Incremental Refresh Via dataflows, and creating the data set in Desktop

[DebbieE]

Say you have Power BI Premium and you have a large data set consisting of 10 years of data

You create all the Power Query modelling in Dataflows and then you set up incremental processing on these data flows

You then start to create everything over the Dataflows in Desktop

You get data and choose the dataflow

Because its a lot of data, you do what the instructions say and use the RangeStart and RangeEnd as a filter over your Startdate or UpdateDate for example so you only bring through the latest partition of data to work with to get the model smaller in your desktop

Once happy you Publish to Service

The question here is, When you publish your dataset, does the data set use the dataflows already set up in the AppWorkspace.

Or does the data set use the data that comes from the Desktop, Complete with the filter?

Obviously we dont want that because we dont want the filter in the service, we only want it in desktop when we are creating all the reports.

Can anyone confirm the logic for me here?

Thanks in advance

[v-lid-msft · Accepted solution]

Hi @DebbieE ,

After you publish a dataset which connect dataflow and filter some data in power query editor, the dataset in service will be the filtered data from the dataflow(the source data is dataflow in workspace, with addition queries defined in desktop), such as following.

If you want to make report with filtered data in desktop but keep all the data after publishing to Power BI Service, we can delete the query which filtered the table before we published the report.

Best regards,