DFS Pro

DFS Pro turns connected data into governed data assets and repeatable data workflows. Use it when operational data needs lifecycle control, stewardship, version history, fusion, review, audit, or BI reporting.

What users do in DFS Pro

Task	UI area	Result
Create governed data assets	Dataset Center	A dataset exists with source type, schema, profile, owner, and lifecycle status.
Validate data assets	Dataset Detail	A stewarded dataset can be used downstream.
Review schema changes	Dataset versions and change impact	Downstream users understand what changed.
Build reusable processing logic	Method Library	A method can be tested, published, versioned, and used in fusion tasks.
Fuse multiple datasets	Data Fusion	Reviewed outputs combine records from multiple sources.
Resolve uncertainty	Review Queue	Conflicts, low-confidence results, source disagreements, and manual flags are reviewed.
Fix rejected rows	Rejected Rows	Upstream corrections can be tracked and reprocessed.
Track evidence	Audit Trail and Metrics	Changes, run outcomes, and operational health are traceable.
Build reports	DFS Pro BI	Reviewed datasets can drive dashboards and scheduled reports.

Lite to Pro workflow

DFS Lite connector -> mapped and synced data -> DFS Pro dataset
  -> method or fusion task -> review queue -> validated output

Move from DFS Lite to DFS Pro when a data feed needs any of the following:

• data steward;
• dataset lifecycle;
• schema versioning;
• profile and preview;
• lineage or change-impact review;
• multi-source fusion;
• review queue;
• BI reporting.

Recommended first path

1. Create a connector in DFS Lite.
2. Map and sync source data.
3. Check data quality.
4. Create a DFS Pro dataset from the connector output or imported data.
5. Preview and profile the dataset.
6. Assign a steward.
7. Validate the dataset.
8. Use it in a fusion task, AI Agent workflow, or BI report.

Read next

Page	Use
Datasets	Create and validate governed datasets.
Fusion Tasks	Combine multiple datasets with reviewable matching logic.
Review Queue	Resolve conflicts, low-confidence outputs, source disagreements, and rejected rows.