[Host] Welcome to the A.E.O. Engine AI Search Show, the A.E.O. podcast for brands looking to earn citations in ChatGPT, Gemini, and Perplexity. I'm your host, Vijay Jacob, Founder and CEO of A.E.O. Engine. Today's episode tackles the biggest argument in search marketing this week. Rand Fishkin, the founder of SparkToro and the guy who founded Moz, posted a five-minute whiteboard video on X arguing that AI visibility tracking very well might be a waste of time. And the industry lost its collective mind. With me to sort through the wreckage is Marcus Reid, industry analyst, ex-Google Ads, and formerly the founder of a martech startup that, quote, taught him humility. Marcus, welcome back. [Guest] Hey everyone. Yes, the startup taught me humility and my accountant taught me paperwork. Good to be here, Vijay. [Host] So let's set the scene, because I think every marketer listening has lived this exact moment. You're in a strategy meeting. Someone asks, are we showing up in ChatGPT? Nobody knows. You check. You're not there. Worse, a competitor is. So someone's boss asks why, and suddenly there's a line item in the budget for an AI visibility tool, and three sales reps from three different startups have emailed you before lunch. That feeling of being behind, plus tool fatigue, plus a hard attribution problem. That's the emotional fuel for this whole debate. [Guest] Right, and Rand Fishkin poured gasoline on it. His argument was simple and it rested on three new studies suggesting that AI search referrals and visibility metrics don't translate into meaningful business results. His framing was that tracking AI visibility, quote, very well might be a waste of time. And because it's Rand, because he has two decades of credibility from Moz, his framing shaped the conversation more than any individual study did. [Host] Which is worth pausing on. Half the people arguing about this never read the studies. They read a tweet about the studies. It's like reading a restaurant review of a restaurant review. [Guest] Exactly. And one honest caveat, which almost nobody in the threads mentioned: the three underlying studies weren't linked in most of the discussion, so their methodology couldn't be independently evaluated. People were fighting over a summary of a summary. That should make everyone slightly uncomfortable. [Host] So let's define terms, because the phrase AI visibility gets used loosely. The cleaner definition I've seen is about whether AI systems, ChatGPT, Google AI Overviews, Perplexity, Gemini, Claude, can correctly understand and represent your business. Not just whether they mention you. Whether they describe you accurately. [Guest] And there's a distinction that matters more than people realize: being cited as a source versus being mentioned in an answer. Those aren't the same outcome, they're not mutually exclusive, and most tools blur them together. If a study measures referral clicks from citations, it's measuring one narrow slice of a much bigger thing. [Host] That's actually the core of the counterargument that emerged. Critics said referral-click studies miss brand search entirely. Someone reads a ChatGPT answer that names your product, doesn't click anything, and later Googles you directly or just signs up. That conversion is invisible to a referral-tracking methodology. [Guest] Right. The debate isn't really about whether AI mentions matter. It's about measurement versus outcome. The skeptics on Reddit put it well. In one digital marketing community, the top sentiment was that we can barely track traditional S.E.O. accurately, and that's a much more mature space with established metrics. So why would we trust AI visibility tools, which are younger and messier? [Host] And the vanity metric question came up hard. A thread in a B2B marketing community titled, is AI visibility useful or is it just another vanity metric, pulled over forty comments. The recurring theme: marketers cannot attribute revenue to AI mentions, so it feels like counting applause. [Guest] Applause is a good analogy, actually. Applause doesn't pay rent. But also, nobody ever sold out a theater by ignoring whether people clapped. The defenders had receipts too, though. One operator who runs multiple brands said that when they actually optimized for AI search, technical foundation plus content precision, they got cited in LLMs. That's a practitioner-first data point. It's one person, it's anecdotal, but it directly contradicts the idea that nothing you do moves the needle. [Host] Let's push on the mechanism, because this is where I land. The answer engines that search the web, ChatGPT Search, Perplexity, AI Overviews, pull heavily from top-ranking results. Which means AI visibility doesn't replace S.E.O., it builds on it. The same authority that ranks you is the authority that gets you quoted. [Guest] Agreed, and there's a compounding effect on top. Optimization strengthens how these systems recognize and cite a brand, and that advantage compounds over time. Early adopters reportedly see around three times more brand mentions than brands that haven't optimized. Now, full disclosure, that stat comes from vendors in the visibility space, so discount it accordingly. Even discounted, it points at something real. [Host] Here's where I think the skeptics underweight one thing: the pre-click moment. AI recommendations increasingly happen before any click, any site visit, any conversation with your sales team. And the AI may position a competitor as the default answer instead of you. You don't get to opt out of that. Your absence is a statement the machine makes on your behalf. [Guest] That's a fair framing, and it's where I'll mildly push you, Vijay. I think the pre-click moment matters for brand positioning, sure. But I'm less convinced it converts for everyone equally. If you're a big brand, being named in an answer probably does drive branded search later. If you're a small Shopify store selling phone cases, I genuinely don't know if a ChatGPT mention moves product. I could be wrong. But the skepticism isn't irrational, it's category-specific. [Host] I'll accept that. I'd counter with the SaaS community thread asking whether AI visibility is even a problem people will pay to solve. Notice that's a different question. Not, does it work, but, is there a durable market here. Even the tool builders aren't sure. That tells you the category is early, not that it's fake. [Guest] And there was a rough consensus middle ground even among sympathetic voices: cautious, low-cost experimentation rather than big platform commitments. Which, coming from an industry that usually buys software the way toddlers buy candy, is almost touching restraint. [Host] Almost like a startup betting everything on a launch that breaks the next day. You'd know nothing about that. [Guest] Nothing whatsoever. We shipped a dashboard once and it broke before the demo call ended. I now consider that a form of training. [Host] Let me bring this to ground for our audience, because there's a practical read here. The whole debate assumes you have to choose between trusting vendor dashboards and ignoring AI search. That's a false choice. The sensible playbook is: fix your technical foundation so AI crawlers can read you, make your content precise about who you are and what you sell, and measure the full funnel, including branded search and direct signups, not just referral clicks. [Guest] And realistically, that work takes weeks to months, not years. Frameworks in the space talk about eight to twelve weeks to become visible across the major platforms. The brands that will dominate AI answers later are the ones building this systematically now. Though I'll flag my own buzzword alarm going off mid-sentence there, so let me just say it plainly: publish clear, structured, honest content and the machines can quote you. That's most of it. [Host] That's honestly the thesis behind A.E.O. Engine. We run always-on AI content agents for ecommerce and B2B brands, doing exactly the unglamorous work Marcus just described, technical cleanup, precise content, structured data, so brands become the cited answer rather than a buried link. Not because dashboards are magic, but because the compounding starts with representation. [Guest] And to be fair to your side of the argument, Vijay, one of the better definitions floating around is that visibility is about whether AI systems can correctly understand and represent your business. That's a real business risk. If the machine describes you wrong, that's a reputational channel you don't control. [Host] Quick recap. Rand Fishkin lit up the week by saying AI visibility might be wasted effort. The studies behind it are hard to independently verify. The skeptics raise a real attribution problem. The defenders show real citation wins and real blind spots in referral-click methodology. My read: the measurement is immature, but the channel is not optional. The honest answer is experiment cheaply, measure the full funnel, and start before your competitor becomes the default answer. Head to A.E.O. Engine dot A.I. if you want to see how we do it. Marcus, good to have you. [Guest] Always a pleasure. Rand, if you're listening, I'd love the links to those studies. Everyone else, stay skeptical, but not so skeptical you become invisible.