Zeotap's composable CDP now runs entirely inside a customer's Snowflake account, eliminating data copies

Zeotap, the European-built enterprise CDP, went live on Snowflake Marketplace on August 5 as a fully composable native app, running identity resolution, audience segmentation, journey orchestration, and AI modeling entirely within a customer's own Snowflake account. No data leaves the Snowflake perimeter at any point, according to GlobeNewswire. That single architectural fact changes the calculus for data and compliance teams that have blocked or delayed CDP deployments for years.

The data-copy problem, finally addressed at the architecture layer

Every conventional CDP shares the same structural flaw: customer data must be extracted from wherever it lives and ingested into the vendor's proprietary cloud environment before the platform can do anything useful. That copy event is where legal, security, and audit friction concentrates. For regulated enterprises, it is often where a deployment stalls.

Zeotap's native app eliminates that step entirely. The platform deploys as containerized services on Snowpark Container Services (SPCS), running directly on the customer's Snowflake instance, as the company described in its launch announcement via GlobeNewswire. Unified customer profiles are materialized as standard Snowflake tables that the customer owns and controls. When an audience is activated, only hashed audience data travels to the downstream destination, never raw personal data.

The practical effect is that the audit perimeter stays intact. The data never migrates to another cloud, never crosses into a vendor's environment, and never leaves the customer's established compliance boundary. For a bank or insurer, that means a CDP deployment can satisfy audit requirements before the first marketing campaign runs, not after.

A full CDP stack, not just a connector

The launch is notable for scope. Zeotap is not offering a lightweight query layer or a simple Snowflake connector. According to GlobeNewswire, the app ships with deterministic and probabilistic identity resolution that generates a persistent unique identifier per person, computed inside the account. Audience building is drag-and-drop with real-time size estimation against live Snowflake data, requiring no SQL from marketing teams.

Journey orchestration supports multi-step, cross-channel sequences triggered in real time by Snowflake events. The AI layer, branded ZeoAI, runs propensity, churn, lifetime value, and lookalike models on Cortex AI, Snowflake's native machine-learning runtime, using the customer's own first-party data and producing no secondary ML export. More than 250 connectors are available at launch, spanning ad platforms, email service providers, CRMs, retail media networks, and customer service tools, as detailed in the GlobeNewswire announcement.

Consent and governance controls are baked into the activation layer, not bolted on afterward. Purpose-based consent is enforced at the moment data is activated, with the platform claiming compliance with GDPR, CCPA, and the EU AI Act. For enterprises operating across jurisdictions, that built-in enforcement is a meaningful reduction in manual governance overhead.

Procurement routed through existing Snowflake spend

Distribution mechanics matter as much as the technology itself for enterprise buyers. Zeotap is available through Snowflake Marketplace's Capacity Drawdown program, which lets eligible customers apply committed Snowflake spend to the purchase. Per the GlobeNewswire release, that means procurement teams can skip a separate vendor onboarding cycle, a new contract negotiation, and an additional budget approval process. For organizations that have already committed significant spend to Snowflake, the marginal friction of adding Zeotap is low.

Time-to-value is quoted at eight weeks from initial Snowflake connection to first operational deployment, according to MarTech Series. That is a compressed timeline relative to enterprise CDP implementations historically, which have routinely stretched to six months or more as data pipelines are built, identity graphs are seeded, and compliance reviews are completed. Running natively inside infrastructure the customer already operates removes several of those setup phases.

Regulated verticals are the immediate opportunity

Zeotap calls out banking, insurance, telecommunications, and the public sector as primary targets, as noted in the GlobeNewswire announcement. These are not accidental choices. They are the verticals where data residency requirements, sectoral regulators, and internal audit functions have most consistently blocked cloud martech adoption. A CDP that runs on infrastructure the enterprise already cleared through its security review is a different conversation than one asking for a new data-sharing agreement.

The composable CDP category has been building momentum across the broader martech market as Snowflake and competing cloud data platforms have expanded their native app ecosystems. Zeotap's launch positions it as one of the more complete CDP offerings in that space, delivering marketer-facing tooling, including the visual audience builder and journey designer, on top of the data-platform architecture rather than alongside it.

For data engineering and marketing operations teams evaluating the platform, the near-term question is whether the 250-plus activation connectors cover their current media and channel mix, and whether the eight-week implementation estimate holds against their specific data model complexity. Both will be visible early in a proof-of-concept running inside an existing Snowflake account.