The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredCompete to become Power BI Data Viz World Champion! First round ends August 18th. Get started.
I'm try to merge 2 queries with sales by customer in differnet years. I tried to turn the merge to "full outer" hoping to capture the customer who didn't have sales in 2020 but did have sales in 2021. However the output of the merge shows (null) for customer 5 and 7. Is there a way to merge the two tables so I have 2 distinct columns for sales by year and all distinct customers? Thanks.
Solved! Go to Solution.
In each of the original tables, unpivot the Sales Column. That will bring Sales 2020 and Sales 2021 into a column. Rename the columns Year.
--
You can then Append the 2 tables.
If you want you can leave the table like that or select the Year column and Pivot it.
In each of the original tables, unpivot the Sales Column. That will bring Sales 2020 and Sales 2021 into a column. Rename the columns Year.
--
You can then Append the 2 tables.
If you want you can leave the table like that or select the Year column and Pivot it.
Thank you