Don't miss your chance to take the Fabric Data Engineer (DP-700) exam on us!
Learn moreWe've captured the moments from FabCon & SQLCon that everyone is talking about, and we are bringing them to the community, live and on-demand. Starts on April 14th. Register now
Hello All,
I am new to Power BI and stuck at a problem:
Attached is an screenshot of an example sample data. I need to find out the count of users who 'only' performed event 'B'
So, as per the example given, the output should be one user, because user 3 performed only event 'B'.
Any idea how can I write DAX query for this problem?
Solved! Go to Solution.
Hi @Anonymous ,
You could create a flag measure:
Flag =
var _dis=CALCULATE(DISTINCTCOUNT('Table'[Event]),ALLEXCEPT('Table','Table'[User]))
var _countB=CALCULATE(COUNTROWS('Table'),FILTER('Table',[User]=MAX('Table'[User]) && [Event]="B"))
RETURN IF(_dis=1 && _countB>0,1,0)
And then apply it to visual-level filter pane, set as "=1":
Best Regards,
Eyelyn Qin
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi @Anonymous ,
Did my method help you ? Please feel free to ask.
Best Regards,
Eyelyn Qin
Hi @Anonymous ,
Did my method help you ? Please feel free to ask.
Best Regards,
Eyelyn Qin
Hi @Anonymous ,
You could create a flag measure:
Flag =
var _dis=CALCULATE(DISTINCTCOUNT('Table'[Event]),ALLEXCEPT('Table','Table'[User]))
var _countB=CALCULATE(COUNTROWS('Table'),FILTER('Table',[User]=MAX('Table'[User]) && [Event]="B"))
RETURN IF(_dis=1 && _countB>0,1,0)
And then apply it to visual-level filter pane, set as "=1":
Best Regards,
Eyelyn Qin
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi @tackytechtom
Thanks for your reply!
But, applying your solution gives an answer of 3, because there are three users in total who performed event = B.
I need users who performed 'only' event 'B', in this case only user '3'
Hi @Anonymous ,
My bad! I did not read your requirement properly.
How about this one here:
Measure =
VAR _helpTable =
SUMMARIZE (
Table,
Table[User],
"CountEventB", COUNTROWS ( FILTER ( Table, Table[Event] = "B") ),
"CountOtherEvents", COUNTROWS ( FILTER ( Table, Table[Event] <> "B") )
)
RETURN
COUNTROWS ( FILTER ( _helpTable, [CountEventB] >= 1 && [CountOtherEvents] = 0 ) )
The idea is to create a _helpTable variable which would return the following result:
From there it is just choosing the one that has value >= 1 in the first column and a value that equals 0 (or BLANK) in the second column.
Let me know if this helps or if you have any questions 🙂
/Tom
https://www.tackytech.blog/
https://www.instagram.com/tackytechtom/
| Did I answer your question❓➡️ Please, mark my post as a solution ✔️ |
| Also happily accepting Kudos 🙂 |
| Feel free to connect with me on LinkedIn! | |
| #proudtobeasuperuser | |
Dear @tackytechtom
Thanks for the answer, I applied the same concept as this and it worked. Please find the DAX below. Although I have two follow-up questions on this, hoping you could answer this as well,
1. What if I have the user column in different table and not in 'Data' table I tried, VAR B_Users = CALCULATETABLE ( DISTINCT ( DifferentTable[User] ), Data[Event] = "B" ) But this is giving me blank output.
2. To extend my original question, what if I want to calculate list of users who performed only event B or C or more than two events? Do I have to manually write the above DAX for each event and then count distinct user id
VAR B_Users = CALCULATETABLE ( DISTINCT ( Data[User] ), Data[Event] = "B" ) VAR Non_B_Users = CALCULATETABLE ( DISTINCT ( Data[User] ), Data[Event] <> "B" ) VAR B_Only_Users = EXCEPT ( B_Users, Non_B_Users ) RETURN COUNTROWS ( B_Only_Users )
NewTable=FILTER(VALUES(table[User]),CALCULATE(DISTINCTCOUNT(table[Event]))=1)
Hi @Anonymous ,
How about this:
Measure = COUNTROWS ( FILTER ( Table, Table[Event] = "B") )
Let mw know if this helps 🙂
/Tom
https://www.tackytech.blog/
https://www.instagram.com/tackytechtom/
| Did I answer your question❓➡️ Please, mark my post as a solution ✔️ |
| Also happily accepting Kudos 🙂 |
| Feel free to connect with me on LinkedIn! | |
| #proudtobeasuperuser | |
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
A new Power BI DataViz World Championship is coming this June! Don't miss out on submitting your entry.
Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.
| User | Count |
|---|---|
| 5 | |
| 3 | |
| 2 | |
| 2 | |
| 2 |
| User | Count |
|---|---|
| 9 | |
| 8 | |
| 7 | |
| 5 | |
| 5 |