Simple Workarounds for Direct Query Limitations

[sotoc]

I am fairly new to Power BI and am requesting a reference or tips on how to workaround the Direct Query limitations, as I've only used Import thus far. The very simple description from MSFT here is not helpful to me as a new user:

https://docs.microsoft.com/en-us/power-bi/desktop-directquery-about

It seems most of the steps I perform in Import mode are exactly those that aren't allowed in Direct Query (duplicate table, unpivot pivot, new columns, and manually creating tables to create relationships for sorting). So I am seeking a source of best practices/ efficient workaround suggestions. Such as, instead of duplicate table/ unpivot use a measure to select columns and union?

Perhaps this is something learned with time and experience with DAX as my forum and internet searches aren't coming up with much.

Thanks in advance, 
Carly

[RMDNA · Accepted solution]

Hi @sotoc,

There really aren't many workarounds - needing to do those kinds of operations is exactly why you would use Import rather than Direct Query. What are the constraints to you using Import?

If changing isn't an option, make your transformations in the datasource so that everything you need is available once you've connected by DQ.

[sotoc]

Thank you @RMDNA.  Unfortunately I am bound to  Direct Query only with a single table per report. They will only pull the rows for me, they won't conduct any transformations in the source data. 

Any idea when the composite model  will be officially launched for the service? We are going to embed our report after I create in the desktop so I can't create the report in the preview version yet.