Best Practice: Data has changed column names, how to adapt PowerBI to slightly revised names

[Anonymous]

Hi all, 

A question of how this should be approached.

The data source (Excel export from database) has changed the names many of the columns meaning that PowerBI no longer automatically picks them up - I can go back into PowerBI refresh data, remove the 'data type' step (after promoting headers) and run this again and that will address most of the issues, but this doesn't automatically update some visuals, and also means that any 'groups' I've created will be lost. 

Is this the best way to approach the change in column names (it's a large amount of names sadly), or is there a way to match old names to new ones meaning that they'll automatically apply to existing visuals and groups? 

The data is a single export and everytime is the whole list, so it's not a period update that could be used as an append or similar. 

Thank you 

[mcgeeks]

I have dealt with this more than once.  It's not fun and with very complicated reports it becomes a nightmare quickly. 

I would love it if MS could remedy this. 

Scenario: Say a report that routinely works starts failing to refresh.  The builder opens it in the desktop version to fix it. 

Suggestion:  If there were a tool to re-link to the newly pulled dataset, old column A name = new column A name and that sort of thing that would be lovely.  The old list of column names and the new list of column names side by side to tie together.  Then when you connect them to the new names it changes all the references throughout the build.  Even when grouping and ungrouping fields.  (I'm fighting that now actually on changed column names.)

[v-easonf-msft · Accepted solution]

Hi, @Anonymous 

There is currently no better way. Once the model changes, I'm afraid you'll have to make these complicated replacements.

Best Regards,
Community Support Team _ Eason

[Anonymous]

A shame, but thanks for the confirmation.