Running Total Issue with blank group (bins)

[Joram]

Dear,

Problem: Running Total metric is not correct for blank values.

I created a running total metric:

#RT_Abandoned = CALCULATE(COUNT(OneQueueCalls[MED_SEG_NUM]),OneQueueCalls[Calls_Abandoned] <> 0,filter(ALLSELECTED(OneQueueCalls),OneQueueCalls[Abandoning_Delay (bins)]<=MAX(OneQueueCalls[Abandoning_Delay (bins)])))

Giving this result:

The raw data for abandoning delay are seconds, which are grouped using:

According to several similar topics the used metric is the solution. Except it doesn't work with group bins.

I hope you can help.

[v-rzhou-msft]

Hi @Joram 

I create a sample and have a test. 

My Sample:

I don't know what does [Calls_Abandoned] looks like. But I think "OneQueueCalls[Calls_Abandoned] <> 0" caused your issue. Here I delete OneQueueCalls[Calls_Abandoned] <> 0 and get correct result.

#RT_Abandoned = 
CALCULATE (
    COUNT ( OneQueueCalls[MED_SEG_NUM] ),
    FILTER (
        ALLSELECTED( OneQueueCalls ),
        OneQueueCalls[Abandoning_Delay (bins)]
            <= MAX ( OneQueueCalls[Abandoning_Delay (bins)] )
    )
)

 Result is as below.

Best Regards,
Rico Zhou

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[TheoC]

Hi @Joram 

I believe the "OneQueueCalls[Calls_Abandoned] <> 0" in your measure may be causing it. Can you switch the 0 to BLANK() or something of the nature to include 0s?

Theo

[Joram]

Thanks for the response,

I tried "OneQueueCalls[Calls_Abandoned] <> BLANK()", but it didn't help.

[v-rzhou-msft]

Hi @Joram 

You can check my sample file and find the difference between yours and mine. I need to know your data model, please share a sample without sensitive data with me. This will make it easier for me to understand your requirement.

Best Regards,
Rico Zhou

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[Joram · Accepted solution]

All,

Thanks for searching a solution.

I removed the visualization filters via the filter pane, and added them in the RT measure itself. I cannot explain why, but this solved the issue I had.