Starting December 3, join live sessions with database experts and the Microsoft product team to learn just how easy it is to get started
Learn moreGet certified in Microsoft Fabric—for free! For a limited time, get a free DP-600 exam voucher to use by the end of 2024. Register now
I have this table:
I am attempting to put together a formula that looks at the latest month for a record and calculates the sum of revenue for that month and multiply it by 12. So in this case, the latest month recorded for revenue is December 2017 which would total 71.28 (271.28-200.00) and multiplying that by 12 would give 855.36 as the projected sum going forward. How can I set this up with the assumption that going forward in each month, there will be new revenue dates and amounts when refreshed? Thanks!
So there is no confusion beforehand, here is a look at my setup:
Account_c from Delta Account and Related_Account_c from Client Revenue were appended into a single table as the unique ID so I can work with fields from both tables. Cross filters on both tables are set to 'Both'. If I do not set these to 'Both', I don't get individual revenue for each account like I'm looking for:
Wrong (single cross filter):
Correct (Both cross filter):
Now, to the formula. I am looking to build a formula based upon the revenue date that looks back two months from the current month, gathers the sum of revenue for that month (based on Revenue_Date_c) and multiplies that number by 12. So in plain terms since it is March, it would be:
(Sum of January Revenue) *12
Once it is April it will then be:
(Sum of February Revenue) *12
etc. etc.
So looking at one company for example:
(Working with December revenue until database is updated to bring in January numbers) The sum of revenue for December would be 71.28 (271.28-200.00) and multiplying that by 12 is 855.36 which is the result I'm looking for. How can I create a formula that gives this?
Well, you can get your EndDate for your calculation like this:
EndDate = EOMONTH(EDATE(TODAY(),-2),0)
And your StartDate like this:
StartDate = EOMONTH(EDATE([EndDate],-1),0)+1
So, you could use these as your filters to calculate 2 months ago from whatever TODAY is and then multiply by 12.
So, what is going on is that you are going back 2 months from today and then getting the end of the month for that month (EndDate). StartDate takes that date, subtracts a month, finds the end of that month and then adds 1 day to give you the start of the month of EndMonth. Lots of different ways of calculating this but this method ensures that you can span years without a problem.
@Anonymous
Try this MEASURE
Measure = VAR lastmonth = MONTH ( LASTDATE ( Table1[Revenue_Date__c] ) ) VAR lastyear = YEAR ( LASTDATE ( Table1[Revenue_Date__c] ) ) RETURN CALCULATE ( SUM ( Table1[Revenue_Amount__c] ), YEAR ( Table1[Revenue_Date__c] ) = lastyear && MONTH ( Table1[Revenue_Date__c] ) = lastmonth )
Hi @Anonymous,
@Zubair_Muhammad's solution should work for you.
Then based on your screenshot, it seems like you are creating a calculated column not a measure. Right? Please refer to following sample, and you can find out the difference between calculated column and measure.
By the way, they are using the same expression:
Thanks,
Xi Jin.
@Anonymous
Could you share your file?
@Zubair_Muhammad Unfortunately I don't think I can since my datasets contains company credentials to our CRM. What could be the issue is how the relationships are set up? Data from two different datasets are connected via a unique ID table and the security filter on both tables is set to 'Both'. The data I'm showing comes from two different places connected by the ID's highlighted:
Starting December 3, join live sessions with database experts and the Fabric product team to learn just how easy it is to get started.
March 31 - April 2, 2025, in Las Vegas, Nevada. Use code MSCUST for a $150 discount! Early Bird pricing ends December 9th.
User | Count |
---|---|
94 | |
92 | |
83 | |
71 | |
49 |
User | Count |
---|---|
143 | |
120 | |
110 | |
59 | |
57 |