Efficiency question - Power Query vs Desktop

[krixtsup3r]

Hi Guys! 

Just wanna ask what is more efficient in different scenarios:

Scenario 1: 

(1) Doing relationships in Power Query wherein you connect your queries to have one summarized query

(2) Doing the connections after the transformations

Scenario 2:

(1) Adding columns in the power query itself, like start of month, end of week, mathematical calculations

(2) Creating calculated columns within the transformed data

I've been pondering which is better, any suggestions is greatly appreciated. 

[Anonymous]

Hi @krixtsup3r 

Could you tell me if your problem has been solved? If it is, kindly Accept it as the solution. More people will benefit from it. Or you are still confused about it, please provide me with more details about your table and your problem or share me with your pbix file from your Onedrive for Business.

Best Regards,

Rico Zhou

[amitchandak]

@krixtsup3r , Scenario 2 is an efficient way.

Scenario 1 dependents. Like summarization should only be done if it is needed.

[krixtsup3r]

@amitchandak  creating a relationship wthin power query to connect data is not efficient as well?  For example connecting your date reference table to your main table?

[Anonymous · Accepted solution]

Hi @krixtsup3r 

There are different options in different situations. Just like we can't do more transform on the column which was built by Dax in Power Query Editor. As below, it tells you when to choose power query or calculated column.

For more info you may refer to this blog:  power query vs calculated column

Best Regards,

Rico Zhou

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.