Power BI Desktop and Power Query in Excel follow the same principles that Power Query Online inside of Dataflow Gen2 does. We try to leverage lazy evaluation as much as possible, which means that we only compute things when they're absolutely necessary in an effort to prevent overloading your data source and/or your own resources in your desktop environment or the cloud.
This means that if you're working with a table that has billions of records and you perform some transformations that could be deemed expensive by your data source, that we will only request to compute around 200 rows of data for the data preview that we showcase, but the rest will not need to be computed and an exact number of rows that could be yielded by that specific step is not required in an effort to not overload the source.
You can, however, click on the "Transform" tab and click the option to "Count rows" which will give you the value that you're after inside of the Data preview
Recent ideas
Data Pipelines - Run only selected activities
For debugging and testing pipeline activities during development, allow us to select one or multiple activities and run only the selected pipeline activities. For example, I'm working on editing ...frithjof_v15 hours agoCommunity ChampionNew617Views11likes2CommentsSemantic model connection bindings should be in source control (Git)
Semantic model data source connection bindings should be source controlled. A semantic model can contain multiple data source references, each of which can be mapped to a separate Fabric data connec...frithjof_v23 hours agoCommunity ChampionNew20Views1like0CommentsBulk changing column names in Visualizations Pane
We often use raw/api column names or measures with a set nomenclature to be consistent and to keep track of them but we do not want to display these names in the visuals. Currently we have to change ...vishal1401971 day agoFrequent VisitorNew6Views0likes0Comments