AI Visibility Starts With Understanding the Conversation
The customer journey moved into a room we can't see into
For twenty years, search marketing gave us a window into demand. By tracking keyword volume, impressions, rankings, and click-through rates, we had an idea of what people were looking for — and at — online, and where they were going in order to find it. The data was always somewhat incomplete, but at least it was visible.
Artificial Intelligence changed that. When a customer asks a tool such as ChatGPT, Gemini, Copilot, or Perplexity for a recommendation, we can’t see the questions. We don't see the volume or context, and we can’t run analytics against them. Further, these AI interactions are complex, evolving through follow-up questions and discussions that shape customers' decision-making in ways we can neither track nor predict.
These AI conversations are responsible for an increasing share of site traffic, at the expense of traditional search. In their efforts to compensate, many brands are responding by chasing tactics. Schema tweaks. Citation counts. The latest “GEO hack.” The problem is, they are trying to optimize performance in AI-powered and generative seach without understanding what actually influences it.
The first step is to understand that AI and generative search aren’t a transaction that responds to tactical levers. Instead, they’re a conversation. At Adcetera, our thinking about AI visibility is built on three core questions.
What Are Users Actually Asking AI?
This is where every GEO program should begin, and where most brands stumble. The instinct — often with encouragement from AI search tools — is to take a keyword list, upload it to the tool, and call it a prompt sample. But keywords aren’t questions. “Lawn tractor” is a phrase that tells you nothing about who’s asking or what they’re trying to decide.
Rather than guessing at prompts, we model them. And instead of starting with the keyword, we start with the customer. What decision are they trying to make? What questions would they ask at different stages of their decision-making process? Classic search data is one input, but so are subject-matter discovery, audience research, competitive research, conversations with sales, site analytics, support tickets, and more.
We’re not looking to find every possible prompt. We use customer and market insights to gain an understanding of the conversations customers are actually having, and model prompts accordingly.
Are We in the Conversations?
Once you know the conversations that matter, you can measure whether you're in them and what part you play in the conversation. Many brands consider this to be "AI visibility," and it's important. But be precise about what you're measuring. Being mentioned, being cited as a source, and being recommended are three different metrics, and they mean different things.
You can be mentioned without being recommended. You can be cited without being named. What we care about most is recommendation, because that's more likely to move a decision.
The uncomfortable truth is your competitors may be winning in conversations you didn’t even know existed — even with strong traditional SEO, a healthy website, and solid web rankings. If customers are relying on AI answers that overlook you, you’re not measuring the right conversations. And you won’t even realize they’re happening.
Are We Considered a Solution?
This is where measurement becomes strategy. One of the most valuable things you can study isn't your own performance. It's the ways other are performing better. When AI recommends a competitor instead of you, it's telling you something, and you need to pay attention.
The underlying question is simple: where does authority ultimately live? AI systems build answers from sources they trust, and your site’s commercial intent will always present a headwind to trust. AI tends to grant more trust to publishers, government sites, industry resources, analysts, and community sites.
But what you do on your own site is still critical. With your understanding of key customer conversations in hand, you must consider how your content positions you. Are you participating in these conversations? Is your content easy to understand, and easy to cite? Outreach to, and engagement with, third parties will open the door, but your own content helps determine if AI sees you in the conversation.
Framework, Not Tactics
We don’t see AI visibility as a technical or tactical challenge. There’s no algorithm to crack, and no magic toolset that will secure performance. Instead, we believe AI visibility is a test of marketing fundamentals. You need to do what good marketers have always done: listen closely, understand your customer, and show up where it counts. The channel may be new, but your instincts are still sound.
About Adcetera
Adcetera is a digital agency based in Houston. We build content strategies that help brands get found, understood, and recommended, across search and AI. If you're ready to understand the conversations shaping your category and claim your share of the empty space, let's talk.