ETL: Split Transformation

Use the Split transformation to divide a single data stream into multiple output streams based on conditional rules in your ETL pipeline.

The Split transformation divides a single data stream into multiple output streams based on conditions you define. Each output can route to a different destination or transformation path.

Overview

• Evaluates each row against one or more conditions
• Routes rows to the matching output branch
• Rows that match no condition go to a default output
• A single row can only go to one output branch (first match wins)

Configuration

1. Add the Split transformation to your dataflow.
2. Define conditions for each output branch using expressions.
3. Connect each output to downstream components.
4. Rows not matching any condition route to the default output.

Use Cases

• Route error records to a separate destination for review
• Split data by region, category, or status for parallel processing
• Separate new records from updates based on a flag field