Recommended Deployment Process for Incremental Refresh Changes

Hi manoj_0911,

For an existing semantic model that already has Incremental Refresh enabled, the usual deployment approach is as follows - 

1. New column added to the source view - 

• Update the model in Power BI Desktop.
• Republish the semantic model.
• Run a refresh in the Power BI Service.

2. Incremental Refresh policy changes (archive period, refresh period, detect data changes, etc.)

• Modify the policy in Desktop.
• Republish the model.
• Trigger a Service refresh so Power BI can re-evaluate and manage partitions based on the updated policy.

3. Changing the partition/date column

• Update the RangeStart/RangeEnd filtering and Incremental Refresh configuration.
• Republish the model.
• Perform a Service refresh to rebuild partitions using the new partitioning logic.

Incremental Refresh partitions are created and maintained by the Power BI Service during refresh operations. 
If you're using Premium/Fabric capacity with XMLA read-write enabled, there are additional deployment options - 

• Using tools such as SSMS or Tabular Editor, you can perform metadata-only deployments, manage partitions directly, and execute targeted partition refreshes rather than republishing the entire model.
• This is often preferred for large models where preserving existing partitions is important. 

For most deployments the recommended sequence is Modify → Publish → Service Refresh.

XMLA/SSMS-based partition management is only needed for advanced scenarios requiring fine-grained control over partitions and refresh operations.

📌Microsoft reference - 

1. Configure incremental refresh for Power BI semantic models - Power BI | Microsoft Learn
2. Configure incremental refresh and real-time data for Power BI semantic models - Power BI | Microsoft Learn
3. Advanced Incremental Refresh and Real-Time Data With the XMLA Endpoint in Power BI - Power BI | Microsoft Learn

💡 Helpful? Give a Kudos 👍 — keep the community growing
✅ Solved your issue? Mark as Solution ✔️ — help others find it faster

Best regards,
Rupasree Achari | BI & Fabric Analytics Engineer