# Share of Model Is a Dashboard, Not Yet a Media Currency *Attribution is breaking because its billing event, the click, is disappearing. Share of Model is the heir apparent. But a metric nobody transacts on is a dashboard, not a currency.* Published: 2026-09-03 | Read time: 5 min read | Author: Smalk AI Research | Source: https://www.smalk.ai/blog/share-of-model-media-currency-ai-visibility --- Attribution did not break because marketers measured badly. It broke because its billing event, the click, is disappearing, and the metric being groomed to replace it, Share of Model, has no billing event at all. - More than half of Google searches now end without a click, and AI answers shape decisions before any site visit registers (The Drum, April 2026). - Marketers are adopting Share of Model, the frequency with which AI engines mention or cite a brand, as the successor KPI to click attribution. - Media history is consistent: a metric becomes a currency only when both sides of a market transact on it, the way impressions did with ad servers and GRPs did with the upfronts. - Until a transaction layer prices brand presence inside AI answers and pays the publishers behind them, Share of Model remains a scoreboard, and scoreboards do not fund content. ## What Share of Model actually measures Share of Model is the percentage of AI-generated answers, across a defined prompt set, that mention or cite a brand, tracked on engines such as ChatGPT, Perplexity, Gemini and Google AI Mode. It is the answer-economy analogue of share of voice. As of mid-2026 there is no certified standard for computing it; frameworks published this year converge on frequency across fixed prompt sets, but vendors still define and calculate it differently (Ellevate and Storylake, July 2026). ## The click had a currency. The citation has a scoreboard. The open-web economy ran on metrics that doubled as invoices. The impression was counted by an ad server both sides accepted, the click carried a CPC, the GRP was priced at the television upfronts. Measurement and money settled on the same unit. The agentic web has inverted that order: influence now happens inside AI answers, the industry can increasingly count it, but no unit of that counting triggers a payment to anyone, least of all to the publishers whose content grounds the answer. ## What The Drum's attribution coverage says, and what we verified The Drum's 2026 industry-insight coverage argues AI discovery has outrun attribution: rankings look healthy while traffic falls, more than half of Google searches end without a click, and AI Overviews now appear on roughly one in five queries (The Drum, April 2026). Semrush research reported by The Drum in June 2026 adds that 43% of consumers now use AI tools for discovery, while 18% of marketers report more leads arriving without clear attribution. One verification note: zero-click estimates range from just over 50% to roughly 60% depending on methodology and date; we use the conservative end. At IAB Connected Commerce, the IAB's Collin Colburn went further, arguing the industry should retire deterministic attribution in favor of measuring commercial influence on both humans and agents (The Drum, April 2026). ## Why measure now, monetize later worked for SEO and will not work here ### The strongest version of the measure-first case SEO built a multi-billion dollar industry on a metric nobody transacted on directly: the ranking. Optimists argue Share of Model will follow the same path, with measurement maturity preceding monetization, and they can point to how quickly 2026 visibility frameworks converged on fixed prompt sets and multi-engine tracking. ### Rankings had a billing event downstream. Citations do not. The analogy fails on one structural fact. A ranking was never the currency; it was a proxy for clicks, and every click carried a CPC or fed an ad impression that paid someone. A citation today terminates inside the answer. Worse, the metric is unstable without a market to anchor it: Semrush tracking recorded ChatGPT citing Reddit in close to 60% of responses in August 2025 and around 10% six weeks later (reported by Storylake, July 2026). No CFO builds a media budget on a signal that can move sixfold in six weeks with no contract attached. ## What this means for brands and for publishers ### For CMOs, media buyers and agencies: track the metric, budget for the transaction Adopt Share of Model now, with a frozen prompt set and multi-engine tracking, but do not confuse the scoreboard with the game. The actionable line is the emerging transaction layer: paid programs that place brands inside AI answers with measurable mention lift. Fund a test budget in 2026 while that inventory is underpriced, and require citation-level reporting from every partner you brief. ### For publishers: your citations are inventory, sell them as such Publishers already supply the evidence that wins recommendations. eMarketer reported in 2026 that 70% of eyewear brand Zenni's LLM citations trace back to affiliate content, proof that third-party pages, not brand homepages, decide the answer. Counting citations is table stakes. The move is converting citation share into rate-card inventory inside an ad model built for AI answers, rather than waiting for engines to volunteer a revenue share. ## Three currency signals to watch through 2027 Watch three formation signals. First, whether a measurement body or industry coalition publishes a certified Share of Model standard, the Nielsen moment for AI visibility. Second, whether the AI engines' expanding ad programs adopt any citation-linked pricing rather than pure audience CPMs. Third, whether the first publisher rate cards denominated in citations or mentions appear. Currencies form when one standard clears both sides of a market; whoever moves first writes the benchmark. ## Conclusion Hold on to this: attribution is not being rethought, it is being replaced, and its successor metric will only matter once a market transacts on it. Smalk AI is building that transaction layer: a Generative Engine Advertising network that places native ads for AI agents, prices brand presence inside AI answers, and routes revenue to the publishers whose content grounds those answers, giving Share of Model the billing event it lacks. What to watch next: the first citation-denominated rate card, because that is the moment the metric becomes a currency. ## FAQ ### What is Share of Model? Share of Model is the percentage of AI-generated answers, across a fixed set of category prompts, that mention or cite a given brand, measured across engines such as ChatGPT, Perplexity, Gemini and Google AI Mode. It is the answer-economy equivalent of share of voice, and as of 2026 it has no single certified calculation standard. ### Is Share of Model replacing attribution? It is replacing attribution as the visibility KPI, not as the payment mechanism. Attribution worked because the click was both a measurement unit and a billing event. Share of Model measures influence inside AI answers but triggers no payment, so it complements attribution reporting rather than restoring what the zero-click shift removed. ### How is Share of Model different from share of voice? Share of voice measured presence in paid or ranked media, where inventory was priced and both sides transacted on the underlying unit. Share of Model measures presence inside synthesized AI answers, where no unit is priced yet. The intent is the same; the market infrastructure underneath it does not exist yet. ### Why is Share of Model not a media currency yet? A media currency requires a standard both buyer and seller accept and transact on, the way impressions and GRPs were. Share of Model lacks a certified standard, shows high volatility across engines and weeks, and has no payment attached to the unit it counts. It becomes a currency when an ad model prices citations and pays the cited source.