marketingdata management
Building for the Agentic Era: How AI and Identity Are Transforming Audience Data Activation
1. The advertising ecosystem is evolving quickly as organizations adapt to new privacy expectations, identity frameworks, and AI-driven technologies. From your vantage point leading revenue at Optable, what are the biggest shifts shaping how brands and publishers approach audience data today?
Three forces of change are converging, creating a real sense of urgency.
The industry’s identity foundation is collapsing. Third-party cookies and shared device IDs aren’t viable anymore. Publishers without a robust first-party identity strategy are already feeling the revenue impact.
Privacy regulation isn’t slowing down and continues to shape how data can be collected, shared, and activated. It’s forcing decisions about consent management and data governance in the infrastructure, not as an afterthought.
And then AI is changing what's possible fast enough that organizations are racing to understand what it means for their workflows before the window for competitive advantage closes.
Brands and publishers are recognizing that these three aren't separate problems. The ones that understood early on that solving identity and privacy is the prerequisite for unlocking AI are in the best position. You can't build intelligent, automated workflows on top of fragmented or non-compliant data. They’re moving away from trying to survive cookie deprecation towards building the foundation that lets them win in an agentic ecosystem.
2. Optable describes its platform as enabling “agentic collaboration.” For readers who may be new to that concept, how does agent-based technology change the way organizations discover, activate, and collaborate around audience data?
Right now, most of the work in audience discovery, planning, and activation is manual. A publisher receives an RFP, a team member spends hours or days querying data, building a proposal, and setting up a deal. Each step requires human intervention.
Agentic collaboration lifts the burden, using AI agents to take over querying data, building audiences, negotiating parameters, and activating campaigns. This shift keeps a human in the loop for quality control rather than execution.
This means publishers can respond to more RFPs, faster, with richer and more tailored audience proposals, and buyers can discover and activate against premium inventory without waiting for a sales rep to call them back.
What makes this work is that the intelligence moves to where the data lives, rather than the data moving to where the intelligence is. That distinction matters enormously for privacy and for data ownership.
3. Many brands and publishers have invested heavily in first-party data, but unlocking its full value can still be challenging. What approaches are you seeing organizations take to turn these data assets into scalable revenue opportunities?
The gap between having first-party data and utilizing it is still wide for a lot of organizations. A recent survey we did with Digiday found that only 4% of publishers have more than half their audience data identifiable via first-party signals. So even the publishers who have invested are still in the early innings.
The organizations that are closing that gap fastest are doing three things:
They're building a proper identity foundation. They’re not just collecting emails. They’re actually resolving those signals across web, mobile, CTV, and audio into a unified identity graph that makes the data actionable at scale.- They're investing in the infrastructure to enrich and activate that data in real time. They’re moving away from manual processes because the programmatic market rewards speed and precision.
- They're putting AI agents to work. They're replacing manual processes by deploying agents to query their first-party assets, build custom audiences, and push them to activation platforms in minutes. They’re ready for the buyer agents that are already searching and evaluating their inventory.
When those pieces come together, they see first-party data working the way it’s supposed to.
4. Collaboration between brands, publishers, and partners has historically required complex integrations and data sharing. How are privacy-enhancing technologies and clean-room environments enabling more secure and efficient data collaboration across the ecosystem?
The old model of manual data sharing is slow, risky, and technically demanding on both sides.
Clean rooms addressed the privacy concern. Advertisers can bring their data into an isolated environment to match against publisher audiences. What comes out is insights and activatable audiences, not raw data.
What's new is that AI and agentic workflows can remove the remaining friction. Data collaboration can feed directly into agentic pipelines that can build audiences, model lookalikes, and automatically push to activation platforms. We call this process agentic collaboration. An advertiser can onboard into an Optable clean room in minutes, and we've had clients go from match to live campaign in hours.
When collaboration is that fast and frictionless, it can become a regular part of how deals get done.
5. Identity continues to be a central component of effective audience activation, particularly as marketers work across multiple channels like web, mobile, and connected TV. How should organizations think about building an identity strategy that supports both reach and accuracy?
The biggest mistake I see is organizations acting like they can resolve identity with a single solution. Different channels operate with different identity signals. Cookies and hashed emails for web, device IDs with consent limitations for mobile. CTV has IP and device IDs but might require server-side integration without the use of tags.
An identity strategy that only works for web leaves revenue on the table. A flexible, multi-channel approach requires a unified identity graph that can span all of those environments from a single platform, with a policy-driven resolution layer that selects the strongest available identifier for each channel and activation path.
Our ID Switchboard manages resolution across UID2, LiveRamp, ID5, Yahoo ConnectID, Epsilon, and others in real time, selecting the best option per environment without requiring publishers to retag or rewrite integrations when they add a new partner or channel.
Identity solutions that optimize purely for match rates often trade precision for scale in ways that hurt performance and confidence. The organizations building durable identity strategies are the ones treating accuracy and consent as non-negotiable. And they’re the ones that will expand their addressable reach over time.
6. Interoperability across platforms, identity frameworks, and marketplaces has become increasingly important in today’s advertising ecosystem. How does enabling collaboration across multiple partners and environments create new opportunities for marketers and publishers?
Historically, interoperability has been waylaid by too many solutions creating dependency rather than connectivity. Proprietary stacks made it hard to work with other partners or adapt as the ecosystem evolved. But interoperability is the key to a successful agentic marketplace.
Optable is a founding member of the Ad Context Protocol (AdCP), which is about enabling AI agents across the advertising ecosystem to communicate, transact, and optimize across organizational boundaries. No single platform should own the entire workflow; the connections between systems working together is where the value is.
A standard like AdCP facilitates agentic collaboration. It means publishers’ inventory and audience data become discoverable and transactable through AI-powered buyer workflows they couldn't access before. On the buy side, marketers can discover premium inventory and act on valuable first-party data from publishers in their own environments.
Agentic buyers are already looking and deciding who gets premium spend, and a publisher who isn't agent-ready is invisible to those workflows.
7. Looking ahead, how do you see AI-driven collaboration and agent-based workflows shaping the future of audience intelligence, monetization, and partnership models across the marketing ecosystem?
The workflows that define how advertising gets planned, transacted, and optimized today were designed for a world where humans painstakingly managed every step. That world is changing fast.
What I'm most excited about, and what I'm seeing early evidence of with our clients, is that agentic workflows go well beyond speeding up existing processes. Agents operating across publisher first-party data can surface audience insights faster than a human analyst, optimize in-flight against real signals rather than relying on post-campaign reports, and close the loop between planning and outcomes in ways that were impossible before.
Publishers with the right infrastructure can offer demand-side access to their audiences and inventory in ways that didn't exist a year ago. Advertisers can run real-time scenario modeling against publisher first-party data without a week of back-and-forth.
The best-performing relationships we're seeing aren't just vendor-client arrangements. We're working with partners like Scope3, Newton Research, and Chalice AI to build agents together that serve use cases none of us could address alone.
The organizations that will lead through this transition are starting with solid data infrastructure and identity foundations, because AI is only as good as the data it operates on.
