um, there's, there's lots of things where we can help, help these LLMs have a better grasp of not just the word, but these worlds behind them and how those worlds are interconnected with one another. And that's very important because it's not that I'm most of the time going into one of these LLM surfaces and asking for a specific brand.

[00:19:54] Jes: I already know the specific brand. I don't really need to do that form of discovery. I'm going to be asking this brand versus that, or what is the best, or I'm going to be doing these top of the funnel, middle funnel research type tasks, and that can very quickly lead to a conversion. If you have those entities really strongly established and you have consensus around the entity.

[00:20:21] Jes: You've managed that conversation in a very healthy way for your brand, your products, your people, your authors, your whatever entity it is that your brand owns or is trying to influence.

[00:20:32] Sani: But this goes beyond your website, right?

[00:20:35] Jes: Yes. 'cause an entity doesn't live on a website. A website is one instance. Of an entity. It's just one situation where it can live.

[00:20:46] Jes: So two is the app. So two is social media. So two is mentions on Reddit. So two is mentions in digital pr. So two is mentions offline that get photographed, and now your brand logo is in an image or a video. There's so many variances of a brand entity or of a product entity that you need to be curating.

[00:21:09] Jes: For whether that's a human user or an AI user.

[00:21:14] Sani: So siloed is bad once again and, and always will be. And treating your website as. I think on its own, it is just, it, it's a way to fail. You mentioned Reddit. Uh, I have a question. How long before we see all those let's spam Reddit to get your brand to be an oms. How long till we see that backfire.

[00:21:35] Jes: Not long, I would say it's already backfiring. So, uh, there was a post on LinkedIn that I saw earlier in the week, and the LinkedIn post was, you know, your, your typical tech bro being so proud of how he pretty much spammed Reddit. Um, with an AI bot that commented on all of the threads and said that his product was the best, and now he was being recommended in LLMs.

[00:21:58] Jes: And I was just sitting there thinking to myself. What now, because you know what's gonna happen now? A whole bunch of people are going to go to what they feel is an authentic community. They're gonna purchase that product. Likely they're going to be disappointed because if it was truly a good product, it would've been organically mentioned, and you wouldn't need to be AI spamming your stuff out there.

[00:22:19] Jes: So let's, let's not give them the benefit of the doubt. Let's say it is a bad product. I use it. I'm gonna use that product and I'm gonna be upset. And do you know the first per place I'm going to go when I am upset? To that Reddit thread that told me to buy it. And I am going to unload

[00:22:36] Sani: Oh

[00:22:36] Jes: a very agitated way at scale.

[00:22:38] Jes: 'cause this is not one person you've talked to. You've talked to all of Reddit and they will gang up. And so now your influencing of the LLM. few weeks later, or however long it takes to consumer product is going to end with a mass of negative sentiment, which, you know, who's gonna pick that up? The LLM and you know, where you're never gonna be able to wipe that out of the memory of the LLM training data.

[00:23:03] Jes: And so you've, you've, by manipulating you've made it worse. And we see this in human behavior. If you lie, you will be caught out. If you manipulate an LLM, you will be caught out because there is. Humans, you, there's always a human that uses the product. A robot, an algorithm, an ai. It doesn't use the thing you sell that product, that service.

[00:23:28] Jes: So you need to understand that you need a good product, you need a good service, and there's no amount of marketing or LLM manipulation or prompt hacking or whatever you wanna do that's going to get you out of making a good product or service.

[00:23:43] Sani: Just that rant was a masterpiece. It was beautiful. I didn't wanna say a word until you were done with it, but I completely agree with the, the only thing this will help you do is fail faster. Maybe there's short term success, a week or two of success, but then it's over forever and your brand is never going to be seen favorably in, in, in an LLM.

[00:24:04] Sani: Uh, so if that's the goal, if you have a thousand products you want to offload and never be in that business again, I'm not saying do it because it's still wrong. Uh, maybe some people will do it, but if the goal is long-term success and winning online and not just online, it's the dumbest thing to do. And there are so many people and it's always bros.

[00:24:27] Sani: There's never a woman suggesting to do that. I don't know why, but it's always AI tech bros who discover 10 and they can now automate whatever they want. I'm not a fan. You're not a fan either, from what I can tell. Uh, but Reddi is a legitimate way to, uh, influence in, in a good way, in an honest way, to influence the discussion around your brand online, because that's what real discussions are happening right now.

[00:24:54] Sani: It's not that Reddi is magic, it's the people are there. And that's it. Good,

[00:25:00] Jes: It's, it's about, um, guiding the conversations, but doing it in an authentic way, same way we respond to customer service. Um, either complaints or compliments. It's another customer service channel. It's another brand management channel, and it is important to be on that brand management. But we need to factor this into a larger strategy like, uh, was it, I think it was Domino's, either Domino's or Pizza Hut, I don't remember which.

[00:25:27] Jes: But they had terrible, like appallingly bad pizza a few years ago, and the CEO stepped up and said. Our pizzas are not his words, but terrible. He used a stronger word than that and he apologized and he says, and we're going to change and we're gonna start using real cheese and not have our pizza dough taste like cardboard.

[00:25:48] Jes: And because he owned it and he owned it in a public space and said, we are changing, that was a really successful campaign. And so if you've had a terrible experience or user's had a terrible experience, own it. Apologize for it. Do better sh and then actually do better. Because if you don't actually do better, it will come back to haunt you.

[00:26:11] Sani: even worse. Yeah.

[00:26:13] Jes: Yeah. But it, it's

[00:26:14] Sani: don't think people are idiots. Yes. And people will figure it out. No matter what you do. People online and people in general are too smart in 2025 to be liked. Well. Most people are too smart to be lied to. Uh, I like this a lot. I, I like the fact that it seems like to influence LLMs, you have to ignore LLMs, and, and that kind of gives me hope for the future.

[00:26:37] Sani: I'm, I'm serious. I, I, I mean, how else are you going to get into an LM other than not mess with LLMs and have it just flow into the LMS organically from the real world? It could be that

[00:26:50] Jes: to it's always been brand and how you are reflecting that brand either offline, where we've got. Much more years of experience or online where we lost our way. But I feel that AI is bringing us back to strong marketing science principles because it's more complex than an algorithm and you can manipulate an algorithm.

[00:27:16] Jes: It's both easier and more difficult to manipulate an LLM.

[00:27:23] Sani: I agree with that. Uh, it, yeah, I mean, you could hack an algorithm. Google, there's so many SEO hacks over the years that you, people come up with something, then Google patches that and it no longer works. I think with ai, the risk of doing that is because not even, even the, the companies making these models, they don't know how, what they will do, so.

[00:27:47] Sani: Hey, good luck. If you want to try to hack it, you may look like an idiot, but if that's what you want to do, play with it and, and see what happens. Uh, you, you've had a very interesting career. You went from startups to enterprise, uh, you worked with ring gear. What should people focus on for the next, the, the rest of this decade?

[00:28:08] Sani: People in marketing and SEO.

[00:28:11] Jes: That's a big question. I think you should focus on what you are uniquely good at. I feel that a lot of the conversation the last 10 years has forced us to be good at everything or expected us to be good at everything. And even if you like to be an SEO, that's, that would be a huge call. That is technical, that is editorial, that is brand, that is strategy, that is pr, that is ux.

[00:28:46] Jes: Like there's so many fields under that umbrella. And I think we lost our way a little bit of do something really well and do that thing well that you're passionate about. So that work isn't work. Work is something that you enjoy doing and. When we do that, you are less likely to have to rely on these.

[00:29:12] Jes: I'm going to go and manipulate a Reddit thread because I don't know how to do digital PR or digital brand management. If you'd really do digital brand management, you will never need feel the need to take that step because you know better ways. And so focusing down on what you're really good at will be overall good for our industry 'cause we see this.

[00:29:35] Jes: Transition and this argument that's happening right now, I-U-S-E-O or I-U-G-E-O. And one of the arguments that I see online is that, okay, well we have to throw off the shackles of the old branding, bad branding of SEO 'cause it was hacky. And then they go and do GEO, which is either SEO under a different name or yeah, manipulating LLMs for a short term win that is there for a week and then gone.

[00:29:59] Jes: Um, why don't we just do our jobs better? I think we can.

[00:30:08] Sani: I, I think we can for sure. Also S-E-O-G-E-O, uh, a IE. Oh, there's so many names. It, it just makes no sense. I, I think it all goes back to maybe not what SEO was 20 years ago, but what. Google has subtly trying to push as the ways to approach SEO for the last five or so years. You know, EAT, the messy middle.

[00:30:31] Sani: They, they've had all those small, subtle clues. Think about the user, the page experience. Uh, you want the user not to go back to the Google search. They've been telling us what to do before ai. Maybe they knew this would happen, but they've been telling us how to optimize for this era. And if you had been paying attention, people had been paying attention, they would know what to do now.

[00:30:50] Sani: So. Basically what you said there, uh, is to, to win in ai, be human and, and just be, be, be creative and like he can, we need to ignore AI a little bit more. I, I, I think we need to forget it exists when we're doing our jobs and just use it as a, as an assistant or whatever else. Two final questions. The one key skill for 2026 that people need to start adopting in digital marketing in general, or SEO, whatever you want.

[00:31:21] Jes: The ability to think this is what AI threatens the most, because when you say. Okay, now I need to do Strategy X. And so you go to Church of PT and say, how do I do strategy X? You've lost the ability to think and then you've lost any creativity. You've lost any brand message. You are going to be, you're commoditizing marketing strategy, and that won't work because you have nothing that will hook a user because you are just regurgitating.

[00:31:54] Jes: Through the LLM, what everybody else has done before, and that's not gonna break through. You have to stop, think, and then yes, use an AI to refine those thoughts to polish a hundred percent. I polish a lot of what I do with ai, but it does not do my thinking, and that's what people will pay for your brain, not your ability to use an ai.

[00:32:23] Sani: I, I absolutely agree. And also, uh, I, I just had a podcast episode two weeks ago about this specific thing. If, if it's too easy to create something, did you really do it? Is it really good? Do you know if it's good? If you, let's say you want to use AI to do some kind of a report for you, and you don't know what a good report in that category looks like.

[00:32:44] Sani: AI will tell you it's good. You will not know if it's good. And then we got, we got flood of terrible, terrible AI created content that people who created it think it's great. And people who know, know it's terrible. And, uh, you lose your ability to think, like you said, or you never developed the ability to think.

[00:33:03] Sani: It's sad that this needs to be done in 2026. We've had a lot of years to start thinking, I, I is the way I feel about this, and, but better late than ever. And what is one old habit people should just drop immediately?

[00:33:21] Jes: Uh, focusing on short term metrics and measuring that on a weekly basis, you will not gain any valuable insight. For your long-term success. By looking at your metrics every day or every week, you need to take a longer perspective and you need to put that in context of the environment in which you compete.

[00:33:45] Jes: I feel like too often SEOs rely on Google Search Console, and when their clicks are going up, they think I'm doing a great job. My clicks are going up. It can just be that the market is going up and your clicks might be going up 1%, but the market is going up 5%. You are losing market share. You will never see that in Google Search Console.

[00:34:07] Jes: So we need to get comfortable with not having precise metrics, which you don't have anywhere. You can just delude yourself into thinking you do. We need to get comfortable with looking at, this is my market share. Not, these are the two competitors I'm measuring myself up against because I feel that I can beat them.

[00:34:26] Jes: It's really looking honestly, at your position in the market and that's going to help you make better business decisions and better investment decisions for your time and your resources.

[00:34:36] Sani: That's just great advice. Jess, thank you so much. Uh, I, I absolutely loved every second, especially at rent. That's going to be the, the trailer for this episode, but I agree with every word of it. Uh, where can people find you and where can people follow your work or

[00:34:52] Jes: So I am most active on LinkedIn. Uh, I will start to write, I wrote the first one and then I took a big break. But I will start to write my newsletter that's called SEO Brief. Uh, it's on Substack, or you can find it through my website, which is jess schultz.com. Um, those are the two, two main ones for next.

[00:35:12] Sani: Excellent. Thank you so much and to everyone listening, please consider rating, reviewing and sharing the episode, and I will talk to you next week.