Edit

Matrix planning visual performance considerations

[This article is prerelease documentation and is subject to change.]

The Matrix planning visual in Business performance planning provides interactive data entry and write-back capabilities directly in Power BI.
Because the visual handles both data rendering and user interactions, performance can vary depending on report design, data model size, and configuration choices.

This article describes recommended practices to improve performance and responsiveness when using the Matrix planning visual.

Optimize data loading using Power BI filters and slicers

To optimized data loading, use Power BI filters and slicers to limit the amount of data shown in the visual. The visual runs inside the Power BI engine, it inherits Power BI’s data-loading constraints. If too many rows are returned, both rendering and write-back operations can slow down or fail. To optimize performance:

  • Use slicers and report/page filters to restrict the visible data to the scenario, period range, entity, or business unit being planned.
  • Avoid loading large unfiltered datasets into the Matrix. DirectQuery imposes strict limits on query size and intermediate results. For more information, see Direct query.
  • Ensure users plan at a manageable level of detail. For example, one scenario and one time horizon at a time.

Manage subtotals and hierarchies

Each subtotal displayed in the visual is calculated on the fly. As the number of subtotals and hierarchical levels increases, the visual needs to perform more computations, which can slow rendering.

To optimize performance:

  • Limit the number of subtotal levels, ideally two or three.
  • Avoid showing totals at every hierarchy level.
  • Disable unused totals in the Format pane.

Note

50 records grouped under one parent loads faster than the same 50 records spread across multiple parent lines.

Conditional formatting and cell locking

Conditional formatting and cell lock rules are applied after the data is loaded but before the rendering completes. These features can increase rendering time, especially when applied to many measures or complex hierarchies.

To improve performance, follow these steps:

  • Apply formatting selectively to key KPIs only.
  • Avoid combining multiple nested conditions or rules in the same column.
  • Use static color schemes where possible.

Edit validation and comments

  • Edit validation rules don’t impact initial rendering because they’re only triggered when users edit data.
  • Comments are retrieved through a separate service call in parallel with the data load, so they have minimal impact on render time.

Both features are optimized for interactive performance.

Authentication and connectivity

Use Microsoft Entra ID (Azure Active Directory) authentication instead of API-based authentication for all Business performance planning visuals.

With Entra ID authentication:

  • Each visual uses the signed-in user’s Power BI credentials.
  • No additional API keys or user mappings are required.
  • Single sign-on improves both security and performance, as visuals no longer need to reauthenticate on every refresh.

This configuration applies to both Power BI Desktop and Power BI Service environments.

General optimization recommendations

To maximize performance of your Matrix planning reports:

  • Keep cube design and data models simple — avoid unnecessary columns or unused relationships.
  • Avoid large multi-select entries that update thousands of rows at once.
  • Limit conditional formatting and cell locks to critical measures only.
  • Minimize use of complex calculated measures inside visuals.
  • Use DirectQuery to Dataverse for transactional scenarios that need near real-time results.
  • Use Import mode for high-volume planning datasets when frequent recalculation isn’t required.
  • Test on production-sized datasets to validate end-user experience before rollout.

Troubleshooting tips

If users experience slow rendering or timeout issues:

Symptom Possible cause Recommended action
Visual loads slowly Missing filter measure Add a filter measure to limit data scope.
Delayed edit response Large data volume or complex validations Reduce edit scope or simplify rules.
Frequent timeouts Excessive subtotals or conditional formatting Disable totals and simplify formatting logic.
UI freezing in edit mode Too many concurrent cell locks or updates Reduce concurrency and batch updates.
  • Last updated on