Get Help

CAT 2.2

Testing data in Power BI workspaces

CAT has supported testing Power BI data for many versions now. You could test both:

  • data in Power BI Desktop, before you deploy changes

  • and data in Power BI semantic models already deployed to Power BI workspaces.

Previously, setting this up required technical knowledge—you had to provide connection strings and configure providers. What’s new in this version is just how easy it is. To create a new data source in CAT Studio, you only need to provide the name of your workspace and the name of your model:

Form with tenant name, workspace name, model name and others

That’s it! You can now create automated tests for your measures and compare the model’s data (DAX) with Databricks, SQL Server, and other data sources. MFA is supported (you can leave User Name and/or Password empty). Service principals are also supported.

Testing MS Fabric data

CAT has supported MS Fabric for a long time, but setting up connections required technical details like connection strings and endpoints. Connecting to MS Fabric is now just as simple:

Icons of MS Fabric Lakehouse, Semantic model and Warehouse

Just copy the SQL Server endpoint from the MS Fabric portal, provide the name of your Lakehouse or Warehouse, and you’re done setting up connections for tests. Here’s an example for Lakehouse:

MS Fabric Lakehouse connection form

CAT Studio provides helpful guidance on where to find each value you need.

Power BI Desktop

If you run lots of automated tests against locally running Power BI Desktop files, you’ll notice a significant performance boost. The tests now work more effectively with connections, which has a huge impact on test performance. Thanks to Roman Lánský for suggesting this improvement!

Is CAT Studio still free to use?

Yes! CAT Studio is still free to use, and it will remain free until March 30th, 2026. No registration needed, just get CAT Studio from MS Store or from our Releases page.

Small Improvements

This version contains bug fixes and lots of small improvements suggested by users—not only for CAT Studio, but for all CAT tools.

For example, CAT will now happily test a query like this:

SELECT ID, FirstName, ID
FROM   Person

Was it our bug or your bug? Either way, it doesn’t matter for testing—now there’s no problem with duplicate column names. There were also many other small improvements throughout.

Output Details

In the last release, we announced a new experimental feature: the Set is empty expectation can now export all erroneous data into a SQL Server table. In this release, there’s a breaking change—the data is now output to the first defined output that points to SQL Server.


That’s it :-). Happy testing!

CAT team