Heading 1
Heading 2
Heading 3
Heading 4
Heading 5
Heading 6
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
Block quote
Ordered list
- Item 1
- Item 2
- Item 3
Unordered list
- Item A
- Item B
- Item C
Bold text
Emphasis
Superscript
Subscript
The decision on the table is where customer data lives and who governs it. Copying everything into a second platform doubles storage, pipelines and the arguments about which number is right. Federating everything can make real-time use cases slow or costly, because every query reaches back into the warehouse. The split you choose is hard to change later, since segments, journeys and AI models all get built on top of it.
Zero copy works in two directions
Zero copy integration covers two flows. Data federation lets Data Cloud query tables in a warehouse such as Snowflake, Databricks, Google BigQuery or Amazon Redshift as if they were local, with no extract, transform and load (ETL) pipeline in between. Data sharing works the other way. The warehouse reads unified profiles and calculated insights from Data Cloud, so analysts work with the same customer view that marketing uses.
Which warehouses support which direction, and with which features, depends on the current Salesforce release and your contract. Check the list for your platform before the design is fixed. Zero copy is one of several ingestion routes in the Data Cloud architecture, next to streaming and batch ingestion, and the choice between them is made per data source.
Federate when the warehouse is already the source of truth
Federation fits when you already run a governed warehouse that finance, analytics and data science trust. Keeping product, transaction and loyalty history there avoids a second copy with its own refresh schedule and its own errors. Data Cloud reads what it needs for segments and insights, and the warehouse team keeps ownership of the data it already manages.
Federation also answers the objection that often blocks a customer data platform: a second copy of everything. Whether Data Cloud belongs in your estate at all is a wider call, and what Salesforce Data Cloud does sets out when it does not. We make this choice with your data team and your marketing team at the same table, because each tends to design for its own queries.
Copy what needs speed or matching
Some data still belongs inside Data Cloud. Real-time behaviour, such as web events that should trigger a journey within minutes, needs to arrive as a stream, because a query back to the warehouse adds delay at the moment it matters. Data that feeds identity resolution often works better ingested as well, depending on volume and refresh needs, since matching rules run across all profiles at once.
The light case needs little. A weekly campaign on loyalty tiers can query the warehouse directly and nobody notices the difference. The heavy case, a basket-abandonment journey that has to fire within the hour, needs its events streamed into Data Cloud. Most organisations end up with both, and that is fine.
