The IAB — the Interactive Advertising Bureau, the group that has set advertising measurement standards for three decades — has published Measuring Visibility in the AI Era, the first standardized framework for measuring whether a brand appears in AI-generated answers.
For agencies, this is the more important development of the year. Not because the metrics are surprising, but because they're now citable.
What the framework actually defines
Four measurement dimensions:
Presence — does the brand appear in the answer at all, for a defined set of queries?
Frequency — how consistently, across repeated runs. Answers vary; single checks aren't measurement.
Representation — accuracy and sentiment of what's said about the brand.
Attribution — whether the brand is cited or linked as a source.
Nothing exotic. But until last week, every vendor and agency in this space was reporting a homemade score, and clients had no way to tell rigor from theater. That ambiguity was suppressing budgets.
Why standards unlock spend
Procurement and marketing directors don't buy things they can't defend upward. "Our AI visibility score went from 41 to 58" is unbuyable. "Presence rate across 40 tracked customer queries rose from 22% to 47%, measured to the IAB framework" is a slide a CMO can put in a board deck.
The same thing happened with viewability in display and with Core Web Vitals in SEO. A named standard turns an experiment into a line item.
How to package it
Two products, not one.
1. AI Visibility Baseline — fixed fee, one-time. Build a query set of 30–50 questions a real buyer would ask (category, location, use case, comparison, and "who's best for X" phrasings — not brand-name lookups). Run them across ChatGPT, Perplexity, Google AI Mode, and Copilot. Report the four dimensions plus a competitor set of three named rivals. Deliver a prioritized fix list. Price it like a technical audit: this is discovery, and it's paid.
2. Monthly AI Visibility Tracking — retainer line item. Same query set, same four metrics, monthly. Report movement, log every fix shipped, flag new inaccuracies. This is the annuity.
Margin note: the profit is in query-set design and interpretation, not in running prompts by hand. If a junior is manually pasting 40 questions into four assistants every month, the retainer loses money by quarter two. Automate collection so your team's hours go to the fixes.
Two cautions
Don't sell rankings. Assistants are non-deterministic — the same question yields different answers on different days. Sell presence rate over time and accuracy correction. Set that expectation in the SOW, not in the QBR.
Lead with representation. The fastest, most visible client win is fixing what assistants say wrongly about them — stale hours, discontinued services, wrong service areas, a competitor's price attached to their name. That's usually traceable to a specific outdated source and fixable in a fortnight. It buys you the runway for the slower presence work.
For your own shop
Update your proposal language to the IAB terms this month, while it's still a differentiator. Run a baseline on your own agency site — you'll be asked. And use free baselines as a new-business wedge: nothing shortens a sales cycle like showing a prospect that a competitor is being recommended and they aren't.
SimpleAIO runs the four-dimension measurement across the major assistants and returns client-ready fixes, if you'd rather not build the pipeline yourself. Agency accounts include multi-client dashboards and white-label reports.