Forrester has published a new analysis that names Databricks CustomerLake as the decisive test case for agentic marketing — a signal that the analyst firm sees the CDP market at an inflection point.

The piece carries a pointed title: "The Canary In The CDP Mine: Databricks CustomerLake Is The Litmus Test For Agentic Marketing." The framing is deliberate. Forrester is treating one product launch as a diagnostic. The question is not whether CustomerLake works in isolation. The question is whether the data foundations most marketing teams built for the CDP era can support what comes next: autonomous agents making decisions on customer data.

Why call it a canary?

Miners once carried canaries to detect gas. If the bird failed, the mine was unsafe. Forrester applies the same logic to customer data platforms. Databricks built CustomerLake on the vendor's data intelligence foundation, and Forrester treats its fortunes as an early warning for the whole category.

The reasoning runs in one direction. Agentic marketing — programs where AI agents plan, activate and optimize campaigns with limited human intervention — demands more from data infrastructure than campaign managers ever did. Agents need clean, governed, well-modeled customer data at every decision point. If a lakehouse-native construct like CustomerLake cannot deliver that substrate, legacy CDP deployments face a harder road.

What does this change for martech practitioners?

For teams evaluating CDPs or lakehouse approaches, the analysis reframes the buying decision. The criterion shifts from "which tool unifies profiles for campaigns" to "which data foundation lets agents operate safely and effectively." Forrester's litmus-test framing suggests practitioners should watch CustomerLake's adoption, integration depth and real-world agentic use cases as a proxy for where the market is heading.

Three practical implications follow from the analyst's framing:

• CDP selection now doubles as agentic-readiness assessment
• Lakehouse architectures compete directly with packaged CDPs for the customer-data layer
• Data quality and governance move from back-office concern to agent-enabling prerequisite

The wider lens

Forrester's intervention arrives as marketing technology vendors race to bolt agentic capabilities onto existing stacks. The analyst's caution is implicit in the metaphor itself. Not every mine is safe. Not every data foundation will survive the shift to autonomous operations.

Databricks, for its part, positions CustomerLake within a broader data-intelligence platform rather than as a standalone marketing tool. That architectural choice is precisely what Forrester puts under examination. Packaged CDPs promised speed to value. Lakehouse approaches promise durability and scale. Agentic marketing, in Forrester's telling, is the test that reveals which promise matters.

The bottom line

For marketing operations and data leaders, the analysis is a prompt to audit current infrastructure against agentic requirements. The canary has entered the mine. Forrester's verdict implies the results will shape CDP roadmaps, vendor positioning and budget priorities across the category.

Practitioners should read the full Forrester post for its detailed recommendations, and track how CustomerLake deployments perform against agentic workloads in production.