[Limited Availability]{class="badge informative"}
Source audiences from Azure storage
Connect Azure Blob Storage or Azure Data Lake Storage (ADLS) Gen2 to Adobe Real-Time CDP Collaboration to source first-party audience data for activation and overlap analysis.
Use this guide to create a reusable Azure data connection and run a one-time import from the configured storage location. Before you begin, confirm that your audience files meet the Audience Sourcing Specification. You will grant Adobe read access to your Azure storage during the setup process.
Choose your Azure source type choose-source-type
Collaboration supports two Azure ingestion options. Use the table below to pick the guide path that matches where your audience files live.
You can only configure one source type per data connection. To source from both Blob and ADLS, create separate data connections.
Prerequisites prerequisites
Before following this guide, complete account onboarding and setup. Then complete the prerequisites in this section before starting the configuration workflow.
Some steps require action by an Azure administrator. If you are not the Azure administrator for your organization, identify the appropriate person before starting.
Azure access and permissions azure-access-and-permissions
Before you configure the connection in Collaboration, you or your Azure administrator must grant Adobe read access to the storage container or ADLS Gen2 filesystem that contains your audience files. After the permission setup is complete, the Collaboration configuration workflow validates access during the Consent step.
Prepare your audience data prepare-audience-data
Your audience files must conform to the Audience Sourcing Specification (v1.2) before sourcing begins.
Key requirements include:
- File format: CSV, using commas as field delimiters and pipes (
|) as separators for multiple values within a single field. - Required fields: Every record must include an
AUDIENCE_IDcolumn and at least one supported match key column. - Supported match keys:
HASHED_EMAIL_SHA_256,HASHED_PHONE_SHA_256,HASHED_IPV4_SHA_256,CRM_ID,LOYALTY_ID,ADFIXUS_ID. - Hashing requirements: All match key values must be trimmed, lowercased, and SHA256-hashed before upload. Collaboration does not hash or normalize data before ingestion.
- Column consistency: All files under your configured path must use identical column structures.
All match keys present in your audience files must also be enabled for your Collaboration account. See Set up match keys for guidance.
Values required before you begin values-required
Have the following values ready before starting the configuration workflow.
customerdatastoredatalake-prodaudience-ingestaudiencessourcing/audiences/path1/sourcing/inbound/00000000-0000-0000-0000-00000000000000000000-0000-0000-0000-000000000000Set up Azure permissions set-up-azure-permissions
Complete the steps in this section to prepare your Azure environment. Adobe requires read access to your storage container before the Collaboration configuration workflow can establish a connection. This work is performed in the Azure portal and may need to be completed by your Azure administrator.
After you complete this section, proceed to Configure your Azure connection.
Obtain Adobe’s Azure service principal identifier obtain-principal-identifier
Before you can complete the role assignment steps below steps, contact your Adobe account team to obtain the Azure service principal identifier for your region (North America, EMEA, or Australia and New Zealand). You will use this identifier to grant Adobe read access to your storage.
Set up Azure Blob Storage permissions set-up-azure-blob-storage-permissions
- In the Azure portal, open the storage account, then go to Containers and select the container that contains your audience files.
- Select Access control (IAM), then select Add role assignment.
- Assign the Storage Blob Data Reader role to Adobe’s principal at the container scope.
- Select Save.
Set up ADLS Gen2 permissions set-up-adls-gen2-permissions
For ADLS Gen2 connections, the Container field in Collaboration corresponds to the ADLS Gen2 filesystem in Azure. Use the filesystem that contains your audience files.
Before assigning permissions, confirm that the storage account has hierarchical namespace enabled and that firewall or private endpoint rules allow Adobe access.
- In the Azure portal, open the storage account that contains your ADLS Gen2 filesystem.
- Open the filesystem that contains your audience files.
- Select Access control (IAM), then select Add role assignment.
- Assign the Storage Blob Data Reader role to Adobe’s principal at the filesystem or directory scope.
- Select Save.
After you complete the permission setup for your source type, proceed to Configure your Azure connection.
Configure your Azure connection configure-your-azure-connection
Use the Collaboration configuration workflow to validate your Azure storage details, confirm Adobe access, review automatically mapped identity fields, and create the data connection.
Add a new data connection add-new-data-connection
Navigate to Setup > My audiences, then select the add icon (
The Add audience workflow appears. Select Add a new data connection, then select Next.
Select your Azure data source select-azure-data-source
Select Azure Blob Storage or Azure Data Lake Storage Gen2, then select Next.
Continue through the remaining steps to validate your Azure connection, confirm Adobe access, review field mappings, and create the data connection.
Enter connection credentials enter-connection-credentials
In the Credentials step, provide the information required to access your Azure storage location.
After you enter the required values, select Connect to Azure.
A confirmation message indicates that the connection was established successfully. Select Next to continue.
Grant Adobe access to your Azure storage grant-adobe-access
In the Consent step, Collaboration validates the Azure permissions that you configured earlier.
Select the launch icon next to Consent URL to open the authorization workflow in Azure. Sign in with an account that has permission to grant consent for the storage location, then complete the Azure authorization prompts that grant Adobe access to the configured storage location. After authorization is complete, return to Collaboration and select Confirm consent to validate Adobe’s access.
When the consent validation succeeds, a Consent granted confirmation message appears. Select Next to continue.
Review field mappings review-field-mappings
In the Field Mapping step, Collaboration automatically maps supported identity fields from your source files.
No manual configuration is required.
Review the displayed mappings and confirm that the source fields match the identity columns in your audience files. Select Next to continue.
Review and complete the connection review-and-complete
In the Review step, verify the storage account, container, source path, tenant ID, and field mappings.
The review page also indicates that the current Azure workflow performs a single sourcing run and does not configure a recurring schedule.
When the configuration is correct, select Complete.
Confirm the connection and monitor sourced audiences confirm-connection-and-monitor-audiences
After you select Complete, Collaboration creates the data connection and navigates you to Setup > My data connections.
Confirm the connection was created confirm-connection-created
The connection card in My data connections confirms that the connection was created successfully. The card displays the source type (Azure Blob Storage or Azure Data Lake Storage Gen2), creation date, match keys, audience count, and current connection status.
View sourced audiences view-sourced-audiences
After the connection is created, Collaboration automatically begins sourcing audiences from the configured Azure location. Navigate to Setup > My audiences to monitor sourcing progress and review sourced audiences.
Sourced audiences appear in the My audiences table. Use the audience status, identity count, source, data connection, and last updated date to confirm that the expected audiences were sourced from your Azure connection.
Known limitations known-limitations
Review the following limitations before you create or manage an Azure data connection.
- Match key constraints: Match keys cannot be removed from an existing connection. To change the active match keys, delete the connection and create a new one.
- One active connection per Azure source type: You can have one active Blob connection and one active ADLS Gen2 connection per account. To change the storage location, delete the existing connection and create a new one.
- Subfolder support: Collaboration ingests only files directly under the configured path. It does not ingest files from nested subfolders.
- Separate source types: Blob and ADLS Gen2 are distinct connections—do not mix configuration between them in a single wizard run.
Troubleshooting troubleshooting
Audiences are not appearing or sourcing is slow audiences-not-appearing
If sourced audiences do not appear after you create the connection, complete the following actions.
- Confirm that the audience files exist directly under the configured path and comply with the Audience Sourcing Specification.
- Check My data connections for errors.
- Contact Adobe support with the connection name, storage account, and container details if issues persist after 24 hours.
Audiences source but show zero or unexpected identities zero-identities
If audiences appear after sourcing but identity counts are zero or lower than expected, complete the following actions.
- Verify that all match key values in your audience files were trimmed, lowercased, and SHA256-hashed before upload. Collaboration does not hash or normalize data on ingestion.
- Confirm that the match keys present in your files are enabled for your Collaboration account. See Set up match keys.
Connection failed after initial success connection-failed
Use these checks when a connection was created successfully but later enters a failed state.
- Verify that the Azure RBAC role assignment for Adobe’s principal was not removed or narrowed.
- Confirm files still exist at the path and match the specification.
Import or format errors format-errors
Use these checks when sourcing fails because of file structure, hashing, or column-format issues.
- Ensure all files keep the same column structure and hashing rules as the initial ingest.
Next steps next-steps
After sourcing completes, audiences are available in My audiences for activation, overlap analysis, and measurement workflows. To activate sourced audiences with collaborators, see Activate audiences.
Other available sourcing methods include Experience Platform, Amazon S3, Google Cloud Storage, Snowflake, and CSV file upload. For other audience sourcing methods, see: