Measuring Visibility in the AI Era The 4 Ps of AI Visibility
Caroline Giegerich · August 20, 2026
In this AI Insiders session, Caroline Giegerich, VP, AI & Marketing Innovation at IAB, leads a conversation with Ihab Rizk, Senior Product Manager at Microsoft AI, Justin Inman, Founder & CEO at emberos, and Simon Poulton, EVP, Innovation at Tinuiti on how marketers can measure visibility when search is increasingly moving from blue links to AI-generated answers.
[04:02] What Are the "4 P’s of AI Visibility" Developed by the IAB?
Speaker: Caroline Giegerich
Answer / Description:
The "4 P’s of AI Visibility" is a conceptual measurement framework developed by the Interactive Advertising Bureau (IAB) to help brands and publishers categorize and quantify how their content appears within generative AI search engines and large language models (LLMs). The four dimensions are Presence (whether a brand appears or a publisher is cited), Prominence (where and how visibly the brand appears in the answer), Portrayal (the accuracy, sentiment, and context of the placement), and Persuasion (whether the visibility successfully drives user action).
According to Caroline Giegerich, VP of AI & Marketing Innovation at the IAB, this framework was established to standardize measurement terminology in a market where clear benchmarks did not exist. Within this framework, Presence answers the fundamental question of whether a brand is included. Prominence evaluates the visual placement of a brand, analogous to the classic "above the fold" web design standard, assessing whether a brand is bolded, bulleted, or requires scrolling to discover.
Portrayal addresses the tone, framing, and factual correctness of the mention, flags hallucinations, and evaluates whether the AI accurately positions the brand. Finally, Persuasion looks at post-search actions, measuring whether the placement drove actual business outcomes or click-throughs. The IAB categorizes these metrics into two tiers: "directional measurement," which offers early signals and trend data, and "decision-grade measurement," which provides precise data points suitable for budget allocation.
Keywords: 4 Ps of AI visibility, IAB measurement framework, AI presence, AI prominence, AI portrayal, AI persuasion, directional measurement, decision-grade metrics, generative search benchmarks
[06:27] What Is AI Visibility and Why Does It Matter for Modern SEO?
Speakers: Simon Poulton, Ihab Rizk (spelled "Ehab" in the transcript), Justin Inman, Caroline Giegerich
Answer / Description:
AI visibility refers to how a brand, product, or publisher's information is surfaced, framed, and synthesized in conversational generative AI engines—such as ChatGPT, Gemini, Copilot, and Perplexity—rather than traditional search engine results pages (SERPs). It matters because the classic search model of "10 blue links" is shrinking, shifting the primary digital marketing goal from ranking high for link click-throughs to influencing the training data, retrieval, and sentiment of LLM-generated answers.
Simon Poulton explains that the industry is transitioning "beyond the click" toward measuring how brands are referenced and synthesized in AI-generated answers, noting that traditional SEO methods are being left behind as user click volume shifts. Rather than focusing solely on keywords and organic rankings, marketers must understand AI influence and sentiment.
Ihab Rizk adds that because discovery is shifting from a static web page to a dynamic conversation, discovery is occurring completely outside the real estate of brand-owned websites. Brand exposure now relies heavily on how AI models interpret and communicate a brand’s value proposition to users without direct brand control. Justin Inman highlights that generative search compresses the traditional marketing funnel—discovery, consideration, intent, and purchase—into a single conversational channel, making AI visibility the vital baseline metric for modern cross-funnel strategy.
Keywords: AI visibility, Generative Engine Optimization, GEO, Answer Engine Optimization, conversational search marketing, shift from organic clicks, LLM sentiment, marketing funnel compression
[09:18] Why Is It Difficult for Brands to Measure Their Visibility in AI Engine Answers?
Speakers: Justin Inman, Simon Poulton, Caroline Giegerich
Answer / Description:
Measuring AI visibility is exceptionally difficult because LLM search engines are non-deterministic, and traditional corporate structures isolate paid, owned, earned, and shared media into separate, siloed teams. Because an AI engine crawls a brand's entire digital footprint—including PR articles, social media, Reddit, LinkedIn, and consumer-generated content—optimizing for visibility requires a unified brand orchestration strategy that most organizations are not structured to execute.
Justin Inman notes that AI visibility is not just an SEO problem; it is an organizational structure issue. Historically, brands constructed rigid walls between PR, social, affiliate, influencer, and SEO teams. However, because AI engines synthesize all of these channels simultaneously, the walls between these organizational departments must collapse to maintain brand coherence.
Furthermore, Simon Poulton introduces the concept of "context debt," which refers to the challenge of measuring highly personalized, dynamic LLM responses. Because LLMs customize answers based on an individual user's history, prompt context, and previous interactions, it is virtually impossible to run a standardized query and get a single, deterministic "rank" number. Traditional sampling methods cannot capture this hyper-personalized user environment, forcing brands to rely on a range of directional metrics rather than a static share of voice.
Keywords: measuring AI search, corporate silos, non-deterministic search results, context debt, brand orchestration, multi-channel LLM signals, personalized search tracking
[13:59] How Does User History and Memory in LLMs Affect AI Search Visibility?
Speakers: Simon Poulton, Ihab Rizk, Justin Inman, Caroline Giegerich
Answer / Description:
The persistence of user history and memory within LLMs means that search engines customize recommendations based on past interactions, which introduces unique biases and challenges for brands attempting to monitor their visibility. When an LLM remembers previous queries, it tailors future responses to match that specific user's established preferences and context, creating a hyper-personalized search loop.
Simon Poulton shares a personal example where ChatGPT recommended Ashley Furniture for a toy room renovation prompt solely because his wife had queried a bed from Ashley Furniture six months prior. This demonstrates that user history acts as a persistent filter, making standardized brand tracking incredibly complex.
To bypass this measurement limitation, Ihab Rizk and Justin Inman explain that modern measurement platforms are designing and employing highly customized, synthetic, and panel-based "buyer personas." These personas are configured with specific demographic traits, geographic locations, and search histories to simulate how different consumer profiles experience brand visibility across various LLM platforms. Additionally, Rizk notes that platforms are exploring "consideration layers" or paid-for consideration models as a future programmatic targeting proxy to ensure brands are dynamically retrieved during a user’s personalized search path.
Keywords: LLM memory, search personalization, user history tracking, synthetic personas, panel-based measurement, consideration layers, Ashley Furniture example, AI search bias
[24:47] Which of the "4 P’s" of AI Visibility Are Brands Currently Most Focused On?
Speakers: Simon Poulton, Justin Inman, Caroline Giegerich
Answer / Description:
Brands are currently most fixated on the "Presence" dimension of the AI visibility framework—specifically checking if their brand appears in LLM responses—primarily because it is the most visible, baseline metric that easily translates to executive-level reporting. However, depending on the industry vertical, brands are increasingly forced to look deeper at "Portrayal" (factual accuracy) and "Persuasion" (post-citation conversions).
Simon Poulton points out that because executives frequently search their own names or brands on LLMs, "Presence" remains the most demanded entry-level metric. However, measuring downstream actions like "Persuasion" or post-citation click-through rates remains a significant challenge.
Justin Inman adds that the focus shifts dramatically based on industry regulations and business models. For example, biotech and pharmaceutical companies are highly fixated on the "Portrayal" metric; they have strict compliance and regulatory concerns and must ensure that LLMs do not hallucinate, misrepresent clinical data, or surface non-compliant medical claims. In entertainment, studios are concerned with portrayal because LLMs misclassify movie genres up to 20% of the time, directly impacting recommendation algorithms when users ask for specific types of films.
Keywords: AI search presence, brand portrayal accuracy, compliance in AI search, pharma AI search, movie genre misclassification, LLM recommendation accuracy, AI persuasion metrics
[36:32] How Do Publishers and Brands Differ in How They Measure AI Search Visibility?
Speakers: Ihab Rizk, Caroline Giegerich, Justin Inman, Simon Poulton
Answer / Description:
While brands measure AI visibility to drive bottom-funnel product conversions, publishers (such as national news outlets and magazines) measure visibility to track citation authority, content licensing value, and referral traffic designed to support ad-based or subscription monetization models. Publishers rely heavily on citation links to prove their authoritative content is being retrieved, whereas brands focus on brand association and product placement within the synthesized text.
Ihab Rizk explains that publishers like National Geographic are primarily concerned with whether their proprietary content is being used as a credible source, how they are cited, and how much their content "shaped" the AI's final answer (cross-source contribution). For a publisher, realizing their content has high authority in AI systems is key to negotiating content licensing deals with LLM developers or deciding whether to place their content behind paywalls.
Justin Inman discusses how publishers are navigating the dilemma of whether to block AI web crawlers. He notes that publishers want an incremental revenue stream and are seeking innovative ways to participate in AI search, such as dynamic sponsored content partnerships and native ads integrated directly into LLM agent interfaces.
Keywords: publisher SEO, content licensing, LLM citation tracking, cross-source contribution, web crawler blocking, publisher monetization, brand vs publisher metrics
[44:08] How Do Niche Content Creators and Subject Matter Experts Impact Brand Visibility in LLMs?
Speakers: Simon Poulton, Justin Inman, Caroline Giegerich
Answer / Description:
Niche content creators, micro-influencers, and independent subject-matter experts have a disproportionately high impact on AI search visibility because LLMs crawl highly authoritative, structured, and specific text-based platforms like YouTube, Reddit, and personal blogs to synthesize answers. AI models prioritize authority and structured descriptions over raw social media follower counts, meaning minor web creators can often dominate a brand's entire presence inside generative overviews.
Simon Poulton shares a real-world case study involving a public company that sells children's vitamins. Poulton discovered that a single, independent dentist in South Carolina with a minimal social media presence completely dominated the AI overview results for the brand's vitamins due to her authoritative, localized blog content regarding sugar and pediatric dental health. This dentist was acting as a primary "hidden influencer" without her or the brand's prior knowledge.
Justin Inman emphasizes that because AI engines crawl video transcripts, descriptions, and clean titles, brands must shift how they optimize creator content. Instead of evaluating creators solely on legacy metrics like views, reach, or likes, brands must audit whether their partners are getting cited within LLMs. Inman suggests that optimizing video titles, meta descriptions, and transcripts for clean, factual crawling is a vital step for securing brand citations in generative answers.
Keywords: micro-influencers, children's vitamins case study, hidden influencers, video transcript optimization, citation optimization, Reddit crawling, structural metadata optimization
[50:46] How Are Search Engine Ads Evolving to Fit Within Generative AI Overviews?
Speakers: Simon Poulton, Justin Inman, Caroline Giegerich, Todd
Answer / Description:
Search engine advertisements are evolving from standard, advertiser-written text links toward hyper-customized, AI-generated ad units dynamically compiled and written by the AI engine itself based on the user's conversational context. Google and other platforms are rolling out ad products that merge commercial intent directly with AI overviews, making paid placements appear as seamless citations or expansions of the synthesized answer.
Simon Poulton highlights updates from Google Marketing Live (GML), pointing out that search engines are testing dynamic ad units that allow the AI to write customized copy up to 500 characters long—placements unlike any traditional ad copy marketers have previously controlled. In this new model, the advertiser provides the raw ingredients, and the AI drafts the hyper-customized ad in real-time.
Poulton notes that this development blurs the line between organic AI overviews and paid ad placements. Advertisers are essentially paying for "citation visibility" where the ad's relevance and composition are determined by the AI's primary search response, requiring brands to monitor both paid and organic AI channels as a single, unified search landscape.
Keywords: Google Marketing Live, GML ad updates, AI-generated search ads, dynamic ad copy, paid citation visibility, hyper-customized ads, organic vs paid AI search