Marketing teams have spent the past two years chasing an unmeasurable target: how visible their brand is inside ChatGPT, Gemini, and the other AI systems that now answer questions their customers used to type into Google. On August 3, the Interactive Advertising Bureau published the industry’s first attempt at a common yardstick for that question, and it arrives at a moment when more than 20 vendors are already selling AI visibility dashboards that rarely agree with each other.
A market with too many rulers
IAB’s new guidance, “Measuring Visibility in the AI Era,” does not endorse a tool or crown a winner. Instead it gives brands, publishers, and agencies a shared vocabulary for a question that has become urgent: is a brand’s content showing up in AI answers, and is it showing up accurately? The bureau built the framework with a working group that included measurement and marketing operations specialists from Walmart, Acxiom, Microsoft, WPP Media, eMarketer, and Tinuiti, spanning both the brand and agency sides of the business.
The trigger is a familiar one in ad tech: fragmentation. IAB found more than 20 companies now sell AI visibility measurement, and their methodologies produce different answers for the same brand asked the same question. Without a shared standard, a marketing leader comparing two vendor reports has no way to know whether a discrepancy reflects reality or just a different counting method.
The four Ps, in order
The framework organizes visibility into a causal hierarchy it calls the “4 Ps,” and the sequencing matters: each layer only counts if the one before it holds up.
Presence
Does the brand or publisher show up at all? IAB tracks this through mention rate, citation rate, share of voice against competitors, and visibility momentum over time.
Prominence
Showing up is not the same as showing up well. Prominence measures where a brand lands in an AI response, whether it is featured or buried in a generic list, and how it ranks relative to competitors.
Portrayal
This is the accuracy layer, and IAB frames it as the one that should worry marketers most. Portrayal covers sentiment and framing, and, critically, hallucination and factual accuracy rates: whether the AI system is describing the brand correctly at all.
Persuasion
The final P ties visibility back to business outcomes: recommendation strength and the click-through rate after a citation. For publishers losing referral traffic to AI summaries, this is the metric that determines whether being cited is worth anything.
Two tiers, not one number
IAB also drew a line between “directional” measurement, useful for internal briefings but built on small query samples, and “decision-grade” measurement, which requires sufficient sample size, query volume, prompt coverage, testing cadence, and data validation before it should influence budget. The bureau set a floor: anything under 50 queries counts as exploratory and, in its words, cannot meaningfully characterize a category. That is a direct challenge to vendors currently selling snapshot reports off a handful of prompts as if they were strategic guidance.
What it means for the marketing leader
The stakes IAB is responding to are not abstract. McKinsey has estimated that brands slow to adapt to AI driven discovery could see traffic declines as steep as 50%, and IAB’s own research found only 16% of brands are systematically tracking AI search performance today. That gap between exposure and measurement is exactly what this framework targets.
For a CMO evaluating vendors, the framework gives a checklist that was missing before: ask any AI visibility tool which of the 4 Ps it actually measures, whether its sample sizes clear the decision grade bar, and how it defines a “citation” versus a passing mention. A vendor that cannot answer those questions in IAB’s vocabulary is still selling directional data as if it were decision grade, and budget decisions built on it deserve extra scrutiny.
It also reframes the measurement conversation for search and content teams already tracking rankings and share of voice in traditional search. The 4 Ps map cleanly onto SEO instincts: presence and prominence will feel familiar, but portrayal is new territory. Brands now need to actively monitor whether AI systems are describing them accurately, not just whether they appear at all.
What to do next
Marketing leaders do not need to overhaul their stack this quarter, but they should start asking their current AI visibility vendor, or the ones they are evaluating, to map its reporting to IAB’s four layers. Teams that already run structured citation audits for SEO have a head start: extend that discipline to portrayal and accuracy checks, since a wrong answer read aloud by an AI assistant does more damage to a brand than a low search ranking ever did. IAB has not endorsed specific tools, and it likely will not soon, given how fragmented the vendor landscape remains. That makes the framework itself, not any single vendor’s dashboard, the reference point worth building internal reporting around this year.
The publishers in IAB’s working group have their own reason to care. As publishers experiment with new inventory built for AI agents, they need the same visibility instrumentation to know whether their content is even reaching those agents in the first place. And the accuracy stakes IAB is flagging echo a problem this publication has tracked before: no AI search engine has produced a reliable, consistent brand leader across repeated queries, which is precisely the inconsistency this framework is built to expose.
Source: IAB
