Editing Semantic Models in the Power BI Service (Generally Available)

Semantic model editing in the Power BI service is now generally available, bringing core modeling capabilities to the web.

Create New Semantic Models in the Web

You can now create new import semantic models from over 100 connectors, all within your browser.

Getting started:

1. Navigate to the Create page and select Get data.


This will launch the modern Power Query Get Data experience for you to discover and connect to data including sources like Excel, SharePoint, SQL Server, and more!

2. Choose your data source and configure connection settings.

3. Select the tables you want included in your semantic model.

From here you have the following 3 options:


Option 1: Create a report

This option takes you into the report editing experience in the Power BI service. Here, you can immediately start building a report on top of your newly created semantic model. The report editor provides all the familiar tools for designing visuals, adding filters, and customizing your report layout—right in the browser.


Option 2: Only create a semantic model

Selecting this option creates just the semantic model, without launching the report editor. You’ll land in the model editor in the web, where you can make further edits to your semantic model. This is ideal if you want to refine your semantic model, add relationships, create DAX measures, or configure row-level security before building reports.

Option 3: Transform data

This option allows you to perform rich data transformations using the modern Power Query editor in the web. After completing your transformations, you can choose to either create a report or a semantic model only, depending on your workflow.


Edit Existing Semantic Models in the Web

Not only can you create new import semantic models in the Power BI service, you can also make core modeling changes to your existing semantic models—all without leaving your browser. This includes many capabilities such as:

Manage relationships: between tables in the properties pane, diagram view, or the Manage relationships dialog. This includes support for autodetecting relationships for faster modeling.

DAX authoring: Create and edit DAX measures, calculated columns, calculated tables, and calculation groups with full IntelliSense support.


Edit properties: in the properties pane, including support for the model explorer.

Row-level security: Define and assign RLS roles to secure your data.

Power Query integration: Just like when creating new semantic models, Power Query is supported for editing existing models. You can add new import tables to your model using the modern Get Data dialog

Perform rich transformations using the Transform data dialog including using the highly demanded diagram view for visualizing query dependencies and structure.


Schema and data refresh: Refresh both the schema and data for your import semantic models by selecting Refresh in the ribbon.


Considerations and limitations

There are a few limitations for this release of editing semantic models in the Power BI service, which fall into a handful of categories.

Considerations with the Power Query editor

Keep in mind the following considerations when interacting with the Power Query editor:

  • Using the Power Query editor to Transform data or connect to new data sources is only supported for import storage mode. These capabilities aren't support for Direct Lake or DirectQuery tables.
  • Adding import tables to the model from custom connectors, Azure Database for PostgreSQL, IBM Informix database (Beta), Essbase, Microsoft Exchange, Hadoop File (HDFS), OLE DB, R, and Python aren't supported. Models using these connectors also do not support query editing in the Power Query editor or refreshing via the ribbon button in the web model editing experience.
  • Semantic models with connected goals do not support editing queries in the Power Query editor or refreshing via the ribbon button in the web model editing experience.
  • If you select Cancel or close the Power Query dialog, any changes made to queries will be discarded. In the web, changes made in the Power Query editor must be explicitly saved and applied to the model for them to persist beyond the editor.
  • You can use existing personal cloud connections in the Power Query editor for the semantic model, but you can't create new ones there. When connecting to a data source in the editor, only on-premises or shared cloud connections can be created. To use a personal cloud connection, link it through the semantic model settings page. Configuration and management of these personal cloud connections can be done in the Power BI Manage Connections and Gateways page.
  • When opening the Power Query editor for a model published from Desktop, connections may initially appear unlinked in the Manage Connections dialog. You'll be able to configure these connections by clickign the "+" sign.
  • data gateway is needed to certain data sources. These gateways can be managed from the semantic model settings page. When publishing from Desktop, gateway connections aren't configured by default for sources that require them. You’ll need to manually set them up under Gateway connections in the semantic model settings.
  • Dynamic data sources aren't supported in the Power Query editor.
  • When adding a new import data source using Power Query on the web, the semantic model doesn't automatically inherit the sensitivity label from that data source.
  • When importing data using Power Query in the Power BI service, relationships defined in the underlying data sources aren't automatically imported. These relationships must be manually recreated in the semantic model.

Unsupported semantic models

The following scenarios don't support opening the semantic model in the service:

  • Semantic models that have incremental refresh.
  • Semantic models deployed through deployment pipelines can only be edited on the web in the development workspace. Editing in test and production workspaces isn't supported.
  • Semantic models that haven't yet been upgraded to enhanced metadata format. You can upgrade to enhanced metadata format by opening the corresponding pbix in Desktop and republishing.
  • Semantic models that have automatic aggregations configured.
  • Semantic models that have a live connection.
  • Semantic models migrated from Azure Analysis Services (AAS).
  • Not all semantic models in Pro workspaces are currently supported in UAE North.

To see which limitation is preventing you from opening your semantic model, hover over the Open semantic model button in the semantic model details page. This displays a tooltip indicating which limitation is causing the Open semantic model button to be disabled.

Screenshot of hovering over the open semantic model button.

Limitations

There are still many functional gaps between the model view in Power BI desktop and service. Functionality not yet supported in the service includes:

  • The refresh button within the web editor for semantic models is disabled for Direct Lake, DirectQuery, and composite models as well as models containing customer connectors or cube data sources.
  • Setting a table as a feature table
  • Configuring any feature table properties
  • Changing the storage mode of a table
  • Changing to and from the data category ‘barcode’
  • View as dialog
  • Q&A setup and configuration including editing synonyms
  • Classifying sensitivity of your report
  • When modifying your semantic model within the Service, changing the name of data fields won't automatically update in existing visuals in downstream artifacts that depend on that semantic model.

 

Comments

Popular Posts

Failed to execute the package or element. Build errors were encountered

Microsoft AI Tour Singapore

Temporary enable and disable SSRS subscriptions