Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
Hi,
I have the following formula to calculate the running total (cumulative) of a measure ([Left CoE]) here, over the quarter:
HI @Anonymous ,
Maybe you can try to use RIGHT function and year field to calculate rolling on the quarter field:
Cumulative on Quarter =
VAR currYear =
MAX ( VW_CoE_OnBoard_Attrition[Year] )
VAR currQuarter =
MAX ( VW_CoE_OnBoard_Attrition[Quarter] )
RETURN
CALCULATE (
SUM ( VW_CoE_OnBoard_Attrition[Left CoE] ),
FILTER (
ALLSELECTED ( VW_CoE_OnBoard_Attrition ),
VW_CoE_OnBoard_Attrition[Year] = currYear
&& RIGHT ( VW_CoE_OnBoard_Attrition[Quarter], 1 ) <= RIGHT ( currQuarter )
)
)
Regards,
Xiaoxin Sheng
Thank you for the reply @v-shex-msft!
The SUM() function throws an error. Please note that [Left CoE] is a measure, and not an existing column.
HI @Anonymous ,
Maybe you can try to use sumx to invoke that measure:
Cumulative on Quarter =
VAR currYear =
MAX ( VW_CoE_OnBoard_Attrition[Year] )
VAR currQuarter =
MAX ( VW_CoE_OnBoard_Attrition[Quarter] )
RETURN
SUMX (
FILTER (
ALLSELECTED ( VW_CoE_OnBoard_Attrition ),
VW_CoE_OnBoard_Attrition[Year] = currYear
&& RIGHT ( VW_CoE_OnBoard_Attrition[Quarter], 1 ) <= RIGHT ( currQuarter )
),
[Left CoE]
)
Regards,
Xiaoxin Sheng
Hi @Anonymous
Try the formula below.
QTD =
CALCULATE(
[Sales],
DATESQTD( 'Calendar'[Date] )
)
Hi @Mariusz,
Thanks for your reply!
DATESQTD throws an error. How do I use the formula with the [Quarter] field instead of the [Date] field?
Thanks!
Hi @Anonymous
The best practice is to design a model with Date Dimension / Calendar Table so you can use built-in time intelligence functions.
However, if you think in your case it's impossible to do so, then please can you provide a small data sample?
Many Thanks
Mariusz
Thank you @Mariusz!
I use the DirectQuery mode, therefore time intelligence functions wouldn't work. 😞 Also, I do not have a date field in this particular table, I would have it use it from other tables.
Here's a sample of the data:
WWID | CW Name | Country | Supplier | Sponsor Name | CoE Discipline | Active | OnBoarded | YearMM | Year | Quarter |
1001 | ABC | India | XYZ | Priya | A | 1 | 0 | 201906 | 2019 | Q2 |
1002 | DEF | India | YZX | Hema | B | 0 | 0 | 201903 | 2019 | Q1 |
1003 | DDD | India | YZX | Hema | C | 1 | 0 | 201903 | 2019 | Q1 |
1004 | YYY | India | XYZ | Priya | A | 0 | 0 | 201906 | 2019 | Q2 |
1002 | DEF | India | YZX | Hema | B | 1 | 0 | 201903 | 2019 | Q1 |
1005 | XXX | India | XYZ | Priya | B | 0 | 0 | 201908 | 2019 | Q3 |
1004 | NNN | India | XYZ | Priya | C | 0 | 0 | 201908 | 2019 | Q3 |
1004 | BBB | India | XYZ | Priya | B | 1 | 0 | 201911 | 2019 | Q4 |
1004 | CCC | India | XYZ | Priya | A | 0 | 0 | 201911 | 2019 | Q4 |
1004 | ZZZ | India | XYZ | Priya | A | 1 | 0 | 201906 | 2019 | Q2 |
I have a measure [Left CoE] that filters on [Active] = 0.
I am trying to get the running totals of [Left CoE] over each quarter, Q1, Q2, Q3, Q4, separately, in order to create a graph like this:
Please let me know if my query makes sense!
User | Count |
---|---|
124 | |
108 | |
99 | |
62 | |
62 |
User | Count |
---|---|
137 | |
115 | |
102 | |
71 | |
61 |