Lesson
Choose A Data Import Method
Once you identify the key data points necessary for your campaign segmentation and triggers, and understand which systems serve as the sources of truth for those data points, you can choose an import method by following the methodology outlined below.
You can collect data about your users from a variety of places, including:
- Directly from websites and apps in real-time as users interact with your product
- Backend databases
- Cloud data warehouses
- Technology partners (Braze Alloys)
- And more
The location of specific data will determine how you import it into Braze. There are many ways to import data into Braze, including:
- Braze SDK
- Braze REST API
- Cloud Data Ingestion (CDI)
- Customer Data Platforms
- Braze Alloy Partner Connectors
Learn more about data import methods in the Data Collection course on Braze Learning.
Historical Data
During onboarding, you can perform a one-time historical backfill of essential data—such as past purchases, favorite brands, preferences, and other key user traits—using CSV, CDI, or API import. This process ensures that this data is readily available in user profiles and can be effectively utilized in your campaign strategies.
Ongoing Data
Depending on its intended use, ongoing data can be imported into Braze through various methods, including:
The Braze SDK
The Braze SDK integrates with your website front-end and mobile apps to collect real-time events and attribute updates from users as they interact with your product. This integration enables immediate actions, such as triggering an order confirmation email when a purchase event is sent to Braze.
REST API
The REST API compliments the SDK by allowing you to send data in real time, or in batches periodically from other sources such as backend databases and data warehouses. This data can be used for audience segmentation or to directly trigger many different types of campaigns. You can also use the REST API to import catalog-access data to personalize your messages with metadata about offerings such as products, discounts, promotions, and more.
You can periodically load non-customer product data into Braze Catalogs or fetch it in real time from the source using Connected Content. Learn more about Catalogs and Connected Content in the Creating Dynamic Messages with Catalogs and Connected Content Fundamentals courses on Braze Learning.
CSV
In addition to API endpoints, you can import catalog data from CSV files to enhance your messages, just as you would access custom attributes or custom event properties using Liquid.
Learn more about Catalogs in the Braze documentation.
Cloud Data Ingestion (CDI)
Cloud Data Ingestion enables you to establish a direct connection between your data storage solution—such as Snowflake, Amazon Redshift, Google BigQuery, Databricks, or Amazon S3—and Braze. This integration allows you to sync relevant user or catalog data and delete users as needed. While CDI does not sync data in real time, it can periodically update at a frequency of your choosing.
Self-Reflection Questions
Each import method has its own unique considerations. When determining which methods are best suited for your brand, it can be helpful to consider the following questions:
For example, a weekly newsletter may require certain data to be updated at a specific time each week before sending.
If so, the SDK would be ideal. Otherwise, periodic updates using the REST API or CDI may suffice.