Skip to content

next-steps.md

Latest commit

 

History

History
66 lines (53 loc) · 2.86 KB

next-steps.md

File metadata and controls

66 lines (53 loc) · 2.86 KB
Raw

Suggested Next Steps

Once you've built and opened your model in Power BI Desktop, here are recommended enhancements:

1. Create Key Measures

The tool auto-generates a record count and URL link measure on your fact table. Build on those with your own business metrics:

Total Revenue = SUM(Orders[Amount])
Average Order Value = DIVIDE([Total Revenue], [Order Count])
Win Rate = DIVIDE(
    COUNTROWS(FILTER(Opportunities, Opportunities[Status] = "Won")),
    COUNTROWS(Opportunities)
)

2. Build Hierarchies

Create drill-down paths in your dimension tables:

  • Date: Year → Quarter → Month → Week → Day
  • Geography: Country → State/Province → City
  • Organization: Business Unit → Team → Owner

3. Apply Any Additional Formatting Desired

4. Hide Technical Columns

By default, this utility follows the best practice of hiding columns that end users don't need:

  • GUID/ID columns (like accountid)
  • They are kept in the model for relationships, but hidden from report view
  • For lookup fields, use sub-column controls (ID/Name/Type/Yomi) to keep only what report authors need
  • If you find that you have additional values that are only needed for formulas, you can hide them as well.

5. Create a Model Documentation Page

Add a report page that shows:

  • Your data model diagram
  • Key measure definitions
  • Data refresh information
  • Known limitations

6. Consider Row-Level Security (for Import or Dual Modes)

If using Import, Dual (All), or Dual (Select) storage modes, implement RLS to control data access:

  1. Go to Modeling → Manage roles
  2. Create roles that filter data based on user context
  3. Assign users to roles in Power BI Service

7. Deploy to Power BI Service

Once you publish your report to Power BI Service:

For DirectQuery (TDS) Models:

⚠️ Critical: Enable Single Sign-On (SSO) in data source credentials:

  • This ensures each user sees only their permitted data
  • Required for Dataverse row-level security to work
  • Enables view filters based on current user context
  • See the Publishing and Deployment section for detailed steps

For Import or Dual (All/Select) Modes:

  • Configure scheduled refresh (up to 8 times per day)
  • Set up failure notifications
  • Consider implementing Power BI RLS if needed (see step 6)

Learn More