Data is more abundant than ever. The question is what to do with it.
According to a February 2026 eMarketer report, three-quarters of U.S. buy-side leaders now believe traditional ad measurement approaches are underperforming, a number that suggests the era of dashboards may be over. Holly Yonosko, chief analytics officer at Omnicom Media Group, said the answer is not more data, but smarter constraints on how AI uses the data already available.
“We now have access to more data than we’ve ever had access to before,” Yonosko said. “But the question then is the measurement and practicality of that data.”
This fix, she argued, requires moving far beyond the vanity metrics built into most client dashboards to a framework built around durable, long-term business outcomes.
Guardrail problem
While faster optimization obviously sounds like a good thing, Yonosko warns that speed without direction is its own risk. This is especially true when AI agents are left alone to pursue efficiency metrics.
“When building an AI agent for a client, it’s really important to first connect AI to business outcomes,” she said. “Make sure you define those things up front: revenue, customer retention, customer growth.” Without that anchoring, the system will optimize for whatever it can most easily measure, which is rarely the same as what the client actually needs, she argued.
“You need to make sure you understand what the efficiency metrics are that you need to recognize, not just business outcomes like revenue or frequency or attention curves,” Yonosko said. “That way you have more context, so it doesn’t just get out of control.”
Unified identity layer
Cross-platform measurements have long been the industry’s beluga whale, but fragmented ID signals have made tracking even more difficult. Yonosko pointed to Acxiom’s Real ID (Acxiom is an Omnicom-owned data and technology company) as the connective tissue that enables a more consistent view of consumers across channels.
“We have a unified identifier, Real ID,” she said. “We can connect cross-channel to identify unique individuals and create clean rooms to take measurements within a secure, closed-loop environment, allowing us to provide true optimization metrics and measurements to our clients.” The clean room model has become the standard for privacy-friendly measurement, allowing brands to analyze audience behavior without exposing raw personal data to third parties.
The IAB survey found that a growing number of ad buyers plan to place greater emphasis on cross-platform measurement. Yonosko said, “We’re not really trying to understand how the individual channels work, but what is the outcome of these channels working together for each consumer?”
You are viewing our coverage of Beet Retreat San Juan 2026 presented by Alliant and TransUnion. Check out this page for more videos in this series.
