Turning AI insights into media activation through open data partnerships

AI is only as useful as the data it can interpret and the systems it can activate. For advertisers, that means AI-powered media starts with connected data that can move across planning, activation, optimization, and measurement workflows. 

This is precisely why open partnerships matter. When marketers can connect paid media signals, first-party data, CRM data, identity, and partner insights across interoperable systems, AI becomes far more effective. As David Wells, industry principal at Snowflake explains, connected data gives AI the context needed to build better audience segments, improve campaign structures, and support more effective media decisions.

In a recent conversation, Wells shared his perspective on how open data partnerships can help advertisers make AI more useful across media workflows.

AI starts with data architecture 

Before AI can improve media performance, marketers need a clear view of where their data lives and whether it can connect to the systems that drive campaign decisions. 

That includes data such as: 

• First-party customer data 

• CRM and loyalty signals 

• Paid media exposure data 

• Owned-site behavior 

• Identity signals 

• Conversion and outcome data 

When those signals live in disconnected systems, AI has a limited view of the customer journey. It may be able to analyze one slice of performance, but it cannot fully explain how media exposure, site behavior, customer value, and outcomes are connected. 

Data composability helps solve that problem. Advertisers should not have to move every dataset into one closed environment to make AI useful. Open integrations allow marketers to work across data clouds, platforms, and partners while preserving visibility and control over how data is accessed and used. 

For partners, this creates a clear opportunity. When data connects seamlessly across systems, it becomes more valuable for advertisers planning, buying, measuring, and optimizing media. 

Data becomes more valuable when AI can review it 

Event-level campaign data can reveal which channels, geographies, audiences, and segments are driving performance. The challenge is making that depth usable for the teams responsible for media decisions. 

That is where AI can help. 

Raw Event Data Stream, or REDS, gives advertisers access to event-level data from their campaigns. When that data can be reviewed through an AI-enabled environment such as Snowflake, it can become easier to surface patterns that may otherwise stay buried in technical reporting. 

For example, advertisers can start to understand: 

• Which channels are contributing to performance 

• Which designated market areas (DMAs) or markets are driving stronger outcomes 

• Which audience segments are converting 

• Which impressions are associated with downstream actions 

• Which signals should inform future audience creation or optimization 

This is a key OpenTTD principle: Data should be useful beyond a single platform report. By making event-level data available in open, partner-friendly environments, advertisers and partners can use it to support planning, measurement, modeling, and activation. 

The real opportunity is connecting paid, owned, and customer signals 

AI becomes more useful when it can read across the signals that matter. Paid media data is valuable, but it becomes more powerful when connected to owned-site behavior, CRM data, identity, and outcomes. 

This is where the Snowflake and The Trade Desk partnership is especially relevant. 

Snowflake supports a governed environment where data can live, connect, and be analyzed. We provide media signals and activation pathways that can help marketers apply those insights to campaign strategy. 

Together, these capabilities can help advertisers move from fragmented data to a more complete view of performance.

That could mean connecting: 

• Paid media exposure to site visitation 

• Site behavior to CRM records 

• CRM data to customer value 

• Customer value to audience scoring 

• Audience scoring back to media activation 

With the right data foundation, marketers can better understand which interactions are creating value and which audiences should receive more investment. 

This fundamentally changes how teams think about media. Instead of treating paid media, owned media, and customer data as separate areas, marketers can now see how advertising connects to business results. 

AI should close the loop between insight and activation 

AI insights are most useful when they can influence what happens next. 

For advertisers, that means moving beyond analysis and into activation. The goal is to use connected signals to build audience segments, inform campaign structure, optimize in-flight, and improve future models. 

Open partnerships help make that possible. 

When advertisers can connect campaign data with customer and outcome signals, they can use those insights to: 

• Create new audience segments 

• Refine existing audiences 

• Identify high-value customer patterns 

This is where composable infrastructure becomes practical. Analysis does not sit in a dashboard waiting for someone to translate it manually. It can inform what gets built, bought, optimized, and measured next. 

For agencies, this can also create a stronger role. AI can reduce manual work, but the value is in helping teams provide better strategic guidance: where to invest, which audiences to prioritize, and how to improve campaign outcomes over time. 

Governance and openness need to work together 

AI needs access to useful data, but advertisers still need privacy, governance, and control. 

That balance matters. Marketers want AI workflows that can connect across systems, but they also need confidence that sensitive data is handled responsibly. They need to control where data lives, how it is accessed, and which partners can use it. 

Open systems work best when they also protect data integrity. 

That means supporting: 

• Data governance 

• Role-based access 

• Privacy-conscious collaboration 

• Interoperability across partners 

• Portability across platforms 

• Clear controls over how data is used 

This is the difference between openness and uncontrolled access. The goal is governed interoperability: the ability to use data across trusted partners without giving up control, portability, or visibility. 

For advertisers, that creates a foundation for AI workflows that can scale beyond one campaign or one use case. For partners, it creates more ways to support marketers without forcing them into closed or disconnected systems. 

As Wells put it: “In ad tech and martech, openness makes the entire system work better.” 

What this means for advertisers and partners 

The next phase of AI-powered media will depend less on isolated tools and more on how well data, partners, and activation systems work together. 

Advertisers need AI that identifies key signals, understands the context, and helps teams act on what they learn. Partners need more open ways to make their data and technology more useful inside the media workflows advertisers already rely on. 

That is the value of open partnerships. When data is connected, governed, and actionable, AI can help move media teams from insight to activation faster. 

To learn more about how The Trade Desk and Snowflake can help advertisers turn connected data into AI-ready media activation, reach out to your partnership manager. 

This information is provided solely for background and is not a representation or guarantee of any future performance. 

David Wells’ perspectives and statistics stem from the discussion “AI & Data Leadership Fireside,” hosted by The Trade Desk at the Possible conference on April 27, 2026.