Before you can begin creating composable audiences, you must:
\nTogether, these steps make up your warehouse setup. You’ll need access to your data warehouse environment and the ability to write and troubleshoot SQL queries.
\nThe warehouse setup process should be completed by a technical team member—such as a data engineer, database administrator, or someone with equivalent experience.
\nSetting up a connection and creating data models involves:
\nA connection links mParticle to your cloud data warehouse so you can run composable audience queries directly on your data. It defines how mParticle authenticates with your warehouse and which datasets it can access. Once a connection is configured, you can begin creating data models that expose warehouse fields for segmentation—without copying any data into mParticle.
\nRead the Connections Overview to get started.
\nA data model defines the structure and logic used to query your warehouse for audience building. Each model is based on a SQL query and represents a specific type of data—such as users, events, or contextual attributes like subscription plans. Data models expose selected warehouse fields in the audience builder and can be reused across multiple audiences. You can also link models together using relationships to power more advanced segmentation logic.
\nRead the Data Models Overview to get started.
\nOnce your connection and data models are in place, you’re ready to build your first audience. You’ll define segmentation criteria, preview your audience size, and schedule regular refreshes to keep your audience up to date.
\nGo to Audience Setup for instructions on how to build an audience.
","frontmatter":{"title":"Warehouse Setup Overview","order":0.5,"seoTitle":null,"seoDescription":null},"parent":{"__typename":"File","modifiedTime":"October 12, 2026"}}},"pageContext":{"slug":"/guides/data-warehouse/composable-audiences/overview/"}},"staticQueryHashes":["2950460867"]}