This is best Fabric, Power BI, SQL and AI community event. How do we know? The last event sold out! Save €200 with code FABCMTY200.
Register nowA new Data Days event is coming soon! This time we’re going bigger than ever. Fabric, Power BI, SQL, AI and more. Don't miss out.
What I am trying to do is get a sum of measure CntVariance (count difference from currentmonth-previousmonth) by a group of existing columns so I can generate SumCntVariance (sum within the group) and eventually work my way to an average count variance (AvgCntVariance) within that same group. The catch is I would really not like to include November because its the initial run, so the differences in cnts should not be used as its the initial load of data.
Here an example of some data:
| BuildPhase | MetricDescription | CntVariable | ReleaseMonthYear | Month_YY.Year | Month_YY.Month | Cnt | PreviousMonthCnt | CntVariance |
| 1 | A | Records | 202211 | 2022 | November | 2218267 | 2218267 | |
| 1 | A | Records | 202212 | 2022 | December | 2400358 | 2218267 | 182091 |
| 1 | A | Records | 202301 | 2023 | January | 2408373 | 2400358 | 8015 |
| 1 | A | Records | 202302 | 2023 | February | 2419882 | 2408373 | 11509 |
| 1 | A | Records | 202303 | 2023 | March | 2457902 | 2419882 | 38020 |
| 1 | A | Records | 202304 | 2023 | April | 2471238 | 2457902 | 13336 |
| 1 | B | Male | 202211 | 2022 | November | 3502 | 3502 | |
| 1 | B | Male | 202212 | 2022 | December | 2975 | 3502 | -527 |
| 1 | B | Male | 202301 | 2023 | January | 2610 | 2975 | -365 |
| 1 | B | Male | 202302 | 2023 | February | 4317 | 2610 | 1707 |
| 1 | B | Male | 202303 | 2023 | March | 4570 | 4317 | 253 |
| 1 | B | Male | 202304 | 2023 | April | 4591 | 4570 | 21 |
| 1 | B | Female | 202211 | 2022 | November | 3402 | 3502 | |
| 1 | B | Female | 202212 | 2022 | December | 2777 | 3502 | -527 |
| 1 | B | Female | 202301 | 2023 | January | 2929 | 2975 | -365 |
| 1 | B | Female | 202302 | 2023 | February | 5002 | 2610 | 1707 |
| 1 | B | Female | 202303 | 2023 | March | 4243 | 4317 | 253 |
| 1 | B | Female | 202304 | 2023 | April | 4520 | 4570 | 21 |
| 1 | C | MemberID | 202211 | 2022 | November | 18237 | 18237 | |
| 1 | C | MemberID | 202212 | 2022 | December | 31011 | 18237 | 12774 |
| 1 | C | MemberID | 202301 | 2023 | January | 8014 | 31011 | -22997 |
| 1 | C | MemberID | 202302 | 2023 | February | 11509 | 8014 | 3495 |
| 1 | C | MemberID | 202303 | 2023 | March | 9990 | 11509 | -1519 |
| 1 | C | MemberID | 202304 | 2023 | April | 13336 | 9990 | 3346 |
My MetricDescription is a QA metric, CntVariable is can be things like , Distinct MemberID, Records, "Y"/"N", "Male"/"Female" ect. I have added a date table that maps the release calle ColumnStackDates where ReleaseMonthYear can be tied to date fields
Releationship is that ReleaseMonthYear=ReleaseCycle in the table below. I am including this in the question because it may be relevant to reason it may not be working:
Relationship set up:
I tried making this a one to one but it freaked out at me an told me that the relationship was invalid.
So I stuck with the one to many and moved on trying to get my groupings and my SumofCntVariance a few difference ways:
I initially tried this equation:
SumofCnt = SUMX(
FILTER(
ALLSELECTED('metadata MetricsTable'[BuildPhase], 'metadata MetricsTable'[MetricDescription], 'metadata MetricsTable'[CntVariable]),
NOT(ISINSCOPE('ColumnStackDates'[Month_YY]) && MONTH('ColumnStackDates'[Month_YY]) = 11) ),
CALCULATE
( [CntVariance],
ALLSELECTED('metadata MetricsTable'[BuildPhase], 'metadata MetricsTable'[MetricDescription], 'metadata MetricsTable'[CntVariable]),
NOT(ISINSCOPE('ColumnStackDates'[Month_YY]) && MONTH('ColumnStackDates'[Month_YY]) = 11) ) )
DAX freaks out saying it could not find [Month_YY], but only in that one area in red above, everywhere else I use [Month_YY] is kosher.
In trying a work around I decided to just use ReleaseCycleMonth
@ssandstrom In a measure, if you reference a column directly you need to wrap it in an aggregation like MAX, MIN, SUM, etc.
Check out the May 2026 Power BI update to learn about new features.
Sign up to receive a private message when registration opens and key events begin.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
| User | Count |
|---|---|
| 26 | |
| 25 | |
| 22 | |
| 19 | |
| 17 |
| User | Count |
|---|---|
| 42 | |
| 41 | |
| 40 | |
| 21 | |
| 20 |