Back to Blog
The IAB Just Published "Measuring Visibility in the AI Era." Here's What Its 4 P's and Decision-Grade Standard Mean for Your 2026 Measurement Stack.
Todd Paris
CEO
5 min read
If you're a CMO, marketing director, or analytics lead who has spent the past year picking an AI visibility vendor with no standard to judge any of them against, that excuse expired yesterday. The IAB published "Measuring Visibility in the AI Era" on August 3, and for the first time this category has a shared vocabulary, a metrics hierarchy, and a named line between numbers you can act on and numbers you can't.
TL;DR
The IAB released "Measuring Visibility in the AI Era" on August 3, 2026. It defines a common metrics hierarchy for AI visibility, the 4 P's, plus shared vocabulary and disclosure requirements for measurement providers.
The guidance draws a hard line between decision-grade and directional data. Decision-grade means rigorously tested and fit for budget decisions. Directional means trend-spotting only. Below 50 queries, the IAB calls the work exploratory, a tier below directional.
More than 20 vendors sell AI visibility measurement today, and their methodologies do not reconcile. The guidance exists because buyers had no yardstick. Now they have one.
IQRush sits on the IAB AI Visibility working team that developed the framework. The IAB stays vendor-neutral and recommends no tools, ours included. What we got from the room was a standard we're prepared to be graded against.
Your move this week: ask your current vendor which tier their data sits in under the IAB definitions, and ask them to put the answer in writing.
What the IAB actually published
"Measuring Visibility in the AI Era" is guidance, not a formal standard, and the IAB is candid about why. Caroline Giegerich, the IAB's VP of AI, told AdExchanger that standards require a stability this market does not have yet. AI answer engines change weekly. Locking a formal standard onto a moving target would date it before the ink dried. Guidance can move with the market.
What the document gives you is a common language. As Giegerich put it to MediaPost, "We are trying to give companies a good idea of what good looks like." Brands get a playbook for judging whether a provider is delivering reliable data. Agencies get shared benchmarks for client reporting. Publishers get standard metrics for how AI platforms ingest and surface their content, which matters the moment a licensing conversation starts. The full framework is on the IAB site.
Full disclosure before we go further: IQRush is an IAB member and sits on the IAB AI Visibility working team that developed this framework. The IAB is vendor-neutral and does not recommend measurement tools, ours included. Read what follows knowing we were in the room.
The 4 P's, translated into buyer terms
The center of the framework is a metrics hierarchy the IAB calls the 4 P's of AI Visibility. Each one answers a different question a marketer is already asking, and each one deserves a different level of scrutiny in a vendor demo.
Metric | What it answers | What to demand from a vendor |
|---|---|---|
Presence | Does my brand appear in AI answers and citations at all? | How many prompts, how many repeat runs per prompt, and the range around every share number. |
Prominence | Where in the answer do I show up, and how am I treated when I do? | Rank positions carry their own uncertainty on top of share uncertainty. Ask for both. |
Portrayal | Is what the engine says about me accurate, and what is the sentiment? | How they separate a hallucination from a sourced factual error, and whether every claim traces to a stored answer. |
Persuasion | Does any of this drive action, like post-citation clickthrough? | Where the outcome data comes from and how it connects back to the specific answer that drove it. |
Portrayal is the one that should get more of your attention than it will. Giegerich told AdExchanger that factual inaccuracies in AI answers are what "would keep me up at night," and she's right to lose sleep. A wrong presence number wastes budget. A wrong claim about your product in front of a buyer does damage you may never see on a dashboard.
Decision-grade or directional: the line that matters
The piece of this framework that will change vendor conversations is the two-tier data classification. Decision-grade measurement is rigorously tested and fit to move budget and strategy. Directional data is for spotting trends, nothing more. And the guidance puts a floor under the whole thing: fewer than 50 queries is exploratory, a tier below even directional.
Think of flipping a fair coin 50 times. Sometimes you'll get 23 heads. Sometimes 27. Sometimes 30. Same coin, just different runs. The variation isn't because the coin changed. It's because 50 flips is a small sample, and small samples bounce around a lot. AI visibility measurement at 50 prompts works the same way, which is why a lift that fits inside the natural fluctuation band is not a lift at all, just noise wearing a case study.
We have been making this argument since we started IQRush, and I'll be honest about how it went early on: it cost us. Walking into a demo and saying "here is the range, and here is what we cannot tell you yet" is a harder sell than a clean hero number. There were quarters when I wondered if rigor was just a slower way to lose. What changed yesterday is that the category's own trade body put the vocabulary in writing. Decision-grade is now a claim a vendor has to defend, not a phrase on our homepage.
More than 20 companies sell AI visibility measurement today. The IAB wrote this guidance because their numbers do not reconcile.
To be precise about what we can and cannot claim: the IAB developed these definitions with a working group that included measurement providers, brand data-science teams, and agencies. We contributed to that work. We did not write the framework, and the IAB endorses nobody. The test now is the same for every vendor on the market, us included: put your methodology where the guidance is.
Marketer-Side Impact
The industry wastes roughly $2 billion a year acting on AI visibility numbers that don't hold up. The unit of waste is familiar: a $50K content reallocation briefed against a shift that was random fluctuation, an agency team rebuilding a report because two tools disagree, a board meeting where the number on slide 4 can't survive one follow-up question.
The IAB guidance doesn't cost you anything to use. It's a free filter. Run your current vendor's reporting through the decision-grade definition and the 50-query floor, and you'll know within one meeting whether the number you defended last quarter was measurement or theater.
Three questions to take into your next vendor conversation
Under the IAB's definitions, is the data on my dashboard decision-grade, directional, or exploratory, and will you state that in writing?
Which of the 4 P's do you actually measure, and for each number, how many prompts and repeat runs sit underneath it?
When your number and another tool's number disagree, what is your documented method for deciding which one is wrong?
None of these questions require a statistics background. They require a vendor willing to be audited against a public document. A vendor who squirms at question one has answered it.
What happens next
The IAB has said the work continues beyond organic visibility, with paid placement and attribution frameworks on the horizon. Nearer in, the IAB is hosting an AI Insiders panel on August 19 to walk through the framework; registration is open. If you're evaluating vendors this quarter, an hour there is worth more than a stack of pitch decks.
My falsifiable bet: by the March 2027 planning cycle, at least one enterprise RFP you receive or issue will require vendors to classify their data as decision-grade or directional under this framework, by name. When that happens, the vendors who spent 2026 shipping hero numbers will be rewriting their decks. We'll be sending ours as-is.
Frequently asked questions
Is "Measuring Visibility in the AI Era" a formal IAB standard?
No. It's guidance. The IAB's position is that formal standards need market stability that AI search doesn't have yet, so the document establishes shared vocabulary, a metrics hierarchy, and disclosure expectations that can evolve with the market. It's still the closest thing to a yardstick this category has ever had.
What are the 4 P's of AI Visibility?
Presence (does your brand appear in AI answers), Prominence (where and how prominently it appears), Portrayal (how accurately and favorably you're represented, including hallucinations vs factual errors), and Persuasion (whether appearances drive action, such as post-citation clickthrough). Together they form the IAB's common metrics hierarchy for AI visibility.
What's the difference between decision-grade and directional data?
Decision-grade data has been rigorously tested and is fit to support budget and strategy decisions. Directional data is suitable for spotting trends but not for acting. The guidance also defines an exploratory tier below directional: fewer than 50 queries. If a vendor can't tell you which tier their reporting sits in, treat it as exploratory.
Does the IAB recommend specific AI visibility vendors?
No. The IAB is vendor-neutral and recommends no tools. IQRush sits on the IAB AI Visibility working team that developed the framework, which is participation in standards work, not an endorsement. Any vendor implying the IAB endorses their product is telling you something useful about how they handle claims.
What should I do with this guidance if I already have an AI visibility vendor?
Ask them to classify their reporting as decision-grade, directional, or exploratory under the IAB definitions, in writing, and to disclose prompt counts and repeat runs for every headline number. If the answer is a deflection, run your next big decision on a second source before you move budget.
Back to Blog
The IAB Just Published "Measuring Visibility in the AI Era." Here's What Its 4 P's and Decision-Grade Standard Mean for Your 2026 Measurement Stack.
Todd Paris
CEO
5 min read
If you're a CMO, marketing director, or analytics lead who has spent the past year picking an AI visibility vendor with no standard to judge any of them against, that excuse expired yesterday. The IAB published "Measuring Visibility in the AI Era" on August 3, and for the first time this category has a shared vocabulary, a metrics hierarchy, and a named line between numbers you can act on and numbers you can't.
TL;DR
The IAB released "Measuring Visibility in the AI Era" on August 3, 2026. It defines a common metrics hierarchy for AI visibility, the 4 P's, plus shared vocabulary and disclosure requirements for measurement providers.
The guidance draws a hard line between decision-grade and directional data. Decision-grade means rigorously tested and fit for budget decisions. Directional means trend-spotting only. Below 50 queries, the IAB calls the work exploratory, a tier below directional.
More than 20 vendors sell AI visibility measurement today, and their methodologies do not reconcile. The guidance exists because buyers had no yardstick. Now they have one.
IQRush sits on the IAB AI Visibility working team that developed the framework. The IAB stays vendor-neutral and recommends no tools, ours included. What we got from the room was a standard we're prepared to be graded against.
Your move this week: ask your current vendor which tier their data sits in under the IAB definitions, and ask them to put the answer in writing.
What the IAB actually published
"Measuring Visibility in the AI Era" is guidance, not a formal standard, and the IAB is candid about why. Caroline Giegerich, the IAB's VP of AI, told AdExchanger that standards require a stability this market does not have yet. AI answer engines change weekly. Locking a formal standard onto a moving target would date it before the ink dried. Guidance can move with the market.
What the document gives you is a common language. As Giegerich put it to MediaPost, "We are trying to give companies a good idea of what good looks like." Brands get a playbook for judging whether a provider is delivering reliable data. Agencies get shared benchmarks for client reporting. Publishers get standard metrics for how AI platforms ingest and surface their content, which matters the moment a licensing conversation starts. The full framework is on the IAB site.
Full disclosure before we go further: IQRush is an IAB member and sits on the IAB AI Visibility working team that developed this framework. The IAB is vendor-neutral and does not recommend measurement tools, ours included. Read what follows knowing we were in the room.
The 4 P's, translated into buyer terms
The center of the framework is a metrics hierarchy the IAB calls the 4 P's of AI Visibility. Each one answers a different question a marketer is already asking, and each one deserves a different level of scrutiny in a vendor demo.
Metric | What it answers | What to demand from a vendor |
|---|---|---|
Presence | Does my brand appear in AI answers and citations at all? | How many prompts, how many repeat runs per prompt, and the range around every share number. |
Prominence | Where in the answer do I show up, and how am I treated when I do? | Rank positions carry their own uncertainty on top of share uncertainty. Ask for both. |
Portrayal | Is what the engine says about me accurate, and what is the sentiment? | How they separate a hallucination from a sourced factual error, and whether every claim traces to a stored answer. |
Persuasion | Does any of this drive action, like post-citation clickthrough? | Where the outcome data comes from and how it connects back to the specific answer that drove it. |
Portrayal is the one that should get more of your attention than it will. Giegerich told AdExchanger that factual inaccuracies in AI answers are what "would keep me up at night," and she's right to lose sleep. A wrong presence number wastes budget. A wrong claim about your product in front of a buyer does damage you may never see on a dashboard.
Decision-grade or directional: the line that matters
The piece of this framework that will change vendor conversations is the two-tier data classification. Decision-grade measurement is rigorously tested and fit to move budget and strategy. Directional data is for spotting trends, nothing more. And the guidance puts a floor under the whole thing: fewer than 50 queries is exploratory, a tier below even directional.
Think of flipping a fair coin 50 times. Sometimes you'll get 23 heads. Sometimes 27. Sometimes 30. Same coin, just different runs. The variation isn't because the coin changed. It's because 50 flips is a small sample, and small samples bounce around a lot. AI visibility measurement at 50 prompts works the same way, which is why a lift that fits inside the natural fluctuation band is not a lift at all, just noise wearing a case study.
We have been making this argument since we started IQRush, and I'll be honest about how it went early on: it cost us. Walking into a demo and saying "here is the range, and here is what we cannot tell you yet" is a harder sell than a clean hero number. There were quarters when I wondered if rigor was just a slower way to lose. What changed yesterday is that the category's own trade body put the vocabulary in writing. Decision-grade is now a claim a vendor has to defend, not a phrase on our homepage.
More than 20 companies sell AI visibility measurement today. The IAB wrote this guidance because their numbers do not reconcile.
To be precise about what we can and cannot claim: the IAB developed these definitions with a working group that included measurement providers, brand data-science teams, and agencies. We contributed to that work. We did not write the framework, and the IAB endorses nobody. The test now is the same for every vendor on the market, us included: put your methodology where the guidance is.
Marketer-Side Impact
The industry wastes roughly $2 billion a year acting on AI visibility numbers that don't hold up. The unit of waste is familiar: a $50K content reallocation briefed against a shift that was random fluctuation, an agency team rebuilding a report because two tools disagree, a board meeting where the number on slide 4 can't survive one follow-up question.
The IAB guidance doesn't cost you anything to use. It's a free filter. Run your current vendor's reporting through the decision-grade definition and the 50-query floor, and you'll know within one meeting whether the number you defended last quarter was measurement or theater.
Three questions to take into your next vendor conversation
Under the IAB's definitions, is the data on my dashboard decision-grade, directional, or exploratory, and will you state that in writing?
Which of the 4 P's do you actually measure, and for each number, how many prompts and repeat runs sit underneath it?
When your number and another tool's number disagree, what is your documented method for deciding which one is wrong?
None of these questions require a statistics background. They require a vendor willing to be audited against a public document. A vendor who squirms at question one has answered it.
What happens next
The IAB has said the work continues beyond organic visibility, with paid placement and attribution frameworks on the horizon. Nearer in, the IAB is hosting an AI Insiders panel on August 19 to walk through the framework; registration is open. If you're evaluating vendors this quarter, an hour there is worth more than a stack of pitch decks.
My falsifiable bet: by the March 2027 planning cycle, at least one enterprise RFP you receive or issue will require vendors to classify their data as decision-grade or directional under this framework, by name. When that happens, the vendors who spent 2026 shipping hero numbers will be rewriting their decks. We'll be sending ours as-is.
Frequently asked questions
Is "Measuring Visibility in the AI Era" a formal IAB standard?
No. It's guidance. The IAB's position is that formal standards need market stability that AI search doesn't have yet, so the document establishes shared vocabulary, a metrics hierarchy, and disclosure expectations that can evolve with the market. It's still the closest thing to a yardstick this category has ever had.
What are the 4 P's of AI Visibility?
Presence (does your brand appear in AI answers), Prominence (where and how prominently it appears), Portrayal (how accurately and favorably you're represented, including hallucinations vs factual errors), and Persuasion (whether appearances drive action, such as post-citation clickthrough). Together they form the IAB's common metrics hierarchy for AI visibility.
What's the difference between decision-grade and directional data?
Decision-grade data has been rigorously tested and is fit to support budget and strategy decisions. Directional data is suitable for spotting trends but not for acting. The guidance also defines an exploratory tier below directional: fewer than 50 queries. If a vendor can't tell you which tier their reporting sits in, treat it as exploratory.
Does the IAB recommend specific AI visibility vendors?
No. The IAB is vendor-neutral and recommends no tools. IQRush sits on the IAB AI Visibility working team that developed the framework, which is participation in standards work, not an endorsement. Any vendor implying the IAB endorses their product is telling you something useful about how they handle claims.
What should I do with this guidance if I already have an AI visibility vendor?
Ask them to classify their reporting as decision-grade, directional, or exploratory under the IAB definitions, in writing, and to disclose prompt counts and repeat runs for every headline number. If the answer is a deflection, run your next big decision on a second source before you move budget.
AI search visibility you can defend
Whether you're building, buying, or briefing on AI search, get decision-grade data that holds.
© 2026 IQRush. All Rights Reserved.
© 2026 IQRush. All Rights Reserved.
Site by ONBOX
AI search visibility you can defend
Whether you're building, buying, or briefing on AI search, get decision-grade data that holds.
Resources
AI search visibility you can defend
Whether you're building, buying, or briefing on AI search, get decision-grade data that holds.
© 2026 IQRush. All Rights Reserved.