The IAB Gave AI Visibility a Common Language. Brands Still Need to Turn It Into Decisions.

TL;DR: The IAB has published its first framework for measuring AI visibility, giving the industry a shared hierarchy (Presence, Prominence, Portrayal, Persuasion) and a clear line between directional signals and decision-grade measurement. A shared vocabulary is only a start, though: McKinsey finds only 16% of Fortune 500 brands track AI search systematically, and for multi-location brands, a healthy-looking average can hide where you're losing, market by market. Jenette Simisky, VP of Product Management at Yext, explains how AI visibility tracking becomes something you can act on: measure consistently, benchmark against local competitors with Scout, fix the gaps through the Knowledge Graph, and measure again.

---

Marketing teams have spent two decades getting good at measuring search. Rankings, impressions, clicks, traffic, conversion: a shared vocabulary everyone trusted.

Then, AI search has dismantled the whole thing.

A customer can now ask ChatGPT, Gemini, or Perplexity a question, get a single synthesized answer, make a decision, and never visit your website at all.

That leaves marketers with a hard question and no standard way to answer it: did we even show up? For a long time, the answers came from more than 20 different vendors, each measuring AI visibility in its own way. McKinsey found only 16% of brands track AI search performance systematically, even as those who fall behind risk losing almost half their traffic.

This month, that started to change. The IAB published its first framework for measuring brand visibility in AI-driven platforms: a common language for a discipline that, until now, everyone was inventing on their own. It's a meaningful step, moving AI visibility from “something we should keep an eye on” to “something we can systematically measure, manage, and improve.”

To unpack what the framework actually changes for brands, and where the harder work still lies, we sat down with Jenette Simisky, VP, Product Management at Yext, who works with brands on exactly this problem every day.

1. The IAB just published its first framework for measuring AI visibility. Why does this matter right now, and what problem is it solving for brands?

The biggest thing the IAB framework does is validate that AI visibility is becoming a real marketing discipline, not an experimental metric.

For years, brands have had a relatively established playbook for measuring search. You had rankings, impressions, clicks, traffic, and conversion. AI search fundamentally changes that journey. A consumer can ask ChatGPT, Gemini, Claude, or Perplexity a question, get an answer, make a decision, and never visit your website.

That creates a huge measurement gap. Brands need to know: Did I show up? How prominently? What did the AI say about me? Did it recommend me over my competitor? And, critically, what can I do to influence the answer?

Until now, vendors have been answering those questions in very different ways. The IAB identifying more than 20 AI visibility measurement providers with little consistency in methodology or results really highlights the problem.

Creating a common language around AI visibility is important because it gives marketers a framework for separating an interesting data point from something they can actually use to make a business decision.

And for me, that's the important shift: we're moving from “AI search is something we should watch” to “AI visibility is something we need to systematically measure, manage, and improve.”

2. Can you walk through Presence, Prominence, Portrayal, and Persuasion in plain terms, and why they work as a hierarchy rather than four separate metrics?

I actually like the hierarchy because it mirrors how a consumer experiences an AI answer.

• Presence is: Are you there? When someone asks an AI engine a question relevant to your business, does your brand appear at all? Are you mentioned or cited, and how often, compared to your competitors?
• Prominence is: If you're there, where are you? Are you the first recommendation? One of several brands mentioned? Are competitors consistently being surfaced ahead of you?
• Portrayal is: What is being said about you? Being visible isn't automatically positive. AI might describe your brand using outdated information, emphasize negative themes, or simply frame a competitor more favorably. Sentiment, accuracy, and context all matter.
• Persuasion is: Does the answer actually make someone want to choose you? Ultimately, visibility should influence a consumer decision. Is the AI engine simply mentioning your brand, or is it actively recommending you?

The hierarchy matters because you can't skip steps. You can't influence a customer through AI if you're absent from the answer. And showing up isn't enough if you're buried below competitors or being portrayed negatively.

That is why I think brands ultimately need to look at AI visibility as a system rather than obsessing over one metric.

3. What's the practical difference between directional and decision-grade measurement, and how should brands know which one they're looking at?

Directional measurement tells you where to look. Decision-grade measurement gives you enough confidence to decide what to do.

Directional data can be incredibly useful. It can tell you that you're appearing less frequently than a competitor, that sentiment around your brand is changing, or that you seem to perform differently across AI engines.

But if I'm going to shift budget, change my content strategy, or roll out an optimization strategy across thousands of locations, I need much more rigor behind that conclusion.

That's where things like sample size, prompt coverage, geography, testing frequency, reproducibility, and competitive context become really important.

One question I'd encourage every marketer to ask their AI visibility vendor is: What exactly is behind this number?

How many prompts were tested? Which prompts? Which models? How frequently? For what locations? Are the prompts branded or unbranded? How are competitors selected? Can I reproduce the result over time?

A beautiful dashboard doesn't necessarily mean you have decision-grade data.

4. McKinsey found only 16% of brands track AI search performance systematically, even though laggards could lose half their traffic. Why is there such a gap?

Because the behavior changed faster than the current measurement infrastructure.

Marketing organizations spent decades building systems around a search journey that was fairly predictable: query, results page, click, website. AI breaks that model. Now the answer itself is becoming the destination.

At the same time, AI responses are probabilistic. You can ask a similar question twice and receive different answers. Results can vary by model, prompt, location, and context. So applying traditional SEO measurement to AI search doesn't really work.

There's also a practical issue: a lot of brands don't know where to start. They're hearing about GEO, AEO, AI visibility, and prompt optimization, but the conversation can quickly become very theoretical.

I think the opportunity is to simplify this for marketers. Start with three questions:

1. How am I performing?
2. How do I compare to my competitors?
3. What should I do next?

If measurement can't ultimately help you answer that third question, it has limited value.

5. The IAB calls out competitive awareness as a primary use case for directional measurement. How does Yext help brands see where they stand against competitors in AI responses today?

This is an area where Scout is particularly differentiated.

Scout doesn't just tell a brand whether it appeared in an AI answer. It benchmarks visibility against the competitors consumers are actually seeing, across AI and traditional search, and does that down to the local level.

That local context matters enormously for multi-location brands. The competitive set for a healthcare provider in Boston might be completely different from the competitive set for that same health system in another market. The same is true for financial services, restaurants, retail, automotive — really any distributed brand.

Scout looks across branded and unbranded queries and tracks signals like AI win rate, share of voice, rank, citation sources and volume, sentiment, and performance over time. Brands can drill from the overall brand down into individual markets and locations to understand exactly where they're winning and losing.

And then we go beyond the metrics.

We can start answering why a competitor is winning. Maybe they have stronger review volume. Maybe their local pages have schema markup. Maybe their structured data is more complete. Maybe the sources AI engines are citing contain better or more consistent information about them.

That's the difference between monitoring AI and actually building an AI visibility strategy.

6. If decision-grade measurement is where the industry is heading, what's Yext's role in getting brands there, and what should they be doing to prepare?

I think getting to decision-grade measurement requires consistency, scale, and actionability.

Consistency comes first. AI responses are inherently variable, so you need a controlled methodology to make meaningful comparisons over time. With Scout, we intentionally use a consistent set of questions and repeat that measurement. If you're constantly changing what you're asking while the AI answers are also changing, it's very difficult to know whether brand performance actually improved or you're simply measuring something different.

Then you need scale. And when I say scale, I don't necessarily mean asking thousands of different questions. I mean generating enough observations using that consistent methodology (across locations, AI engines, competitors, and over time) that you can distinguish a real pattern from the natural variability of generative AI.

That's particularly important for enterprise and multi-location brands. A brand-level average can look healthy while masking significant differences by market. At scale, you can start to understand where you're consistently winning or losing, which competitors are showing up instead, and whether those patterns are actually changing over time.

And finally, the measurement has to be actionable. The goal isn't another dashboard. It's understanding what is driving your visibility and what you can do to improve it.

That's where I think Yext is particularly well-positioned. Scout can identify where a brand is winning and losing, and Yext can connect those insights back to the digital presence that influences AI answers in the first place. If the opportunity is related to business information, reviews, local content, or other visibility signals, brands can take action through Yext and then measure again.

Ultimately, that's the loop we're trying to create: measure consistently, observe at scale, understand what's driving the outcome, take action, and measure again.

7. Anything else to add?

One thing I would add is that AI visibility isn't really a new channel to manage in isolation. It's an outcome of your entire digital presence.

AI engines are synthesizing information from your website, listings, reviews, social content, and many other sources. You can't optimize one page for ChatGPT and declare victory.

That's also why I think Yext is uniquely positioned here.

We have spent years helping large, complex brands manage the underlying information about their businesses across the digital ecosystem. Now Scout gives those brands the intelligence layer to understand how that information is being interpreted by AI and traditional search — and how they compare to competitors. Yext's Knowledge Graph and brand visibility products then give them a way to act on what they learn.

The other differentiator I'd emphasize is local intelligence.

There are a lot of tools that can tell a CMO, “Your brand appeared in 37% of these prompts.” That's interesting.

But if you manage 5,000 locations, the more valuable questions are: Where are we losing? Who are we losing to? Why are they winning? What should we fix first? And can we do it at scale?

That's the problem we're focused on with Scout.

The IAB framework is an important step because it gives the industry a common vocabulary for measuring this new world. The next step is turning those measurements into something marketers can actually use.

For me, that's where this gets exciting: AI visibility shouldn't just be measurable. It should be manageable.

The bottom line

The IAB framework matters because it finally gives the industry a shared way to talk about AI visibility: a common vocabulary where there were 20 competing ones. But a vocabulary is a starting point, not a finish line. The brands that pull ahead won't be the ones with the most dashboards. They'll be the ones who can answer the third question, “What should I do next?”, and then prove the action worked.

That's the shift worth internalizing: AI visibility shouldn't just be something you measure. It should be something you manage: measure consistently, observe at scale, understand what's driving the outcome, act on it, and measure again. The framework tells you what to look at. Closing the loop is what turns a number you watch into a result you own.

AI visibility should be manageable, not just measurable. See how Yext helps brands close the loop.