Power Apps Dataflow - result in wrong column order

I am using a Dataflow in Power Apps to combine a number of imported Excel spreadsheets, using Merge Queries as New to end up with a combined table at the end with columns from each source file, all in the correct order. After publishing and refreshing, instead of my original order, the resulting table has all of the columns arranged in alphabetical order by column name.

Is there any way to ensure that the table I am left with at the end of the merge comes out of the Dataflow looking like the version I've meticulously arranged within the query? I can't see why this is happening and it will create so much more work to manually rearrange the columns after extraction to a spreadsheet.

Any help or advice at all is appreciated, thanks in advance.

Categories:

General Comment

Yes, Power Apps Dataflows tend to reorder columns alphabetically upon loading data into Dataverse or other destinations. Unfortunately, this is an automatic behavior and not explicitly documented in Microsoft's official resources. However, there are a few ways you can work around this issue to retain your desired column order.

Workarounds to Maintain Column Order in Dataflows:

1. Explicitly Reorder Columns at the End of Your Query

After performing the merge, add a "Reorder Columns" step in Power Query before saving the Dataflow:

In Power Query Editor, select Transform → Reorder Columns.
Manually drag and arrange the columns into the desired order.
Click Close & Apply to ensure this step is explicitly recorded.

This should help enforce the correct order before publishing.

2. Use a Custom Column Order in Power Automate (If Exporting)

If you're exporting the data (e.g., to Excel), Power Automate can be used to fetch the records and arrange the columns manually in the desired order before writing them to an output.

3. Recreate the Column Order in Dataverse (If Applicable)

If you are loading data into a Dataverse table, and you control its schema, you can:

Manually create the table in Dataverse with the exact column order you need before importing.
Map the Dataflow fields correctly to the corresponding Dataverse fields.

Dataverse enforces its own schema ordering when data is loaded, so this might help.