Connected Sources

In your data warehouse

1. Create a role and grant permissions to query and create tables in a schema.
2. Set up your warehouse and give access to that role.
3. Create a user for that role.
4. Depending on your configuration, you may need to allow Braze IPs in your Snowflake network policy.

In the Braze dashboard

5. Create a new connected source in the Braze dashboard.
6. Configure the sync details for the connected source.
7. Retrieve the public key provided in the Braze dashboard.

In your data warehouse

8. Append the public key from the Braze dashboard to the Snowflake user for authentication. When you’re finished, you can use the connected source to create one or more CDI Segments.

Step 2.1: Create a role and grant permissions

Create a role for your connected source to use. This role will be used to generate the list of tables available in your CDI segments, and to query source tables to create new segments. After the connected source is created, Braze will discover the names and description of all tables available to the user in the source schema.

You may choose to grant access to all tables in a schema, or grant privileges only to specific tables. Whichever tables the Braze role has access to will be available to query in the CDI segment.

The create table permission is required so Braze can create a table with your CDI Segment query results before updating the segment in Braze. Braze will create a temporary table per segment, and the table will only persist while Braze is updating the segment.

1
2
3
4
5
6
7
8
9
10
11
12
13
CREATE ROLE BRAZE_INGESTION_ROLE;

GRANT USAGE ON DATABASE BRAZE_CLOUD_PRODUCTION TO ROLE BRAZE_INGESTION_ROLE;
GRANT USAGE ON SCHEMA BRAZE_CLOUD_PRODUCTION.INGESTION TO ROLE BRAZE_INGESTION_ROLE;
GRANT CREATE TABLE ON SCHEMA BRAZE_CLOUD_PRODUCTION.INGESTION TO ROLE BRAZE_INGESTION_ROLE;

-- grant access to all current and future tables or views in the schema
GRANT SELECT ON ALL TABLES IN SCHEMA BRAZE_CLOUD_PRODUCTION.INGESTION TO ROLE BRAZE_INGESTION_ROLE;
GRANT SELECT ON FUTURE TABLES IN SCHEMA BRAZE_CLOUD_PRODUCTION.INGESTION TO ROLE BRAZE_INGESTION_ROLE;

-- grant access to specific tables or views in the schema
GRANT SELECT ON TABLE BRAZE_CLOUD_PRODUCTION.INGESTION.USERS_ATTRIBUTES_SYNC TO ROLE BRAZE_INGESTION_ROLE;

Step 2.2: Set up the warehouse and give access to Braze role

1
2
3
CREATE WAREHOUSE BRAZE_INGESTION_WAREHOUSE;

GRANT USAGE ON WAREHOUSE BRAZE_INGESTION_WAREHOUSE TO ROLE BRAZE_INGESTION_ROLE;

Step 2.3: Set up the user

1
2
3
CREATE USER BRAZE_INGESTION_USER;

GRANT ROLE BRAZE_INGESTION_ROLE TO USER BRAZE_INGESTION_USER;

You will share connection information with Braze and receive a public key to append to the user in a later step.

Step 2.4: Allow Braze IPs in your Snowflake network policy (optional)

Depending on the configuration of your Snowflake account, you may need to allow the following IP addresses in your Snowflake network policy. For more information on doing this, view the relevant Snowflake documentation on modifying a network policy.

• united states (us)
• european union (eu)

Step 3.1: Add Snowflake connection information and source table

Create a connected source in the Braze dashboard. Go to Data Settings > Cloud Data Ingestion > Connected Sources, and then select Create new data sync > Snowflake Import.

Input the information for your Snowflake data warehouse and source schema, then proceed to the next step.

Step 3.2: Configure sync details

Choose a name for the connected source. This name will be used in the list of available sources when you create a new CDI segment.

Configure a maximum runtime for this source. Braze will automatically abort any queries that exceed the maximum runtime when it’s creating or refreshing a segment. The maximum runtime allowed is 60 minutes; a lower runtime will reduce costs incurred on your Snowflake account.

Step 3.3: Note the public key

In the Test connection step, take note of the RSA public key. You’ll need it to complete the integration in Snowflake.

Add the public key you noted during the last step to your user in Snowflake. This will allow Braze to connect to Snowflake. For details on how to do this, see the Snowflake documentation.

If you want to rotate the keys at any point, you can create a new public key by going to Data Access Management in Cloud Data Ingestion and selecting Generate New Key for the respective account.

1
ALTER USER BRAZE_INGESTION_USER SET rsa_public_key='{INSERT_YOUR_KEY}';

After you add the key to the user in Snowflake, select Test Connection in Braze, and then select Done. Your connected source is now created and ready to use in CDI segments.

You may set up multiple integrations with Braze, but each integration should be configured to connect a different schema. When creating additional connections, you may reuse existing credentials if connecting to the same Snowflake account.

If you reuse the same user and role across integrations, you won’t need to add the public key again.