Source File Columns I Don't Want In My Data Model

[Anonymous]

Dear Community, 

I am using source file as the below one:

However, my organization is making changes constantly, and starting today i have 2 more columns, which i don't want to see at all. As you can imagine my whole data model suffers because of this change. Can someone help me keep my original structure of the report? Below are the new ones:

Thank you!

Atanas

[nvprasad]

Hi

The other workaround within PowerBI is, In PowerQuery Editor do the following steps.

1. Do the unpivot transformation to get all the groups into one column

2. Split using delimiter then regroup using first & last value as per your exiting group logic

3. Pivot it back.

Thanks 

Venkata Nalla

[ImkeF]

Hi @Anonymous ,

as the first step before all other transformations:

• Check the column you want to keep
• Righclick the mouse
• Click: Remove other columns

This will keep your column names clean and managable.

[Anonymous]

@ImkeF 

thank you, but how do i put that in context? I already applied many steps to transform my data (with the old structure). Should i apply that as an additional step? 

thank you,

Atanas

[ImkeF · Accepted solution]

Hi @Anonymous ,

click on the first step of your query with your mouse and do the selection.

This will automatically insert a step for you.

[Anonymous]

@ImkeF 

Dear Imke, 

It worked like a charm! Just one adjustment on my end. It worked pretty well after the "promoted headers" step!

I don't experience the referesh problem anymore. Thank you for this information!!!

@amitchandak 

Thank you for your quick reply and adding Imke to to the discussion!

One last thing. Here's a link to a webinar hosted by Chriss Webb: https://www.youtube.com/watch?v=mGnNn6gnrKk

It focuses on pretty same problems. Hope it will help you as well. 

Stay safe!

Atanas

[nvprasad]

Hi

The other workaround within PowerBI is, In PowerQuery Editor do the following steps.

1. Do the unpivot transformation to get all the groups into one column

2. Split using delimiter then regroup as per your exiting logic

3. Pivot it back.

Thanks 

Venkata Nalla

[amitchandak]

@Anonymous , The best way is to keep the data model stable or add columns at the end. In this way, I do not think you can avoid changes to data model.

@ImkeF , Any suggestions on this

[Anonymous]

@amitchandak 

Thank you Sir!

I do realize that, but no one asked me when they submitted the request to IT...

I would really appreciate any help...

Atanas