Data Cloud can ingest data in real-time using your existing Ingestion API integrations. To do so for an Ingestion API Data Lake Object (DLO), make sure that the Data Model Object (DMO) the DLO is mapped to, is a member of a real-time data graph.
After ingestion, the data is available in the real-time layer instantly so that it can power real-time identity resolution, real-time calculated insights, real-time segmentation, and real-time actions.
Things to Consider
Only streaming data ingestion is supported: Bulk data ingestion is not supported for real-time data ingestion.
Events need Individual ID to be ingested: Any event that doesn’t have an individual ID won’t get ingested to the real-time data graph. Such events are synced through the standard Data Cloud pipeline, which is not real-time.
Cannot delete data in real-time: Data can only be appended or updated in real-time. Any delete operation is performed using the standard Data Cloud pipeline ingestion process.
During Upsert, old records aren’t deleted in real-time: Any new data that is upserted will be done so in real-time. However, any deletes to be made on the old records are not in real-time and are only performed using the standard Data Cloud pipeline.
Billing is based on SSRT events and API usage type: Your Data Cloud Services card is charged for events ingested in real-time.
Real-time data ingestion is enabled by default for all newly created real-time data graphs: If you don’t want to ingest data in real-time, reach out to your Salesforce administrator.
Example
Imagine you work for an auto manufacturer. Your objective is to create a service where you can identify when a vehicle has broken down, and proactively reach out to the customer who owns the vehicle, and send roadside assistance.
Using Real-Time Ingestion API, you can ingest the vehicle telemetry data in real-time into Data Cloud. With this data, you can identify a break-down in real-time and create a support case in Service Cloud to send assistance.