CDPs are Becoming the Infrastructure of Decision Engines
We need to move past the broken notion that you must choose between a legacy storage-based CDP and a modern, composable platform that only one vendor understands, says Zack Wenthe, Director of Product Marketing & Customer Data Evangelist at Tealium.
The customer data conversation has always revolved around accumulation. More data. More signals. More storage. The assumption was simple: if organisations could collect enough information about their customers, clarity would eventually follow.
It rarely did.
Despite unprecedented access to customer data, many brands still struggle to answer the questions that matter most: who their customers are, what they need right now, and how to respond in ways that feel relevant rather than reactive. The growing challenge is the ability to create meaning from it.
“Clarity doesn’t come from collecting and storing more data. It comes from turning that data into something more meaningful, both from an insight’s perspective and as a tool to deliver better CXs,” says Zack Wenthe, Director of Product Marketing & Customer Data Evangelist at Tealium.
That shift is changing the role of the Customer Data Platform. Once designed to organise and store information, CDPs are increasingly being asked to do something far more ambitious: Help organisations make better decisions in real time.
Data Without Context: Why Volume is not the Problem
The assumption in many organisations is that better customer understanding comes from more data. In practice, the opposite problem has emerged.
- Most brands are no longer short of data: Customer interactions, behavioural signals, transaction histories, and engagement metrics are being collected at unprecedented scale. The challenge is rarely data availability.
- They are short of usable context: Data often exists in isolation, spread across systems and teams, making it difficult to understand the customer’s intent, needs, or journey in a meaningful way.
- And increasingly, short of timely activation: Insights lose value when they arrive too late. The real challenge is turning customer signals into relevant actions while the opportunity to influence the experience still exists.
“The volume of data was never the problem,” says Wenthe. “Most brands are drowning in data but still starving for context.”
Across enterprises, customer signals are stored in warehouses, duplicated across tools, and visualised in dashboards that explain what happened, often long after the moment has passed. The result is a growing gap between insight and action.
Customer data, in this framing, behaves less like an asset and more like a perishable input. The longer it sits idle, the less useful it becomes for real-time engagement.
As Wenthe puts it, “We need to move past the broken notion that you must choose between a legacy storage-based CDP and a modern, composable platform that only one vendor understands.”
Clarity, therefore, is an activation problem, converting fragmented signals into something meaningful at the point of interaction.
The Structural Problem: When Organisation Design Becomes the Bottleneck
“For as long as technology has been part of the enterprise, the ‘Holy Trinity’ has always been people, process, and technology. Yet too much focus is often given to the technology and not enough to the people and processes,” says Wenthe.
As CDPs evolve, the constraint is shifting away from technology and toward organisational design. While most enterprises frame personalisation and decisioning challenges as tooling issues, Wenthe pointed to a deeper friction point: how teams are structured to operate around data.
“It’s all three — technology, expectations, and the way organisations think about data have all changed,” he said. However, organisational alignment has not kept pace.
- Data teams optimise for infrastructure
- Marketing teams optimise for outcomes
- Product teams optimise for experience
These misaligned incentives create friction even when systems are centralised. Data may be unified in theory, but execution remains fragmented in practice.
“Most organisational charts are a bottleneck,” Wenthe noted. “Even when data is centralised, processes are still scattered across the organisation.”
The implication is clear: CDPs cannot function as decision engines in isolation. They require shared accountability across marketing, data, and product functions. Without that alignment, even the most advanced platform becomes another layer of infrastructure rather than an operating system for decisions.
AI Inside the CDP: Intelligence Amplifier or Illusion of Autonomy?
AI is now being positioned as the layer that transforms CDPs into decision engines. But the reality, according to Wenthe, is more constrained than the narrative suggests. AI is already valuable in several areas:
- Identifying customer patterns at scale
- Predicting intent and behaviour in real time
- Generating personalised creative variations faster than manual workflows
However, its limitations are increasingly visible at the data layer. “Where AI is falling short is actually beyond the models,” says Wenthe. “These failures are data failures. Garbage in, garbage out.”
In other words, AI does not eliminate the need for structured, high-quality customer data. It intensifies it. Without consistent inputs, even the most advanced models struggle to deliver reliable decisions across channels.
This is why the idea of a fully autonomous, agentic CDP remains largely aspirational today. Most organisations lack the data depth, consistency, and contextual richness required for systems to operate without human oversight.
AI, in this sense, is not replacing decision-making inside CDPs. It is accelerating it — but only when the underlying data foundation is strong enough to support it.
From Data Systems to Decision Systems
The evolution of CDPs is not just a technological upgrade. It represents a shift in how organisations define the role of customer data itself.
Storage is no longer the differentiator. Many organisations already rely on cloud warehouses and composable architectures to manage data at scale. What now matters is how quickly that data can be turned into action.
Wenthe argues that flexibility is becoming more important than rigid platform definitions. Different organisations require different combinations of capabilities, and those requirements are changing faster than traditional system roadmaps can adapt.
“There is no one-size-fits-all approach,” he said. “Flexibility within your vendor is more important than a shiny new feature.” As CDPs move toward decision-making roles, they are becoming less about consolidation and more about orchestration, connecting signals, context, and activation in real time.
The organisations that succeed in this next phase will not be the ones that collect the most data. They will be the ones who can convert it into decisions fast enough to match customer expectations.
And in that shift, CDPs are becoming the infrastructure of decisions.
ALSO READ: The Live Moment Effect: How Emotion in Live Sports Boosts Brand Recall
