Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowData Days is here! Join us now for 60+ days of learning, challenges, and connection. Learn more
I have report I need to build but layout requested is to have month over month aggregated over 3 month period. How could i do this in PBI? With matrix, the groupings and subtotals automatically get incorporated into each column group.
see below example of how i need it
Issues i found using matrix
Solved! Go to Solution.
Hi, @Anonymous ;
Base on my understand , you could have two ways to achieve it, as follows:
Method1:
1. create a Auxiliary table(Table2)
Table2 =
var _a= SUMMARIZE('Table','Table'[cate],'Table'[date],"month",FORMAT('Table'[date],"Mmm"))
var _b=SUMMARIZE('Table','Table'[cate],"1",BLANK(),"month","MOM%")
return UNION(_a,_b)
2.create a measure2,
Measure 2 =
VAR _PER=CALCULATE(SUM('Table'[value]),FILTER(ALLEXCEPT('Table','Table'[client]) ,[cate]=MAX('Table2'[cate])))/CALCULATE(SUM('Table'[value]),ALL('Table2'))
VAR _B= IF(MAX('Table2'[month])="MOM%",FORMAT(_PER,"0%"), CALCULATE(SUM('Table'[value]),FILTER('Table',[cate]=MAX('Table2'[cate])&&[date]=MAX('Table2'[date]))))
RETURN IF(ISFILTERED('Table2'[month]), _B,SUM('Table'[value]))
The final output is shown below:
Method2:
1. add three rows in power query .
2.create a measure1
Measure1 =
VAR _mom =
CALCULATE (
SUM ( 'another'[value] ),
ALLEXCEPT ( 'another', 'another'[client], 'another'[cate] ))
RETURN
IF (
ISFILTERED ( 'another'[date] ),
IF (
MAX ( 'another'[date] ) = "MOM%",
FORMAT (_mom/ CALCULATE ( SUM ( 'another'[value] ),ALLEXCEPT ( 'another', 'another'[client] )),"0%"),
SUM ( 'another'[value] )),
_mom)
The final output is shown below:
if not right, can you share me simple file after removing the sensitive information?
Best Regards,
Community Support Team_ Yalan Wu
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi, @Anonymous ;
Base on my understand , you could have two ways to achieve it, as follows:
Method1:
1. create a Auxiliary table(Table2)
Table2 =
var _a= SUMMARIZE('Table','Table'[cate],'Table'[date],"month",FORMAT('Table'[date],"Mmm"))
var _b=SUMMARIZE('Table','Table'[cate],"1",BLANK(),"month","MOM%")
return UNION(_a,_b)
2.create a measure2,
Measure 2 =
VAR _PER=CALCULATE(SUM('Table'[value]),FILTER(ALLEXCEPT('Table','Table'[client]) ,[cate]=MAX('Table2'[cate])))/CALCULATE(SUM('Table'[value]),ALL('Table2'))
VAR _B= IF(MAX('Table2'[month])="MOM%",FORMAT(_PER,"0%"), CALCULATE(SUM('Table'[value]),FILTER('Table',[cate]=MAX('Table2'[cate])&&[date]=MAX('Table2'[date]))))
RETURN IF(ISFILTERED('Table2'[month]), _B,SUM('Table'[value]))
The final output is shown below:
Method2:
1. add three rows in power query .
2.create a measure1
Measure1 =
VAR _mom =
CALCULATE (
SUM ( 'another'[value] ),
ALLEXCEPT ( 'another', 'another'[client], 'another'[cate] ))
RETURN
IF (
ISFILTERED ( 'another'[date] ),
IF (
MAX ( 'another'[date] ) = "MOM%",
FORMAT (_mom/ CALCULATE ( SUM ( 'another'[value] ),ALLEXCEPT ( 'another', 'another'[client] )),"0%"),
SUM ( 'another'[value] )),
_mom)
The final output is shown below:
if not right, can you share me simple file after removing the sensitive information?
Best Regards,
Community Support Team_ Yalan Wu
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
@Anonymous Hard to be sure but sounds like Rolling Months: https://community.powerbi.com/t5/Quick-Measures-Gallery/Rolling-Months/m-p/391499#M124
If not, Sorry, having trouble following, can you post sample data as text and expected output?
Not really enough information to go on, please first check if your issue is a common issue listed here: https://community.powerbi.com/t5/Community-Blog/Before-You-Post-Read-This/ba-p/1116882
Also, please see this post regarding How to Get Your Question Answered Quickly: https://community.powerbi.com/t5/Community-Blog/How-to-Get-Your-Question-Answered-Quickly/ba-p/38490
The most important parts are:
1. Sample data as text, use the table tool in the editing bar
2. Expected output from sample data
3. Explanation in words of how to get from 1. to 2.
Don't miss out on Data Days, June 15 through August 7. Learn Fabric, Power BI, SQL, AI and more.
Check out the May 2026 Power BI update to learn about new features.
| User | Count |
|---|---|
| 23 | |
| 23 | |
| 21 | |
| 17 | |
| 14 |
| User | Count |
|---|---|
| 58 | |
| 50 | |
| 37 | |
| 29 | |
| 24 |