IAB's New AI Visibility Guidelines: What They Mean for Your Brand

The Interactive Advertising Bureau published its first industry-wide guidelines for measuring AI search visibility on 3 August 2026. The framework, called "Measuring Visibility in the AI Era", gives brands a shared vocabulary and a two-tier quality standard for judging whether the AI visibility data they buy is actually reliable. It matters because more than 20 vendors now sell tools that measure the same brand differently, and until now there was no agreed way to tell which numbers to trust.

What changed?

On 3 August, the IAB released Measuring Visibility in the AI Era, which it describes as the industry's standardised set of guidelines for tracking brand and publisher visibility in AI-powered discovery platforms, designed to help organisations evaluate measurement approaches with confidence.

The 36-page document was built by an IAB working group with representation from Walmart, EMARKETER, Acxiom, WPP Media, Microsoft Clarity, PMG, emberos, the Alliance for Audited Media, Tinuiti and IQRush.ai, according to ppc.land's coverage. As that outlet notes, a market has formed around a question nobody had agreed how to answer: more than 20 companies now sell tools claiming to measure how brands and publishers appear inside AI-generated responses, and those tools use different query sets, different platform coverage, different scoring rubrics and different definitions of the events they count.

The centrepiece is a new metrics hierarchy the IAB calls the "4 P's of AI Visibility": Presence, Prominence, Portrayal and Persuasion, as reported by MediaPost. Playwire's breakdown explains the first layer plainly: Presence is simply how often your content is cited in AI search responses, the baseline that everything else builds on. The framework then moves up through Prominence and Portrayal to Persuasion, which asks whether that visibility actually drives action. As the IAB's own release puts it, Persuasion asks whether AI visibility drives action, with metrics including Recommendation Strength and Post-Citation Click-Through Rate, which bridges to the IAB's forthcoming attribution framework.

Just as important is a two-tier quality classification designed to help brands and agencies determine whether the data they're buying is reliable enough for its intended use. "Directional" measurement identifies patterns and supports early signal detection and competitive awareness, but isn't sufficient for budget allocation or executive strategy decisions. "Decision-grade" measurement meets a stricter bar. Per AdExchanger's reporting, the IAB classifies anything below 50 queries as "exploratory", a step below even directional measurement, because fewer than 50 prompts "cannot meaningfully characterize a category".

Crucially, the IAB is careful not to overclaim its own authority here. IAB VP of AI Caroline Giegerich told AdExchanger that the guidelines are deliberately not called a "standard": a standard requires stability, and right now marketers are in a "mass transition space", with the ad industry unable to settle on fixed standards until AI search behaviour itself becomes more consistent.

Why does this matter for your business?

For years, brands trying to track their presence in ChatGPT, Perplexity, Gemini and Google's AI Overviews have had no consistent way to compare vendors or interpret their own numbers. Two tools could analyse the same brand in the same week and produce completely different scores, with no way to know which one to believe.

That confusion has real commercial consequences. Giegerich told MediaPost: "We are trying to give companies a good idea of what good looks like." Without a shared vocabulary, marketing leaders have been forced to make budget and strategy decisions on data they had no way to sanity-check.

The framework also has direct implications for publishers, not just brands. Playwire's analysis argues that Persuasion is the most critical metric, because it tracks whether AI citations actually drive clicks back to a site. Presence alone, being cited without traffic following, doesn't pay the bills.

Not everyone thinks the framework goes far enough. One critical response calls the document genuinely useful in a narrow sense, but argues it is, by its own admission, a framework for a discovery model that no longer describes how AI-native consumers actually behave, according to aivojournal.org. That's a fair challenge: a measurement standard is only as good as the behaviour it measures, and AI search interfaces are still changing month to month.

What should you do now?

First, stop treating every AI visibility report as equally trustworthy. If a report is based on fewer than 50 prompts, the IAB's own guidance says to treat it as exploratory at best, not something to base a content or budget decision on.

Second, ask any vendor you use, or are considering, to disclose their query volume, sample size, testing cadence, reproducibility and platform coverage. Machine Relations' analysis notes that decision-grade measurement meets standards across query volume, sample size, prompt coverage, testing cadence, reproducibility and platform coverage, producing data that can actually support operational decisions. That's the bar to hold providers to.

Third, use the 4 P's hierarchy as a diagnostic, not just a scorecard. Knowing you have Presence, that you get mentioned at all, tells you far less than knowing your Portrayal, how accurately you're described, or your Persuasion, whether citations lead to clicks or signups. If you're strong on Presence but weak on Persuasion, the fix is different content than if you're invisible altogether.

Finally, this is a good moment to get a baseline reading of where you actually stand. Checking your own AI visibility with Sited's free audit at https://sited.online gives you a starting point before you compare vendor claims against the IAB's new quality tiers.

Frequently asked questions

What is the IAB's "Measuring Visibility in the AI Era" framework?

It's a set of guidelines published on 3 August 2026 that provides shared vocabulary, quality criteria and disclosure requirements for AI visibility measurement, without prescribing specific tools or ranking individual providers.

What are the "4 P's of AI Visibility"?

They are Presence, Prominence, Portrayal and Persuasion, a hierarchy running from simply being cited by an AI system through to whether that citation actually influences a buying decision or click, as outlined by MediaPost.

Is this a mandatory industry standard?

No. The IAB's own VP of AI told AdExchanger the guidance is deliberately not called a standard, because a standard requires stability, and marketers are currently in a "mass transition space".

Why does the directional versus decision-grade distinction matter?

Because a lot of AI visibility data on the market today is based on small sample sizes. The guidelines state that data based on fewer than 50 queries "cannot meaningfully characterize a category", so it shouldn't be used to justify major spending decisions.

Does this framework address how AI-native users actually search?

That's a live debate. Critics argue the framework, while useful for comparing vendors, is by its own admission built around a discovery model that no longer describes how AI-native consumers actually behave, so it's best treated as a starting point rather than the final word.