Zeotap launches Open CDP: the first fully composable CDP that runs on open-source compute and the customer's own lakehouse — sovereign by design

European enterprises can now run their full customer data platform on open Apache Spark and open Iceberg tables in their own jurisdiction, with no proprietary warehouse underneath

Berlin, Sept. 15, 2026 (GLOBE NEWSWIRE) -- Zeotap today announced Open CDP, a new way to deploy its composable Customer Data Platform (CDP) that runs entirely on open-source compute over the customer's own lakehouse. The offering activates customer data using open-source Apache Spark and open Apache Iceberg tables on the customer's own object storage, governed through the customer's existing catalog with no proprietary warehouse, no copy of the data into a vendor's cloud, and no per-credit compute tax. The tangible benefits for enterprises include greater control, data sovereignty and residency compliance, and more predictable total cost of ownership (TCO).

 

Zeotap’s InfraFlex platform was already designed to offer enterprises full flexibility of deployment model across packaged, composable, and hybrid systems on warehouses such as Google BigQuery, Snowflake, Databricks, Amazon Redshift, and others. The Open CDP capability takes it even further by addressing the remaining gap that warehouse-native composable CDPs have long left open for regulated European enterprises. 

 

Composable CDPs keep customer data in the customer's warehouse and activate it in place but they still rent a full cloud warehouse to do the work. That ties the architecture to a single vendor's cloud, prices continuous audience computation as metered consumption, and cannot satisfy hard data-residency or sovereignty requirements. Open CDP removes these constraints.

Open at every layer

Open Compute is built so the customer chooses each layer of the stack rather than inheriting a bundled dependency:

 

• Engine — open Apache Spark. The proven open-source engine for large-scale identity resolution, audience computation, and activation, run on bare-metal compute. No proprietary runtime to lock into.
• Storage — the customer's lakehouse. Profiles, identity graph, events, computed traits, and segment memberships live as open Apache Iceberg tables on compatible object storage the customer owns, readable by any engine.
• Catalog — bring your own. The platform federates with the customer's existing catalog rather than imposing a new one, with support at launch for Databricks Unity Catalog, Google Cloud, Snowflake, and open-source Unity Catalog. Governance stays where the customer already manages it.
• Deployment — operated by Zeotap or by a partner. Customers can have Zeotap operate the compute on their behalf for a fully managed experience, or run it on their own or a partner's infrastructure for maximum control.

 

On top of this open substrate sits the CDP itself: identity resolution, audience and segment building, computed traits, journey orchestration, reverse-ETL activation, and consent and PII governance built for regulated B2C.

Sovereignty as an outcome of the architecture, not an add-on

Because the data sits in the customer's own storage, the compute runs on bare metal in a known location, and the catalog stays with the customer, nothing has to traverse a foreign vendor's cloud to do its job. For organisations facing DORA, national data-residency mandates, and CLOUD Act exposure, this is the difference between a control that can be evidenced to a regulator and one that cannot.

 

The sharpest fit is the regulated, cost-sensitive, high-volume B2C enterprise —telecommunications, retail and fashion, and financial services— where high event volumes, hard residency requirements, and continuous audience computation all apply at once.

 

"Composable was the right instinct, but the industry stopped one layer too high. It kept renting a proprietary warehouse to run the work. Open CDP takes composability down to the compute itself. Customers keep their data, their storage, and their catalog; we bring the open engine and the CDP intelligence on top. Openness becomes the lock-in story, told in reverse — because every layer is replaceable, customers can leave whenever they want, which is precisely why they stay," said Projjol Banerjea, founder and Chief Product Officer at Zeotap.

Availability

The Zeotap Open CDP is available to qualified enterprise customers in Europe and elsewhere in the world. Zeotap works with each customer to model the deployment against their existing lakehouse, catalog, and residency requirements. For more information, visit https://zeotap.com/

About Zeotap

Zeotap is the AI-powered Customer Data Platform that helps enterprises unlock the true value of their customer data - faster, smarter, and with enterprise-grade precision. With plug-and-play connectors and a flexible, modular design, Zeotap integrates effortlessly into existing ecosystems, delivering 3x faster time-to-value than traditional CDPs. Trusted by leading global brands, Zeotap combines the industry’s most advanced ID Extension and identity resolution with AI-powered orchestration - enabling unmatched match rates, smarter segmentation, and real-time activation. Designed for business teams and loved by IT, Zeotap makes it easy to unify, activate, and scale customer data - while staying fully compliant with global privacy regulations. Smarter data. Stronger outcomes. Built for business.

Press Inquiries

Ketty Gobbo
Head of Marketing
ketty.gobbo [at] zeotap.com
https://zeotap.com/