Pipelines in Fabric Data Factory: no-code data orchestration

Data Factory in Microsoft Fabric is the data integration solution — equivalent to Azure Data Factory but native to Fabric. It lets you build visual pipelines to move, transform and orchestrate data across sources.

Create your first pipeline

In the Fabric workspace, select New → Data pipeline. The visual canvas presents activities you can drag and connect:

• Copy Data — copy data between sources (SQL, REST, Blob, etc.) and destinations (Lakehouse, Warehouse).
• Notebook — run a Spark notebook as part of the pipeline.
• Stored Procedure — invoke an SP in a Warehouse.
• If Condition / Until / ForEach — flow control logic.

Example: incremental ingestion

Pipeline: Incremental_Ingestion
1. Lookup → SELECT MAX(last_load) FROM watermark_table
2. Copy Data (SQL → Lakehouse)
   Source: SELECT * FROM sales WHERE updated_at > @{activity('Lookup').output.firstRow.last_load}
   Sink: LH_Bronze/Tables/Sales (Append)
3. Stored Procedure → UPDATE watermark_table SET last_load = GETDATE()

Parameterization

Define parameters in the pipeline (gear icon) and use them in expressions with the syntax @pipeline().parameters.ParamName. This lets you reuse the same pipeline for multiple tables.

Triggers

Pipelines can be triggered by:

• Schedule — cron-like, with configurable recurrence.
• Event-based — when a file arrives in OneLake/Blob Storage.
• Manual / API — via REST API for integration with external systems.

Conclusion

Fabric Data Factory pipelines cover the vast majority of ingestion and orchestration cases with no code. For complex transformations, combine pipelines with Spark notebooks — the best of both worlds.