RAMP data puts Profound at 52% of GEO software spend among buyers while only 1.2% of SaaS buyers purchased any GEO tool in September 2026. On episode 36 of The GEO Show, Paris Childress covers that concentration-vs-opportunity open alongside AppsFlyer click-only AI attribution, PeakMetrics reputation GEO, Bluefish x Bazaarvoice UGC for e-comm citations, Rankability cross-engine citation study, Stacker low-DR earned media, FogLift tool-recommendation concentration, Ahrefs Brand Radar scaling to 30M prompts, QueryLift Japan seed, and RadIntel Radar.
Paris Childress opens episode 36 of The GEO Show with RAMP spend data: Profound at 52% of GEO software purchases among buyers, Peak at 18%, AirOps at 17%, while only 1.2% of SaaS buyers purchased any GEO tool in September 2026. The briefing then covers AppsFlyer click-only AI attribution across eight engines, PeakMetrics reputation GEO for communications teams, and Bluefish routing visibility gaps into Bazaarvoice review workflows.
The Rankability throughline lands mid-episode: across 1,645 observations on 10 platforms, no engine pair shared more than 24.1% of cited pages, and 55.2% of top-10 AI citations did not rank in Google's traditional top 10. Stacker shows low-DR earned media still earning citations. FogLift finds tool recommendations concentrating while one-third of citations come from YouTube, Reddit, SitePoint, and G2. Ahrefs expands Brand Radar to 30 million prompts. QueryLift raises a Japan-focused GEO seed. RadIntel Radar turns predictive audience intelligence into authority content.
[00:00:00] Paris Childress: Hi, everybody. Welcome back to another episode of The GEO Show. I'm your host, Paris Childress, and The GEO Show is brought to you by GeoForge, full self-driving for AI visibility We have got some great stories on this Friday, September 18th. Let's get right into it. Some of the themes here will be about attribution, GEO broadening beyond SEO, and power concentration into some of the major players.
[00:00:38] So these are some themes that we have seen and reported on in recent episodes, and they continue. So let's start with our first story of the day Coming from RAMP data. Fresh spend data from RAMP shows AEO power concentrating around three vendors. RAMP's updated [00:01:00] vendor data places Profound at fifty-two percent adoption among RAMP businesses purchasing AEO software.
[00:01:09] Peak is at eighteen percent, and AirOps is right on its heels at seventeen percent. Peak moved to number two and is labeled the category's fastest-growing vendor. AirOps has the highest competitor switch rate. RAMP's data comes from anonymized procurement activity across its customer base. So these numbers are not overall market share, but they are pretty good indicators.
[00:01:35] As we can see in the chart on this slide, if we go back just to April of last year, Profound was almost nowhere and, and really in the last, uh, year and a half or less, uh, Profound has really surged into a dominant market share position, uh, nearing fifty percent market share. And then it looks [00:02:00] like Peak and AirOps are battling it out for a distant number two spot.
[00:02:04] Uh, the biggest loser here is ClearScope, and I don't know if ClearScope really belongs in this category. I'm not sure what they've-- I haven't taken a, a deep look, but I don't really consider them in the AEO, GEO platform space. And then we have Scrunch and Athena, both flat for the last year or so, uh, at sub ten percent.
[00:02:31] So it's interesting to predict how this market is gonna shake out, how this SaaS category is gonna shake out. It looks like Profound right now is clearly winning. Uh, they made a major funding announcement, a hundred and eighty million raised at a one point eight billion valuation earlier this week. Um, but if you look at their growth, it is flat in the last few months, and really the only player who is consistently growing month over month is Peak AI.
[00:02:59] [00:03:00] So let's see if Peak can pull away from AirOps, and then let's see if Profound can continue to grow. Uh, one thing that I wanna point out here, though, is this figure. One point two percent of RAMP businesses using SaaS bought an AEO tool in September of twenty twenty-six, which is this month That is nothing.
[00:03:25] So when we look at these market shares and Profound is winning and Peak and AirOps are number two, number three, but let's put this all in perspective that just about only 1% of companies that buy SaaS software has purchased in this category, and that is nothing. So that means I think it's still a wide-open race for people, and I do think that stickiness is hard to come by and people are gonna be changing and testing new tools quite a bit On to the next story.
[00:03:58] AppsFlyer turns [00:04:00] AEO into an attributable acquisition channel. AppsFlyer now attributes organic discovery across eight AI engines and paid ChatGPT ads separately, carrying both through app re-engagement, installs, and web/mobile conversions. So this is about app attribution here, app re-engagement, app installs, and then web and mobile conversions.
[00:04:29] The supported AI engines include ChatGPT, Claude, Perplexity, Gemini, Copilot, Groq, Manus, and DeepSeek. I took a deeper look into this, and it's not what I had hoped. Uh, attribution isn't solved by any stretch. These are still click-based. So when someone clicks on a link from, let's say, a ChatGPT result, they may be, uh, on the web, or they may be on mobile, uh, on a [00:05:00] mobile app.
[00:05:00] Most likely, this is a click in a mobile app, and that trick... that click is tracked all the way through to conversion, and then that gets attributed. So this is still click-based attribution only. Uh, this does not count if someone has some so-called impression-based conversion, where they just see a brand mentioned, and then they go and, uh, do a search for that brand, or they arrive through other means other than the click.
[00:05:29] So I, I got a little bit excited when I read this headline, but when I dug into it, it really is just click-based attribution at the mobile layer. But it is a positive step. All right, on to our next story. PeakMetrics, not to be confused with Peak AI, the platform, P-E-A-K, PeakMetrics, brings GEO into communications and reputation intelligence.
[00:05:55] PeakMetrics launched AI Perceptions on September 17th for [00:06:00] communications teams. It tracks how ChatGPT, Gemini, Claude, Groq, and Perplexity characterize brands, then connects those perceptions to narratives spreading through the news, social media, and online communities So this is not about visibility.
[00:06:18] This is about reputation and narrative intelligence. It's basically looking at how are the current stories and social media discussions around my brand affecting, um, the reputation and the way AI is talking about my brand right now. So it's a little bit more real-time and reputational based. But I think this is an interesting divergence of GEO from a pure visibility standpoint to more of a reputation monitoring and narrative intelligence.
[00:06:54] And I think that we'll see in the last story today s-something similar, that a lot of [00:07:00] PR agencies are moving into this space with this angle, not purely visibility, but reputation and accuracy of presenting the brand and then connecting that to how, uh, the current news stories, current social media buzz, and also within online communities is shaping that narrative in real time
[00:07:25] Next story, Bluefish and Bazaarvoice connect AI monitoring directly to UGC execution Bluefish and Bazaarvoice announced an enterprise partnership in which Bluefish detects AI visibility gaps and routes recommendations into Bazaarvoice review, sampling, and structured content workflows. Their joint research found that 90% of surveyed consumers consider it important to know an AI recommendation is sourced from real reviews or customer photos.
[00:07:59] In [00:08:00] Bluefish's dataset, fifty-nine percent of AI-cited product pages had a hundred plus reviews, ninety-two percent had ratings of four stars or higher, and sixty-five percent displayed a review summary near the top. The research combines a seventeen hundred plus consumer survey with two hundred and thirty-seven thousand eight hundred and four citation instances, and it is vendor-sponsored.
[00:08:26] So this is a legit report. Structured UGC, user-generated content, is becoming part of the machine-readable recommendation stack. And what jumps out at me here is that this is really GEO for e-commerce primarily. Um, Bazaarvoice is a platform that can capture reviews and then allow brands to, um, use that strategically, use those reviews strategically for marketing purposes.
[00:08:55] So ninety percent of, of the surveyed consumers [00:09:00] want to know that what AI is recommending is sourced from real reviews, and that means that those AI engines will prioritize the discovery and crawling of those reviews. So those reviews have to be structured well for discovery and crawling and ingestion by those AI bots.
[00:09:22] And then we have also the review count, one hundred plus reviews in almost sixty percent of those citations, ninety-two percent with four stars or higher. So if you're an e-commerce company, um, the... probably the top priority for citation building within GEO is going to be reviews. Getting those reviews, of course, you want them high, a number of re-reviews, but you also want those reviews to be as positive as possible, and they need to be structured well for AI bots to quickly discover them and, and crawl and [00:10:00] ingest them.
[00:10:00] And so that is what Bazaarvoice is doing. So Bluefish is routing the visibility gaps and telling Bazaarvoice, "We have these gaps, and we need to get some more reviews in these areas." And then Bazaarvoice is responsible for getting those reviews. A very interesting partnership here. On to the next story.
[00:10:21] Cross-engine citation overlap tops out at just twenty-four percent Rankability analyzed 1,645 AI query page observations across ten platforms, and it found that no platform pair shared more than 24.1% of cited pages. Very interesting. It also found that 55.2% of top ten AI citations did not rank in Google's traditional top ten.
[00:10:54] Rankability explains cautions that these are observed relationships rather than causa- [00:11:00] causal ranking factors. Two things definitely jump out here. Among ten different platforms, a- and when you pair them up in any possible combination, you still could not get to a higher than a 24% overlap in cited pages.
[00:11:15] So the AI engines are all very, very different in how they source and surface citations. Um, there is no one single, um, AI ranking here. Another fact that jumps out in a big way is that more than half, 55% of the top ten citations did not rank at all in Google's traditional top ten. So a lot of people that are still saying that SEO is the same as GEO or, uh, you need to do SEO in order to do GEO, meaning that you need to rank in Google, ideally rank in the, in the first page top ten in order to, uh, really have a good chance at citations.
[00:11:59] Uh, s- study [00:12:00] after study has proven that to be incorrect. It's more about relevance than it is about Google rankings. You do need to be indexed, but you don't need to rank on page one to have a good shot at getting, uh, citations in AI. On to the next story we go. Lower authority earned media still wins AI citations, but differently by engine.
[00:12:25] Stacker tracked eight hundred and ninety-five distributed stories and roughly two hundred and twelve thousand pickups across six AI services. Sites with domain rating below forty still generated one-fifth of Stacker network citations. Claude and Google AI Overviews each cited sub-DR forty domains in one point eight percent of measured responses versus only zero point zero five percent for ChatGPT.
[00:12:57] Citation gains also appear to plateau around [00:13:00] ninety pickups for lower authority domains. All right. We are looking at the relationship here between, uh, probably the, the most durable traditional SEO metric, um, which was kind of the currency for link building and link exchange, which is domain rating. Um, an equivalent would be domain authority, DA.
[00:13:24] Domain rating comes from Ahrefs, but it's-- these are, these are very similar. It has to do with your domain strength as measured by the authority of sites that link to your domain. So a domain rating of forty is kind of a baseline above which is pretty good. Uh, of course, when you get into the seventies and eighties, you're in very elite territory for domain rating.
[00:13:47] Below forty is considered relatively weak. And what this is saying is that domain rating is not a strong predictive factor in citation appearance, and this [00:14:00] is strong evidence of the divergence of classic SEO and today's GEO. You don't necessarily need to get citations. You don't need to get brand mentions in high DR sites in order to be cited.
[00:14:14] Um, one thing that does seem interesting, though, is that ChatGPT, uh, differs quite a lot from Claude and Google AI Overviews in that they're only showing, um, about half of a percentage of citations from less than DR forty sites. So ChatGPT seems to weigh the site authority, the classic SEO site authority metric, more heavily than Claude and Google AI Overviews.
[00:14:41] Um, but it's still worth going out and building citations in relevant sites that may not have a DR of forty-plus All right, next story AI tool recommendations are concentrating around a few incumbents. [00:15:00] FogLift's unbranded buyer intent benchmark found that-- found Otterly in fifty-eight percent of usable answers, Profound in fifty-four percent, Semrush in forty-nine percent, Peak and H- Ahrefs in thirty-six percent each.
[00:15:21] Its September checkpoint also found thirty-three point three percent of the leading citation source layer came from off-vendor sources such as YouTube, Reddit, SitePoint, and G2. All right. There's two things happening here, and we are talking about AI tool recommendations here. Um, similar to our first story about the market-- relative market share of Profound and Peak and AirOps and the others.
[00:15:49] But here this is AI tool recommendation, so it's a little bit broader, and we have now Semrush and Ahrefs in the mix. And we also have Otterly in the mix, which did not appear [00:16:00] in that first story in Ramp's market share estimates. So Otterly appeared actually number one here in fifty-eight percent of usable answers.
[00:16:10] So what I'm wondering here is: what were the prompts that were used in this study? Um, it's unbranded buyer intent benchmark for AI tools. So Otterly is one of the lowest cost providers. They appeared in fifty-eight percent, followed by Profound, fifty-four percent Semrush, Peak, and Ahrefs. Um, but then of course, one-third of all the citations came from non- non-brand sources.
[00:16:38] So YouTube, Reddit, SitePoint, and G2. Uh, this is very consistent with what we're seeing. YouTube, especially within Google AI Overviews, is by far the number one citation source. It suggests that one of the strongest things you can be doing to build citations for AI search is to be [00:17:00] creating and publishing YouTube videos.
[00:17:02] Reddit, of course, has always been there, uh, is a partner of ChatGPT and, and very prominent, especially on ChatGPT. SitePoint, I'm not sure about. Uh, G2, uh, just announced recently a partnership with ChatGPT, a data partnership, which is gonna mean that G2 reviews are gonna have even stronger presence and will be serving as a solid source for all B2B SaaS-related, uh, AI search.
[00:17:30] So I think if you're a B2B SaaS company, uh, you want to be leaning in heavily to YouTube videos, Reddit presence, and G2 reviews. I think those are the top three citation activities, and maybe look at li- listicle inclusion as well. On to the next story. Ahrefs doubles a brand radar prompt data set And adds entity extraction.
[00:17:56] Ahrefs on September seventeenth announced that its [00:18:00] Brand Radar People Also Ask prompt data set expanded from roughly fourteen million to thirty million prompts. While AI answers now include entity extraction and a top entities widget, it also introduced an optional traffic estimation model that accounts for search engine results page features, including AI overviews.
[00:18:24] All right. This is an example of a giant SEO platform flexing its muscles, showing that, uh, it can increasingly win on raw data scale and bundled workflows. I personally don't use Brand Radar, even though our agency has an Ahrefs account. We're still using it primarily for more traditional SEO work like, uh, keyword research and SERP analysis.
[00:18:54] Uh, I, I haven't really figured out the Brand Radar product. Um, one of the big weaknesses [00:19:00] in my opinion is that you can't go in and specify which prompts that you wanna track. You can only go in with your, a topic or your category and just look at broad visibility metrics that come from their gargantuan data set of prompts.
[00:19:16] Um, but to me that's not really good enough and I don't see Ahrefs really winning this transition from SEO, SEO tool to GEO tool. They are clearly a number, number one or number two in SEO tools right there with Semrush. Although I just don't see them making the crossover the way that, uh, in the territory where say, Profound is right now.
[00:19:43] So we'll see how this develops but, but clearly they have, um, massive infrastructure and massive data sets that they can leverage that other, other players just don't have. Next story. Query, QueryLift raises one hundred million Japanese [00:20:00] yen to build a Japan-focused GEO platform. Tokyo-based QueryLift raised a hundred million yen in seed funding from JAFCO and Z Venture Capital.
[00:20:11] The company monitors mentions, citations, comparisons, and recommendations across ChatGPT, Gemini, Claude, and Google AI Overviews, then feeds observations into content improvement workflows. Funding will support natural language processing and information retrieval R&D, plus enterprise sales. All right. The first major player that we have seen or that I've read about in this space coming out of Japan, QueryLift.
[00:20:38] So more and more we're seeing, uh, not only regional, but company specific players emerge in this space, and this is a, a major announcement coming from Japan. All right, let's move now to our last story of the day. Radar turns predictive intelligence into authority-- into an authority-building engine.[00:21:00]
[00:21:01] RadIntel launched Radar on September seventeenth, combining predictive audience intelligence with original reporting, executive perspectives, research, and multimedia content. The company explicitly frames the platform across GEO, AEO, and SEO and brand-own-owned channels and plans to offer it to select outside brands and agencies.
[00:21:26] All right, this is a different twist. It's an interesting twist because they, uh, the, the company here, uh, RadIntel has all this audience intelligence, predictive audience intelligence, and they're leveraging that to generate content which is reporting original reporting content around this audience intelligence, including executive perspectives, their unique research, and multimedia content with the goal, not the only goal, but with one stated goal of [00:22:00] influencing GEO and AEO.
[00:22:03] So that's a different, uh, that's a different approach than, let's say, tracking prompts and then trying to, uh, generate content that can close gaps in share of voice and AI visibility among a prompt set. Here they're deriving, uh, they're deriving their content strategy, uh, not from any sort of prompt tracking, but from predictive audience intelligence.
[00:22:25] So that is an interesting play and something that is gonna be worth following. All right. That will do it for this Friday, September eighteenth, episode thirty-six, and hope to see you all in the next one.
