[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 we're covering something that landed recently and made a lot of marketers sit up: OpenAI appears to be testing a business analytics dashboard for ChatGPT — a kind of Search Console for ChatGPT. My guest is Marcus Reid, industry analyst, ex-Google Ads, and a man with strong opinions about dashboards. Marcus, welcome.

[Guest] Hey everyone. Strong opinions is one way to put it. I've spent a decade watching people fall in love with a number they can't explain, so, yeah. Good to be here.

[Host] Let's start where our listeners actually live. You run a brand. You've been publishing, you've been doing the S.E.O. work, and someone in a Monday meeting says, hey, are we showing up when people ask ChatGPT about us? And you go look — and there's nothing. No dashboard, no clicks, no impressions. You're guessing off screenshots your cousin sent you.

[Guest] It's worse than guessing, because you're guessing with confidence. I've watched teams write entire quarterly plans off three ChatGPT answers they got on a Tuesday. One of them said our product was discontinued. It wasn't. We'd shipped a new SKU that month.

[Host] That's the gap. And there's a name for the work that's supposed to close it. People call it G.E.O. — Generative Engine Optimization — and A.E.O., Answer Engine Optimization. How do you get cited inside an AI answer instead of ranked in a list of links. The problem has always been measurement. You could do the work, but you couldn't prove it happened.

[Guest] And that's what this whole story is about. Not the optimization. The scoreboard.

[Host] So what actually happened? Walk us through it.

[Guest] Early this year, people spotted what looks like a business hub inside ChatGPT — reportedly at chatgpt dot com slash business-hub. There's a banner inviting businesses to quote, claim your business in ChatGPT. The first person to surface it publicly was Juozas Kaziukėnas, the founder of Marketplace Pulse, posting on LinkedIn. Trade coverage followed — Search Engine Land, Search Engine Roundtable, and a cluster of G.E.O.-focused outlets.

[Host] And what does it reportedly do?

[Guest] Three things. You claim and verify your business. You see daily impressions — how often ChatGPT mentions you in answers — and clicks, meaning how many people actually clicked through to your site. And you get control over your products and business details, plus something they're calling a business agent. It's a test, not a launched product. I want to be clear on sourcing: much of the detail rests on screenshots and one researcher's read, not an official OpenAI measurement contract.

[Host] Which matters, because the internet has already written the press release.

[Guest] The internet writes the press release before the product ships. That's the business model.

[Host] Fair. How does it work under the hood — because the measurement is the interesting part.

[Guest] Mechanically it's close to something you already know: Google Business Profile. You claim, you verify, you manage details. Then the dashboard reports two daily events — you appeared in a response, and someone clicked. That's it. The clever bit is on OpenAI's side. By getting businesses to claim and verify, they get clean, structured, first-party data straight from the businesses themselves. Which is exactly what makes shopping answers inside ChatGPT better. You're not scraping a product feed; the merchant handed it over.

[Host] So the analytics is almost the byproduct. The real asset is the verified data pipeline.

[Guest] That's my read. The dashboard is the carrot that gets you to fill in the form.

[Host] Now the criticism, and I think it's fair. This is not Google Search Console. There's no query report.

[Guest] No prompt data, no user data, no competitive benchmarking, no rankings. Nothing that tells you why the number moved. Kaziukėnas put it best, and I'll paraphrase him accurately: it won't help you understand why mentions or clicks are going up or down, but combined with external tools that track prompts, it will show accurate first-party data on whether your G.E.O. and A.E.O. work is landing.

[Host] It's a scoreboard without a play-by-play.

[Guest] And honestly it's a little like Netflix telling you you're top ten in your area. Great. Something happened. No idea what.

[Host] Before this existed, how were people even measuring?

[Guest] Sampling. You'd run a few hundred prompts through a tool, count how often you appeared, and hope the sample held. That's proxy data. First-party click data from the source is a different category of evidence entirely.

[Host]  Here's where I'll push back a little. A thin scoreboard is still a big deal. The reason S.E.O. became a budget line at every company is that Search Console gave you a number you could take to a C.F.O. Before that you had rank trackers and vibes. This gives AI visibility a number. That changes the internal argument.

[Guest] I'll half-agree. The number helps. But look at who's building it. One piece in Aaron's Newsletter framed the whole thing as ChatGPT building its own Search Console, and it's really about ads. The claim flow, the verified account, the daily reporting — that's infrastructure you need before you can sell placement. This isn't a gift to merchants. It's the foundation of a media business.

[Host] I don't think those are in conflict. You can build ad infrastructure and hand merchants better data on the way.

[Guest] Sure. I just don't want people building a strategy on the assumption that OpenAI's roadmap is designed to make their lives easier. I could be wrong. I actually don't know if this holds in six months — whether it ships broadly, gets folded into ads, or quietly disappears the way a lot of ChatGPT plugins did.

[Host] That's fair.  Let's get to what operators should do now, because that's the part that compounds while the dashboard is still a test.

[Guest] Four things, none exotic. First, query fan-out. One prompt from a user spawns a bunch of sub-queries behind the scenes. A question like best running shoes for flat feet turns into several retrieval passes. You need coverage across the whole fan, not one keyword. Second, third-party citations. AI answers lean on sources that aren't you — reviews, listicles, forum threads, trade pubs. If those describe you badly or don't mention you, your own site copy won't save you.

[Host] A lot of brands learned that the hard way with Google AI Overviews.

[Guest] Same muscle. Third, structured data. Clean, machine-readable facts about your products and business. If OpenAI is literally asking businesses to verify their details, the value of being legible to machines just went up. Fourth — the boring one — S.E.O. foundations still matter. Crawlability, speed, internal links, real content. AI recommendations don't come from nowhere. They come from a web a crawler can read.

[Host] And that's the bridge to our world. At A.E.O. Engine, most of what we do with ecommerce and B2B clients is exactly this — making a brand legible to answer engines, not just rankable in blue links. Our content agents run continuously, we do the technical work, but the outcome we're chasing is being the cited answer inside ChatGPT, Google AI Overviews, Perplexity. This news says the measurement layer is finally catching up to the work.

[Guest] And it raises the tooling question. There's a live debate about whether third-party G.E.O. platforms like Peec and Profound get disintermediated by first-party OpenAI data. My take: they don't, because the missing why layer is exactly what they sell.

[Host] Agreed. The prompt tracking is the diagnostic. The first-party data is the ground truth. You need both. And the practical move is simple: claim your business the moment that banner shows up in your market. Treat it like Google Business Profile in 2012 — the people who did the boring verification early got a decade of compounding. Some observers think the claim mechanism is more important than the analytics for exactly that reason: verified identity inside the answer.

[Guest] And remember the shopping piece is US-only right now. If you're outside that, you've got runway to get your house in order before the dashboard is even looking at you.

[Host] Marcus, this was good. Quick recap of the topic: OpenAI is testing a business hub inside ChatGPT that reports mentions and clicks, it's thin on diagnostics, and it's likely the opening move of a much bigger commercial plan. Where the field goes next is measurement getting real on both sides — first-party from OpenAI, prompt-level from everyone else.  If you want the operator playbook for getting cited in AI answers, head to A.E.O. Engine dot A.I. — that's aeoengine.ai. See you next episode.