The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredCompete to become Power BI Data Viz World Champion! First round ends August 18th. Get started.
Greetings,
I'm trying to create a measure with an if statements and some actions.
For example I am looking to create a measure that checks if date column is blank on a row, then that row is skipped.
Name Date
John 5/5/2017
Alice 6/1/2017
Ralf (blank)
Jeff 8/9/2017
Here, Ralf is skipped since there is no date in that row and moves on to Jeff.
Solved! Go to Solution.
Here is a mock calculation with some sample data I used:
Measure Value = CALCULATE(COUNT(Data[Date]),Data[Date]<>BLANK())
I think I would need more context into what you are trying to compute. Getting DAX/Power BI to move over the rows with (Blank) values is easy enough, but providing a comprehensive solution requires more information from you.
@CamFlawless wrote:I think I would need more context into what you are trying to compute. Getting DAX/Power BI to move over the rows with (Blank) values is easy enough, but providing a comprehensive solution requires more information from you.
This is basically to show an amount of system incidents occurred over a period of 1 month (all this is an example). Then throw it into a bar graph that shows the number of issues on that month but skips those without a date. so if there are total of 20 rows and 3 have empty dates then the graph would show 17 instead of 20.
Here is a mock calculation with some sample data I used:
Measure Value = CALCULATE(COUNT(Data[Date]),Data[Date]<>BLANK())
Hi,
I think you do not even need the filter, because COUNT counts only values that are not blank.
So :
COUNT( Data[Date] )
should give you the expected result.
Have a nice day
User | Count |
---|---|
86 | |
84 | |
35 | |
35 | |
35 |
User | Count |
---|---|
94 | |
79 | |
63 | |
55 | |
52 |