The GEO Trust Gap: SEOs Want The Data, But Not The Platforms Selling It via @sejournal, @DuaneForrester
All Paid Media PPC News Social MediaAdvertising Video Advertising Columns Ask A PPC ExpertNEW PPC Pulse Webinar Google Local Services Ads Are Moving To PMax: What To Check First Prepare for Google's LSA move into Performance Max with a before-and-after audit checklist from CallRail. Digital Marketing All Digital Marketing Analytics & Data Ecommerce Lead Generation Content Marketing Social Platforms Google YouTube Reddit LinkedIn TikTokNEW WordPress Other CMS Platforms Webinar Google Local Services Ads Are Moving To PMax: What To Check First Prepare for Google's LSA move into Performance Max with a before-and-after audit checklist from CallRail. SEJ Live Webinars Rundowns PodcastNEW Ebooks All Resources Rundown Why Your Brand Drops Out of AI Answers Five articles on the signals behind citation fluctuation, where models source citations, and how to audit your content for AI visibility gap Advertising On SEJ Case Study: B2B SaaS Banner Ads Webinar AI Cites Your Brand. Now What? Turn AI Visibility Data Into Actions Turn your AI visibility reports into a prioritized AEO action plan with practical guidance from Ahrefs. Not a pricing problem. 163 respondents value the data and won’t fund the tools that produce it. Duane Forrester 34 seconds ago ⋅ 10 min read Duane Forrester Founder and CEO at UnboundAnswers.com Bio Follow Over three weeks in July, I ran a survey asking people who work on AI search visibility what they think of the platforms built to measure it. 163 responses. This is a self-selected sample recruited through my own network and its re-shares (as well as paid ads on LinkedIn and X), so it describes engaged practitioners in and around one corner of the industry, not the entire industry. Percentages here carry roughly a seven-point margin, and I’ll come back to the sample size at the end, because it turned out to be part of the story. I asked people to rate how valuable various kinds of AI visibility data would be: Query alignment beyond just keywords. Competitor comparison. Whether a mention comes from training or retrieval. Chunk-level attribution. Citation status. Then I asked whether investing budget in a dedicated platform for this feels worthwhile right now. That’s the survey in two numbers. A third of respondents rated the data highly valuable and platform investment lukewarm or worse, in the same sitting, minutes apart. Only 44% think buying a tool in this category is worthwhile at all. Nearly a third rate it 1 or 2 (5 being highest). I collected the data in four snapshots as responses came in: 36, then 75, then 100, then 163. Neither number moved more than a tenth of a point across the whole run. Whatever this is, it isn’t a sampling artifact. It settled early and stayed put while the sample quadrupled. 123 people (75%) wrote something in the open text box. I asked what their biggest unanswered question was, or their biggest issue with the platforms they’d tried. I expected a feature request list. That isn’t what I got. (Many people raised more than one issue, and each was counted under every theme it touched, so these add to more than 100%.) The texture of responses matters more than the counts, I think. Several people described the same structural problem with prompt-list tracking: you choose the prompts, which means you decide in advance what you should be visible for, then measure yourself against your own list. One called it a self-fulfilling prophecy. Related, and sharper: these tools have no denominator. Scores come from invented prompt lists rather than observed query volume, so ordinary model variance gets reported to a client as a win or a loss with nothing underneath it to say which. One respondent argued citation tools are a fundamentally different animal from rank trackers: you can’t reverse-engineer what’s working when the answer changes every time you ask, so what you’re left with is closer to a brand awareness signal than a diagnostic. An agency running 50-plus clients laid out the squeeze plainly: they can’t sell AI visibility as a service without measurement tools, and can’t justify the tools until they’re selling the service. One person with two decades in the industry framed it more evenly than anyone else: Third-party SEO data was always directional rather than gospel, and that’s fine, as long as nobody pretends otherwise. And the sharpest one, aimed straight at vendors: That you can’t do what these tools purport to do, because every user of every model gets a different experience. Snake oil. Magic beans. I’m not going to argue with any of that here. It’s what practitioners said when asked, and the value of asking is diminished if I spend the space explaining why they are wrong. 7% raised cost. 57% raised either “I don’t believe the number” or “I can’t connect this to money.” That ratio is the most useful thing in the survey. Whatever is holding this category back, the answer is not that the tools are too expensive. Reading individual answers gives you complaints. Cross-referencing them gives you something else, and three patterns held up when I tested them. People who raised trust concerns value the data exactly as much as everyone else, and would spend exactly as much. Their rating of the underlying data value: 4.20, against 4.19 for everyone else (a difference of one hundredth of a point). Their budget: statistically indistinguishable. But their willingness to invest in a platform drops to 2.76 against 3.36 for everyone else, and they’re markedly less likely to be paying for anything. Same valuation. Same money available. Different conclusion. Whatever is blocking this segment, it isn’t what the data is worth to them, and it isn’t what they can afford. It’s whether they believe it. Almost nobody is building the alternative. 8% of respondents built their own tooling. Among the people who raised trust or non-determinism, 9%. Among the 78% who call accuracy essential in a vendor, 6%. I don’t read this as hypocrisy. Building this is genuinely hard (I know!), and most practitioners have a job that isn’t engineering, but it does reframe the objection. “The numbers can’t be trusted” isn’t functioning as a diagnosis anyone acts on. It’s a request for someone else to solve it properly. The gap between valuing the data and funding a platform is identical across every role. Agencies, in-house teams, independent consultants are all within a rounding error of each other. It isn’t agencies being cheap or in-house teams being spoiled. It’s the whole market. I can’t tell you which direction that runs. Buying may resolve the doubt, or people without the doubt may be the ones who buy. The survey can’t distinguish those, and I’m not going to pretend otherwise. But it’s the single largest split in the dataset, and it suggests the objection looks different from inside a subscription than outside one. Ranked by share rating each 4 or 5: query alignment beyond just keywords, 90%. Competitor comparison on the same query, 83%. Whether a mention comes from training or retrieval, 83%. Chunk-level attribution, 75%. Citation status, 71%. On which systems matter: Google’s AI Overviews and AI Mode at 95%, ChatGPT at 94%, Gemini 75%, Claude 64%, Perplexity 34%, Copilot 25%. Nothing else cleared 5%. I capped that question at five selections, and 46% of respondents used all five, so treat those as floors. 87% describe their practice as actively working on this or established in their work. This is not an audience that needs convincing the problem is real. The most-raised objection was methodology opacity: Show me where this data comes from and why I should believe it. It’s a reasonable thing to want. It’s also, as stated, not a thing any vendor in this category can give you. Now, I should remind everyone that I built one of these platforms. That’s a conflict, and you should read what follows knowing it. It’s also why I have a view on what vendors can and can’t disclose, as I’ve had to make that call myself.
Source: Search Engine Journal
This article has been carefully curated and reformatted for educational and informational purposes. Full credit goes to the original publisher.
📚 Visit more helpful articles on Joab Peters Blog
No comments