Cloud Data Ingestion sources

Each Cloud Data Ingestion source has its own setup page covering everything you need for that source: the resources to create, how to connect it to Braze, the source table or file format for each data type, and source-specific troubleshooting.

Data warehouse sources

• 

  Snowflake
• 

  Amazon Redshift
• 

  Google BigQuery
• 

  Databricks
• 

  Microsoft Fabric

File storage sources

• 

  Amazon S3
• 

  Google Cloud Storage
• 

  Azure Blob Storage

Choose a source

Cloud Data Ingestion connects to two categories of source, and the category determines how syncs are triggered.

Source category	Sources	How syncs run
Data warehouse	Snowflake, Amazon Redshift, Google BigQuery, Databricks, Microsoft Fabric	On a schedule you set, as often as every 5 minutes or as rarely as once per month, or on demand through the API. Braze queries your source table for rows where UPDATED_AT is later than the last synced value.
File storage	Amazon S3, Google Cloud Storage, Azure Blob Storage	Event-driven. Your provider publishes a notification when a new file lands, and Braze ingests that file. File storage sources don’t use UPDATED_AT.

What each source page covers

Every source page follows the same shape, so you can set up any source without reading across multiple articles:

• The resources to create in your warehouse or cloud account, including roles, users, service accounts, and network access.
• A Set up authentication section that lists the authentication methods the source supports, with setup steps for each method.
• How to create the source and the sync in the Braze dashboard.
• A tab for each data type you can sync from that source, covering user attributes, custom events, purchase events, catalog items, Canvas triggers, and user deletions.
• Source-specific query optimization and troubleshooting.

Authentication methods by source

Each source supports one or more ways for Braze to authenticate. Select a source to go to its Set up authentication section.

Source	Authentication methods
Snowflake	Key pair authentication
Amazon Redshift	Username and password
Google BigQuery	Workload Identity Federation, Service account key
Databricks	OAuth M2M (recommended), Personal access token
Microsoft Fabric	Service principal
Amazon S3	IAM role
Google Cloud Storage	Workload Identity Federation, Service account key
Azure Blob Storage	Service principal

Data types by source

Not every data type is available for every source. Use this table to confirm what a source supports before you start.

Data type	Data warehouse sources	File storage sources
User attributes	Supported	Supported
Custom events	Supported	Supported
Purchase events	Supported	Supported
Catalog items	Supported	Supported
User deletions	Supported	Supported
Canvas triggers	Supported	Supported
Connected sources	Supported	Not supported

Shared reference

Some Cloud Data Ingestion behavior is the same no matter which source you use:

• Table setup explains required columns and how PAYLOAD formatting maps to the /users/track object formats.
• Best practices covers sync scheduling, UPDATED_AT behavior, and how to minimize data point consumption.
• Sync logs and observability covers monitoring sync runs and exporting logs.
• FAQ answers common questions about sync behavior, errors, and security.