Why Your Marketing Dashboard Is Lying to You (And What AI Search Has to Do With It) | AJ Ghergich, Botify [Part 1 of 3]

Hey, real quick before we get into this, here's something that I've been noticing lately, and I'm seeing this with a lot of love. A lot of really smart marketers are jumping straight to the tools. They're on LinkedIn, they're asking, "Hey, what's the best say I've visibility tracker? Should we be using this tool or should we be using that one?" And these are all really good questions, but I'm seeing a pattern when we're reaching for the tools before we actually understand the fundamentals of how AI search works, like why some brands get cited and others don't, or what signals AI engines are actually looking for, or how citation and trust scores actually function. So I'm working to fix that. I'm in the planning stages of a free GEO course, a how-to GEO kind of thing, that walks you through the foundations before the tactics, because one to understand how it actually works, the tools start making sense. Without that foundation, you're really just buying a dashboard. The course is free, it's going to go live on YouTube and the podcast, and there's going to be a workbook to go with it, so that you can take notes and apply what you're learning. The newsletter is the first place I'm announcing it when it goes live, so sign up for the visibility report to get on the list, and you'll know the minute that it drops. Okay, to the episode. Hey, welcome back to found an AI. I'm Cassie Clark, a fractional content strategist in an AI search optimization expert. I'm also your guide through what's actually happening in search right now. Not the hype, but the real stuff, and if you've been hanging around LinkedIn or read it the last couple of days with this Google guide that just came out, you know exactly what I'm talking about. Today I'm kicking off a three-part series with AJ Gurgich, who leads a consultant in the AI team of Bodify. Bodify is an AI and traditional search platform, and AJ has been paying attention to how these AI engines work, since way back when GPT-2 could barely form a coherent sentence. If in this first conversation, well, let me just get straight into it. Your dashboards are lying to you. Bod traffic is out, and it's sorting your data in ways that are making leadership ask the wrong questions. AJ, bless them. Also says something that I've been thinking for a while now. Added Planet Wide Scale. People are building their AI Memory Foundations, and if your brand isn't part of that, you are not dismissing clicks. You're missing the story entirely. Let's get into that conversation. [Music] So I made you Gurgich. I lead the consulting and AI team at Bodify. Bodify is a AI and traditional search platform that helps brands be found everywhere. We work with a lot of e-commerce and retailers. Everyone from Big Box retailers, you know, the shoes you wear on your feet, the clothes that are in your closet, to mid-market, smaller growth brands that are trying to break out. We're an international company, so we have offices in London, France, and New York. So I like to ask this to nearly everyone that comes on the show. When did you start paying attention to how these AI engines work? Because I know you have the SEO handle over on X or Twitter, so you've been into it for a while. From the very beginning, I was super fascinated. I remember when GPP 2 came out and it could barely kind of string together coherent sentences. And I was hooked. This is the future. I was sold. I honestly, I was way too sold. I think I got some psychosis or something. I was like, "We'll definitely have AGI by 2026." I was very on it early and very in it early. So it's been fascinating to watch it go from, "Hey, that's just an auto-complete text." "Oh my gosh, this is AGI. Oh wait, it's not." Pump the brakes to it's changing the world infrastructure right now. Pretty cool to be involved all along the process. Yeah. So you were also part of recently Retail Economics Future of Search and Discovery Research with Richard Lim. Do you want to tell us about that before I start asking some questions? Yeah, honestly, we did a study, modify, we tried to do a lot of thought leadership. And so we worked with Datadome in AWS and did a lot of research around bots, agent-to-comers, put out a lot of information. It's a free study. I don't know if we can link it or mention it. Yeah, absolutely. But yeah, you can download it, Snagget. It's a lot of those, I'd be honest. I've been in the industry a long time. A lot of those things are pretty fluffy. This is data dense. It's got a lot of charts and graphs. A lot of takeaways in it. It's worth the download and it's hard to beat the price. It's free. So, it's Snagget. The brain is always nice. But there is one metric in there where you mentioned that bot traffic is up 4.5 times in 25. Okay, so what does that actually mean for brain analytics? And what are marketers getting wrong when they're looking at their dashboards right now? Yeah, I mean, I'll come in hot on this one. It means your dashboards are lying to you. That's what it means in plain speak. So bot traffic is inflating impressions and it's masking performance. So, I'll give you an example. In the report that I'm talking about, we mentioned this stat, there is an imbalance right now in kind of activity and impressions to visit ratio. So, every one visit that OpenAI sends to a retailer, right? A retailer website. It performs 198 rolls. So, let's just say it's one to 200 ratio, right? Google's ratio is one to six. And that's what we're more used to, right? Because search is that kind of thought to keystroke moment where you have active intent. And so, that's why you see this kind of like one to six ratio. It's around the intent. Right now, AI chat bots are a lot more in the discovery realm. So, what's happening is there's this flood of impressions that one visit to 200 impressions. It's making everybody's dashboards off. It's like, hey, our conversion rates are lower, our impressions are up, but our clicks are down. What's going on, and that's what's going on is the discovery has taken off with AI, but not yet the transaction. And so, it's not closing the loop. And so, there's a donut hole in everyone's data. So, what do you suggest when we're looking at these? Like, how do we think about it? Yeah, so I think it's similar to how you would think about early days of social. At first, you were like, hey, you know, well, I'm old. So, you know, it's like Facebook, my space, you know, we got likes. And then some board members like, well, we can't cash the likes at the bank, AJ. And I'm like, yeah, but they're going to be great. This is worthwhile visibility. And then now today, you know, it's like, who wouldn't launch a campaign on Insta, right? Like, of course you would, right? And who wouldn't do that? We know that there is a transactional value that occurs, whereas before, it was all about discoverability. We're just in that cycle right now, where most folks have adopted AI for research and discovery, but the analytics aren't there, the tracking isn't there yet. And so, it's really hard to close the loop. My point would be do the same things that you did. Like, the smart brands that jumped on social, do we feel like that was a bad move? No, it was a good move. The smart brands that jump on AI search now are going to be rewarded later. The analytics will come. It will, that donut hole will get closed. Don't sit on the sidelines because you can't track things perfectly, but do know that there's a hole in your data and you have to be okay with that. So, my husband buzzes at me all the time because he'll say something and I'll be like, this is what I'm hearing you say. So, this is what I'm hearing you say. Don't worry about the numbers right now, but do the research and do the work to optimize for AI search visibility and just, it'll catch up later. Yeah, exactly. And know that you're going to see outside impression data versus concrete, like, transactional data, either clicks, transactions, and know that that's normal and that it's not only happening to you, it's happening to everybody. So, if I'm listening to this, my question is, okay, I'm hearing, I'm an SEO practitioner or I'm a content manager and I'm hearing you say this, but leadership is not on board yet. What am I saying to them to make them feel comfortable that, hey, clicks are down, but this is what's happening. Yeah, I would say that they have to update their worldview or data worldview. So, the good thing and the bad thing is these executives have, you know, she's a CMO, she's been trained on all this data. She's been trained that she has row as from paid media. She's been trained that she actually can attribute like in-store purchases to Instagram ads. It's not perfect, but she can do it, right? And so, we're actually in this world where we're not used to not having data. And we got to go back to that uncomfortable space where we don't have the data to back up every single thing that we track every dollar to every you know every dollar spent to every outcome. They have to update their world model and they they're data model and it goes even more important than that. It's how their mindset is like a lot of CMOs are in that click mindset and they need to move to a visibility mindset. So she's tracking clicks and conversions. That's 2020 thinking right? That's because that's the world that she lived in the data reports that she get. How many clicks did I get from Facebook from Google? How many conversions did I get? What was my click-through rate? What was my call? Like all that's logical. You don't rank in AI. It's it's stochastic. It's not deterministic. Whereas Google you ranked. We I ranked number two for this query. I get clicks. I track that. Here's my revenue. So if if she stays in that mindset she's going to continuously get reports that make her unhappy because she's not updating her world view to say wait a minute there's this whole other thing that's happening outside of my data collection stream that is important. These are the customers I want. Those are the customers that are buying. That's the 30 year old demographic that we're going for that are using chat. That's the buyer. You cannot ignore it. And so yeah if you keep that same you know clicks mindset I think it's going to harm you down the road. Yeah I was doing some podcast research this morning and the stat out now the big one is like 93% zero clicks in AI mode. It's a huge number and it's the feature that's not a bug like the whole answer is the product now from Google. It's the job to be done. The job to be done was never to click. Right. So like why do we want to force the customer there? That's not what the customer doesn't wake up and think I want to click some things today like she wants outcomes. And so if the AI can help us achieve an outcome faster and more informed it wins. It's good for the customer. We just got it. We have to we like I said you have to update your mindset or else you just live it. You're living in 2020 just like pretending that we're in a clicks world that you can just track that way. So you have to update and evolve or you're going to you're going to have a bad day every day because your numbers are going to keep looking like they're down and you're going to miss out on all the discoverability that could you could be celebrating if you update your mindset. Gotcha. Gotcha. Okay. So there's another question I want to ask you. So retail economics found that 73% of consumers have used AI assistants and 38% are using them for shopping. At what point does this stop being the nice to have and like an existential issue for brands? I feel like we're there. What do you think? Yeah. We're there. It's just the very last mile. Consumers are almost always there before the technology. Consumers were there before Uber. They just didn't know it yet. Consumers were there before Netflix. And so the moment Netflix dropped right everyone went well that's better. I joke about it. The V-mean. I do miss walking the aisles of Blockbuster. Trying to see which Rambo I'm going to rent or whatever. So I missed that but apparently not everyone else did. So the second new process, new technology comes along. They will drop the old thing. They will drop the cab and go to Uber. They will drop Blockbuster, go to Netflix. They will drop me doing all the research to buy a pair of socks and let the AI do it. That's been the pattern of human history. And so it's here. The only thing that is still a bit of hype is the last mile. The consumer. If you talk to anybody go talk to some you know few folks at a bar and ask them which one of you would allow an agent to run your checking account. You'll get no hands race. Right. Like no one will be like yeah. But like in a couple years I think that's common place. Like we won't even think twice. Like oh, Bank of America, I'm sure they have an agent running them. The whole thing right. But we're not there at the last mile. But all of this is pent up demand. Pent up consumer demand. They are waiting. She is ready to buy those socks with an agent. But the technology isn't quite there for the last mile for the trust. The moment it happens, it'll be a flood. And everyone will be like oh my gosh, it came out of nowhere. No, it didn't. It's been there the whole time. There's just been a lack of trust so that it can do the last mile properly. But it's there. The pent up demand is there because all the discoverability is moving to chat anyhow. And once discoverability is there, it's like to again go back to the Facebook effect. They started a business with no monetary value, but they got an audience. And if you capture an audience, you can monetize an audience. So they capture everybody on discoverability and then you move them to transaction by easy connections to Shopify and so on. You win. So I think what's interesting about all of this is that with like traditional Google with the Tim Blue links, the burden of research is on the user. You have to be the one to go find it. Now with chat GPT or Gemini or whichever one you're using, it does it for you. And I think that's something about the instant gratification that we're getting used to. Quick word before we get back to that conversation. If you're listening to this and you're thinking oh man, I really need someone to help lead this for me. That's what I do. I'm an AI search optimization expert and a fractional content strategist for startups and enterprise brands. I build content programs that connect strategy with execution and actually account for how AI search is reshaping visibility right now. This is not a 2016 SEO playbook with a fresh coat of paint. This is fresh for 2026 and where user search behavior is headed. This sounds like the kind of helping that you're looking for. The fastest way to reach me is to just email me directly. I'm at [email protected]. That's C-A-S-S-S-I-E at CassieClarkmarketing.com. Send me a note, tell me what you're working on and we'll go from there. Okay, back to the episode. Yeah, I think it's also just human nature. People are lazy by nature and that's not a bad thing. You're 100% right. Your instinct is right on. I sound like an AI. You're absolutely right. Tell me you interact with you without don't be hurt. Right, right. Insight that most people are missing. Now, the as soon as you can offload that you will, you'll drop it like a bad habit. So right now you have to think of all the work you have to do. You have to think of the idea. You have to formulate the question in a way that Google will get it right. You have to look and scan and summarize all the links. Decide which one doesn't look like a scam. Click on that one. You have to read the article. You have to go back. You have to search and sort the Amazon reviews and see which one is fake and which one is real. And all of that is a lot of like, there's like an effort economy that's been going on that's been running on you as the GPU. And if you can offload that effort to an AI, if you trust it and if you believe it knows you, that's the other thing. The personalization layer. If it knows you well and it knows your habits and you can trust it, you'll offload that like laundry. Right, like the second they come out with a robot that could do laundry, I think, you know, who's going to be like, no, but I really like doing laundry. Right. No one likes doing any of this. They will offload it in a heartbeat. So yeah, you're 100% right. Absolutely right. Yeah. So I think as marketers, we need to think about hate to say it this way, but I'm going to say it this way, I used to teach seven grade students. You have to think about, you have to think about the user as a seventh grader, like they will find the shortcut if it's there. Yeah. Yeah. Yeah. Because again, it's jobs to be done. It's not the job to be done. The job to be done isn't to click on blue links, summarize things, read reviews. That's not the job to be done. Right. So yeah, anything that's not the job to be done, the customer will drop for the actual outcome. Right. So what's the worst case scenario for a brand that ignores all this for another 12-ish months? I would say the thing that people should really be thinking about. If I was a CMO and I talked to a lot of, I get to talk to a lot of CMOs as a perk of the job. So I get to get a little bit of insight into how, you know, how she's thinking across companies. One thing that I'm having a conversation with is that everybody is making building the memory foundation with their chats right now. And so right now, I'm talking to my chat. It's learning about, you know, hey, I have, you know, three kids and here's their ages and like, I like this and I don't like that. And, you know, he writes this way. So my memory layer is being established and it's happening at a planet-wide scale. No one had AI a few years ago and they didn't have memory. So it's all being built now. I just can't conceive of a world where you wouldn't want your brand part of her memory foundation. Right. Like if you If you're, you know, Zappos or somebody and you're, it's like, you can't just sit it out and go, oh, okay, I'll just let her memory blocks be built with you buy shoes not here. Right? You must become part of that story. And obviously it's the same thing with marketing. You don't need to, if you're a smaller brand, it's not everybody's story. But I really think getting into the foundation of building as everyone's building up their memory layer is a sneaky way to think about it. Obviously that could go bad, you could be manipulative. I think about it in a positive way, right? But I would not let that sit out and let everybody else be part of this her story and my brand not be. That's a good way to think about it. And as you're talking, I'm thinking about how Gmail has the personalization thing now where it's like reading all of your emails and then pulling the Gemini. There's so many things that go into this that I don't think we've all connected just yet. Yes, yes. Now we just need one to read all of my LinkedIn messages and tell people no, but yes. You know, I was in here thinking this morning, like I should probably took it my inbox. No, don't do it. That's a tomorrow project. Okay, so you also mentioned in this report, this is the last question I want to ask you about. I mentioned in that report that JavaScript rendered content is largely invisible to AI crawlers. I have seen gosh, I don't know so many people on LinkedIn within the last maybe three weeks talking about this very specific thing. Now that feels like it's a bigger problem than I'm sure it is, but how common is this and how would a brand even know if they're being invisible because of the JavaScript rendering? Yeah, obviously tools like modify, we handle this for you, plug, plug, but do simple things. Like right now, if you're like, I'm not technical at all, do you have a Chrome browser? If you do, download the DevTools plugin or extension. I don't want to do that, just do it. It's a little extension. You click it and you can turn it off JavaScript. Turn off JavaScript on your PDP and then refresh it. Now see what content is there and what's not there. Obviously a company like Botify, we're doing this at scale for hundreds of bots and particularly. But I'm just saying just right now, if you're like, I'm on shop, I have a Shopify side. I want to look at your site with JavaScript on and off. You'll be very, very surprised. A lot of times like reviews and other things that are loaded externally will go away. Not care of sales and things like that. I was just with, I'll leave their name out, but just a major shoe brand at their headquarters we all know who they are. We all have them in our closet. And what we were doing was we were showing them their products side by side. And we're like, this is what Google sees. Here's your product. They see 5,000 words. Here's what chatGPTCs, it sees 3,000 words. And my point is, what are you comfortable with being lost in the context in those 2,000 words? What if that was like the use case, like someone's looking for shoes for walking with knee pain? And that, are you okay with 2,000 words of content being lost, context being lost? So I think about it as an accessibility issue. So if you had a user who has accessibility issues, maybe you maybe vision issues, you might give them larger fonts. Like AI has accessibility issues. It can't access everything. And so you really need to think about serving it just like you would a customer that has accessibility issues. And a lot of what you see where AI hallucinates and makes things up, it's just simply a lack of context. It'll improve over time, but it's a major issue right now because most of the web is in modern web is built on JavaScript. And so it having rendering issues causes a lot of headache for brands. That's part 1 with AJ. And as always, I want to leave you with the one thing I actually do this week if I were you. Go pull up your analytics. Look at your impressions versus clicks. At this point, that Google glitch that we talked about a couple of weeks ago, it should be fixed. So if your impressions are up and clicks are down, that's not a performance problem. That is AI discovery working on your behalf and your current measurement framework might not be keeping up with it. The move here is to update what you're measuring. Start tracking visibility alongside those clicks. Are you showing up in the AI generated answers? Is your brand being cited? Those are the new baseline questions. As AJ put it, the smart brands that jumped on social early were rewarded. That same thing is happening right now with AI search. You don't have to have it all figured out. You just have to stop sitting on the sign lines waiting for the perfect data. This is your sign to go experiment. It's a lot of fun tracking it draws me. Part 2 drops next Tuesday. AJ touches on AI governance and who really need to see that table when figuring that out. If you want help figuring out where your brand stands in AI search today, head over to CASTICLARKmarketing.com to get started with an audit or check out the info in the show notes below. As always, I would love you forever if you left a review or shared this with a friend. Thank you so much. That's it for today's episode. Until next time, stay Pizzaboo.