Joining multiple datasets so they add new rows instead of columns

I’m trying to join multiple datasets into one master dataset. Each dataset is formatted in the same way and comprised of paid ad campaign stats broken out by date for each network. Instead of doing any of the joins available in the UI, I would simply like to add the rows from each dataset to the bottom of the master - Just like copy and pasting new rows to the bottom of an excel doc. Is this possible?

Hey Nick. This is known as a ‘union’ in the database world. You can do this by creating your master dataset using SQL. Your SQL statement should look something like this:

select * from <location>.<dataset_view_1>
UNION ALL
select * from <location>.<dataset_view_2>
UNION ALL
select * from <location>.<dataset_view_3>
;

the location and dataset view names of each of your datasets can be found by going to your dataset and looking at the “warehouse views” section of your dataset overview in the top right corner.

SQL is the best way to do this today but we are working to make unions natively available in Sigma in the future.

This worked perfectly! The screenshot of the Paid Ads CSV Import you attached is exactly what I’m replacing with this. Now I have live data and no more manual imports!!!

1 Like

Hi Tyler,

Is it possible to ‘union’ two data sources in Sigma worksheet world? As I know, there are only ‘join’ ways.

Thanks,
Hugo

Hey Hugo! Currently the best way to do unions is to create a new dataset using the SQL runner. We do plan to introduce native support for unions in the next 3 - 6 months which will allow you to invoke the union straight from the worksheet.

1 Like