Rank Prompt's study of nearly 30,000 YouTube citations finds only 14% of AI-cited videos also sit in YouTube's top 10 for the same query, leaving 86% outside that ranking surface. On The GEO Show, Paris Childress treats that as the top story and pairs it with Peec's Meta Muse Spark coverage, ChatGPT Ads expansion across Asia, TechMagnate's citation-stability data, an AI Mode CTR experiment, BrandGhost cross-engine overlap near 10.6%, and vertical plays from Fullspan Health, RevEO, Trebelino, and Ulerity.
Welcome to episode 43 of The GEO Show, the briefing on Generative Engine Optimization and AI visibility. Brought to you by GEOforge.
In this solo roundup, Paris Childress, founder of Hop AI and co-founder of GEOforge, leads with Rank Prompt's finding that 86% of AI-cited YouTube videos sit outside YouTube's own top 10, then covers Peec on Meta Muse Spark, ChatGPT Ads in Asia, TechMagnate citation stability, an AI Mode CTR experiment, BrandGhost overlap, Fullspan Health AIQ, RevEO, Trebelino Visibility One, and Ulerity's local OS.
Rank Prompt analyzed almost 30,000 YouTube citations across 13,788 prompts and nearly 1,600 brands. Only 14% of cited videos also appeared in YouTube's top 10 for the equivalent query; 86% came from further down or were not surfaced there at all. Paris treats this as the top story and argues YouTube is the strongest single GEO play right now, because freshness and topic fit can beat YouTube rank and view counts inside AI Overviews and AI Mode. That democratizing force should encourage marketers to publish more topical video even without legacy traction.
TechMagnate tracked 83,633 citations across 2,191 domains over eight weeks in India's personal loan category. Average weekly domain churn hit 39% (ChatGPT 44.8%, Perplexity 35%, Google AI Mode 31.8%). Only 15% of domains appeared in all eight weeks, yet those domains captured 93.4% of citations. Paris stresses weekly trend lines and replacing lost citations with new content rather than celebrating one-day wins.
Peec added Muse Spark coverage so teams can track visibility, sentiment, and cited sources as GEO measurement moves from chat UIs into agents that book, negotiate, and purchase. Paris frames this as an early sign of an agent-surface taxonomy beyond chatbots.
OpenAI expanded ChatGPT Ads to Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam, and Taiwan, taking the product past 60 countries. Paris says paid and organic AI discovery now need joint reporting, similar to the old SEO and PPC split.
A pre-registered randomized field experiment with 1,100 US Google users found an AI Mode-only experience cut external publisher click-throughs by almost 19 percentage points, lowered Search frequency by about 0.92 sessions per day, and reduced reported trust and satisfaction. More evidence that answer engines suppress publisher clicks while influencing buying decisions before a site visit.
BrandGhost's September 23 observatory run across ChatGPT, Claude, Gemini, and Perplexity put cross-engine source overlap near 10.6%, with 49.4% of comparable recommendation sets showing zero brand overlap. Paris reinforces platform-by-platform reporting, at minimum ChatGPT, Google AI Overviews, and Google AI Mode.
Closing beats cover Fullspan Health AIQ's proprietary publisher stack (Healthline, Healthgrades, Medical News Today), RevEO's closed-loop task OS that ranks interventions by economic value then remeasures, Trebelino's Visibility One PR practice, and Ulerity folding AI search into a local marketing OS for multi-location brands. GEO keeps blending into PR, local, and SEO-plus workflows rather than staying a standalone discipline.
Subscribe to The GEO Show wherever you get your podcasts, and watch the full episode on YouTube: https://youtu.be/JvMqrn7tzVw
Audio: https://share.transistor.fm/s/85427bfd
Question: Are you still optimizing YouTube for YouTube rank, or for AI Overview citations?
Paris Childress: Hi, everybody. Welcome back to The GEO Show, brought to you by GEO Forge: Full Self-Driving for AI Visibility. I am re-recording episode forty-three with some great stories. So let's get right into the number one story of the day. It's about Peec, and that is Peec, P-E-E-C. Peec extends AI visibility into Meta's personal agent layer.
Paris Childress: All right, this is the first instance that I've seen of an AI visibility platform integrating directly with Meta's Muse Spark. Peec added support for Muse Spark, the model powering Meta's new Muse personal AI agent, and it's the hottest AI product on the market. Peec says that teams can measure visibility, sentiment, and cited sources alongside its existing answer engine coverage.
Paris Childress: Importantly, Peec itself describes Spark track- Spark tracking as the closest approximation to Muse. I think Spark here refers to Google Spark. So it describes that the Spark tracking, which I guess it already does, as the closest approximation to Muse because the agent is personalized and cannot be directly reproduced by a monitoring tool.
Paris Childress: So I wonder how they are monitoring a brand visibility inside of both Spark and Muse. it's probably through, through the login, I guess, of the users. But clearly now GEO measurement, this is an early sign that GEO measurement is moving beyond chat interfaces like ChatGPT and into agents that can actually take action on behalf of those users.
Paris Childress: They can choose things, they can book, they can negotiate, they can purchase. And so this is a distinct agent surface taxonomy that goes beyond, chatbots, basically. So I, I think this is the, the future here. It's gonna be very interesting when a lot of the AI usage shifts from chatbots into these personal assistants like Muse and GrokBot All right, next story.
Paris Childress: ChatBot, I'm sorry, ChatGPT Ads expands to seven more Asian markets. OpenAI announced expansion of ChatGPT Ads across Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam, and Taiwan, taking ChatGPT Ads to more than 60 countries now So paid AI discovery is scaling geographically very quickly alongside organic recommendations.
Paris Childress: So we are, just barreling ahead here. ChatGPT Ads, it's happening. that paid surface area is becoming more and more important, and I think anyone that is measuring AI visibility, especially if you're an advertiser in ChatGPT Ads, you really do need to be looking and reporting across both the paid layer and the organic layer, just like the SEO PPC split of old days On to the next story.
Paris Childress: Citation stability is becoming more important than citation presence. TechMagnate tracked eighty-three thousand six hundred and thirty-three citations across two thousand one hundred and ninety-one domains over an eight-week period in India's personal loan category. Average weekly domain churn was thirty-nine percent.
Paris Childress: ChatGPT churned forty-four point eight percent, Perplexity thirty-five percent, and Google AI mode thirty-one point eight percent. Only fifteen percent of domains appeared in all eight weeks, yet those fifteen percent of domains captured ninety-three point four percent of all citations. So what this is showing us here is that citation tracking and citation presence in AI answers is extremely volatile.
Paris Childress: You might, you might be winning a citation today. That citation might even hold for a few days, but chances are over the long term, it'll drop. And it's not very stable, and it's very, very volatile. So the important thing with AI visibility tracking, share of voice tracking, citation rate tracking, it is to establish long-term trend lines and to really observe the volatility versus the relative stability of your citation presence.
Paris Childress: Ideally, what you wanna have is citations that are stuck, just like in the SEO world, where you had rankings that were stuck and in place, high rankings that were stuck and permanent and stable. That just unfortunately doesn't exist with citations, so you have to track it, every... ideally every week, and you have to watch how it's moving.
Paris Childress: And when you lose a citation, you may never get that back, but you might have to replace that with a new piece of content. So citation stability, I think, implies a whole lot more content than it did before with SEO. Next story. Randomized AI evidence. randomized evidence shows that AI mode reduces publisher traffic.
Paris Childress: Well, this is kind of like a big duh, 'cause we've said that before. A pre-registered randomized field experiment involving eleven hundred US Google users found that an AI mode-only experience reduced external publisher click-throughs by almost nineteen percent-- nineteen percentage points, lowered Google Search frequency by almost one session per day, zero point nine two sessions per day, and reduced reported trust and satisfaction.
Paris Childress: So, this is not that surprising that AI mode is reducing clicks because it is providing the answer. And, yet more evidence keeps coming that, GEO is increasingly influencing the buying decision before any website visit happens Next story. 86% of AI-cited YouTube videos are outside of YouTube's own top 10.
Paris Childress: Now, this is, this is pretty big. I consider this to be our top story of today's episode. Rank Prompt analyzed almost 30,000 YouTube citations across 13,788 prompts and nearly 1,600 brands. What they found was that only 14% of cited videos appeared in YouTube's top 10 for the same query. So this is YouTube search, this is not Google search, and they were converting the prompts into queries, and they saw that only 14% of cited videos that were cited in, I, I presume this is predominantly AI overviews, but it doesn't say here.
Paris Childress: But only 14% of those were also ranked top 10 in YouTube search for the same or the equivalent query. 86% came from further down in the YouTube results or were not surfaced there at all. And this is fascinating to me. This is similar to the finding that citations do not correlate with SEO rankings, and now we're seeing the same thing with video.
Paris Childress: And I think right now, the, the strongest GEO-- the strongest single GEO play that you can make right now is YouTube, YouTube videos. Because YouTube videos are getting absolutely stuffed into Google AI overviews and AI mode, especially Google AI overviews. And what this is telling us is that it doesn't have to be a top 10 ranking YouTube video in YouTube search.
Paris Childress: mostly what, what I'm seeing is it's a freshness factor. You can have a brand-new video with practically no views at all, but if you're nailing the topic that is the topic of the prompt, then you'll be there. You're gonna win that citation. Doesn't matter if you're ranking or not on YouTube for that video, or even just how much traction that video has gotten in terms of views or engagement.
Paris Childress: And that's great because that's a really democratizing force, and that should encourage, you all listening who are, marketing and thinking about video. It should really encourage you to push harder and harder on video because you're not at a disadvantage by creating new videos that don't have any history or don't have any, any views or YouTube rankings.
Paris Childress: You can still win in AI overviews with those videos On to the next story. Cross-engine source overlap has fallen to roughly 10.6%. BrandGhost's September 23rd observ- observatory run analyzed, about 7,800 citations across ChatGPT, Claude, Gemini, and Perplexity. Cross-engine source overlap was approximately 10.6%.
Paris Childress: we've seen this before, the cross-engine source overlap being about ten percent. forty-nine point four percent of comparable recommendation sets had zero brand overlap, and average recommendation overlap was nine point nine percent. So again, there is increasingly no meaning-- mean-- there is no meaningful universal AI ranking.
Paris Childress: You really have to look platform by platform, and at a minimum, I would be looking at ChatGPT, Google AI Overviews, and Google AI Mode. Those are the three AI engines that we report on, and, I think that is, is adequate coverage. What we see consistently the-- is when you look at only ChatGPT versus only AI Overviews or AI Mode, you also get very, very different results in terms of share of voice.
Paris Childress: And share of voice is driven by, mention rate of brands in the answers plus citation rate. All right. On to the next story. Fullspan Health creates a vertical AI visibility platform around proprietary authority. RVO Health launched AIQ by Full Health-- by Fullspan Health, combining AI visibility measurement with its publisher portfolio, including Healthline, Healthgrades, and Medical News Today.
Paris Childress: So that is quite a deep, vertically integrated set of proprietary data that they have. And Fullspan reports category reach as high as eighty-one percent for migraine prompts and up to three point six times competitor citation frequency in cardiovascular prompts. So these vertical platforms that have all this proprietary data like RVO Health, 'cause they have Healthline, Healthgrades, and Medical News Today, they're really able to leverage that proprietary data for GEO results, and they're able to build a much stronger GEO moat when they own that trusted data, that authoritative content.
Paris Childress: They control, parts of audience distribution and even commercial relationships. So I think for those who are able to do this, to keep building proprietary data, that is only gonna put you in a stronger and more competitive position for AI visibility in GEO Next up, RevEO launches as a revenue-first SEO plus AI search operating system.
Paris Childress: Revim commercially launched RevEO, powered by IndexR. The product connects search and AI visibility data, ranks recommended tasks by economic value and effort. It assigns owners and checklists, then recrawls to verify implementation and measure outcomes. It tracks ChatGPT, Google AI Overviews, AI Mode, Gemini, and Perplexity.
Paris Childress: It says that its existing channel includes four hundred plus enterprise customers and one hundred plus digital marketing agencies. So again, this category is converging around prioritized execution. This is much more than just another reporting dashboard. It is another closed-loop workflow, and, that is becoming table stakes here.
Paris Childress: I think what they're doing also seems to be another blend of SEO plus AI search. And it is also-- What I find interesting here is that it is creating specific tasks for, I guess, for the user based on whatever they're finding with AI visibility data. And then it's ranking those tasks by economic value and effort, and then assigning those to owners with checklists.
Paris Childress: So it's like full-on project management. And then after those tasks are completed, it comes back to recall and verify whether those interventions had any impact, any measurement impact. So this is, I think it's the right process where you start with measurement, you establish a baseline of measurement, you have then recommended actions and interventions, which typically means creating content and building citations, off-site citations.
Paris Childress: And then you remeasure, you revisit the impact of those interventions on your initial baseline measurements, and you see if it moved the needle or not. And then you get some sort of a learning out of that, and you reapply that learning to the next cycle. And this is how a closed-loop system can, over time, with that data and that learning through every loop, it can turn into a flywheel All right, next story.
Paris Childress: Revisiting PR agencies getting into the GEO game. PR agencies are turning AI visibility into a permanent practice area. Trebelino/Keller launched Visibility One as its fourth core practice alongside public relations, growth marketing, and creative services. The service includes AI visibility audits, ongoing monitoring, content designed to be surfaced and cited, and technical guidance.
Paris Childress: This was reported by PR Newswire. So GEO is definitely escaping the SEO department. It's becoming a broader communications and reputation discipline, and PR firms, classic PR firms are absolutely jumping into the GEO game and taking ownership. This is another example of why I believe GEO is much, much bigger than SEO, because it is not performance marketing.
Paris Childress: SEO is performance marketing. GEO is brand and it's PR, and that's why clearly PR agencies are discovering this and latching onto it and adding it to their arsenals. Last story of the day Ulerity, I think that's how it's pronounced, U-L-E-R-I-T-Y, absorbs AI search into an autonomous local marketing OS. Ulerity expanded its, its agentic workforce for multi- multi-location and franchise brands.
Paris Childress: Its new performance agents combine listings, reviews, localized creative, paid media, AI search/AEO visibility, call tracking, and reporting into an observe, act, and measure loop. A very, very interesting story and a very interesting GEO local play by Ulerity. They're pulling in all the local signals. They're pulling in listings, reviews, localized creative, also paid media, and of course, the AI search and GEO visibility, adding call tracking to that into a full suite.
Paris Childress: And, I think that's a very, very interesting loop. So vertical operating systems can make standalone AI visibility really disappear into a broader marketing workflow. And this is, this is what is-- What, what I take from this is that GEO isn't really a black-and-white discipline where there's a clear workflow.
Paris Childress: GEO can be blended into other workflows. It can be blended into PR workflows, as we just talked about. It can be blended into local marketing workflows. It, of course, is often blended into legacy SEO workflows, and that is why it's, it's bigger. It's more fungible, more malleable than SEO, and that's what makes it so interesting.
Paris Childress: All right, that's all for today. Thanks to everyone who's been listening, and we'll see you all on the next episode.
