Forum Discussion
Extracting date from poor source data format
Hi all,
I have a live query to an SQL server returning some finance fields and the source data is as per table below.
- B_Jan, B_Feb, B_Mar are the amounts invoiced to a customer in a given month.
- C_Jan, C_Feb, C_Mar are the costs incurred on a job in a given month.
I have only used the Jan, Feb, Mar field as an example but I have a full year dataset. There is no primary key, but the individual record distinction is based on "JobNo, Year"
What I am trying to do is plot the mopnthly invoices and costs for a given year but I am limited on doing this in PowerBI due to the poor source table setup. Ideally, I'd have a seperate table for billings, seperate table for costs etc. I don't have access to change the source layout in SQL.
At a stretch, I figure I could turn off live query and do some external manipulation of the tables in Access or a seperate SQL instance and import that, but thought I'd see if anyone else had ideas first
Thanks
3 Replies
- mike_honeyMemorable Member
I would turn off direct query and import the data, but I would transform it in the PBI Desktop Edit Queries window. The Unpivot Columns transformation will twist those columns into rows, to get a useable result. Queries can also automate splitting the column headers and converting the Months into dates.
- MattAllingtonCommunity Champion
Do what Mike says, then
You can unpivot the data and convert it to your needs
select the first 2 columns and then select "unpivot other columns"
Split the attribute column in the "_" character
replace b with billing
replace c with cost
rename columns
this gives a data table that is more usable. Does this make sense?
- aob1Frequent Visitor
Thanks guys. I get the concept, I'll have a play with the unpivot feature and see what I can come up with. Much appreciated