Help Needed: Pipeline->Dataflows->Lakehouse->PowerBI

In the pre Fabric days is when I was fairly good with PowerBI and would use the Desktop for all the steps of importing data, transforming and then creating reports.

The client I am working with has Fabric and we want to do it "properly" but find I am getting lost with a few stages.

I have a workspace with the premium feature enabled with the diamond icon

Can someone explain, if this is possible? I may have the steps or technical terms mixed up, but this is my general understanding of what I'm trying to achieve:

Import an on premise SQL into Fabric (Datapipeline?)
create a Lakehouse for this data
Transform and clean the data (Dataflow)
have a custom (or default) semantic model attached
Import the Lakehouse as a data source into PowerBI Desktop so that it inherits the semantic model AND data
Create reports/dashboards in Desktop
Publish: Once reports/dashboards are published they are refreshed based on the Lakehouse (frequency set by the Dataflow?)
Be able to modify the entire workflow as the needs evolve

At the moment this last step (modify the workflow) seems to be the hardest part...

If this is too vague then I can provide some specific examples of the steps in which I feel like I am close to achieving this but am blocked.

Thanks!

4 comments