How to Rank and get list of user for making segments dyanamic

[Anonymous]

Hi , I am trying to make the user list dyanamic based on their frequency.

Screenshot What I trying to get:

In above image my ranking function is not working 

ie 

FreqeuncyRanking = 
	Var r = RANKX(Fact_GA_DATA,[% GT_Frequency],,DESC,Skip)
return r

My Frequency_GT% is like :

% GT_Frequency = 
VAR CurrentFrequency = [M_Frequency]
VAR TotalFrequency =
	CALCULATE(
		[M_Frequency],
		REMOVEFILTERS(Fact_GA_DATA)
    )
RETURN 
    DIVIDE(
		CurrentFrequency,
		TotalFrequency
	)

 Let me know how can I make those segments dyanamic?

Thanks

[v-eqin-msft]

Hi @Anonymous ,

Sorry for my misunderstand... You could use SUMX() instead of SUM( )

SUMX('Fact_GA_DATA',[% GT_Frequency measure])

If you are still confused about it, please provide me with more details about your table and your problem or share me with your pbix file from your Onedrive for Business. Thank you

Best Regards,
Eyelyn Qin

[Anonymous]

Hi @Eyelyn9 , Many thanks for your help, but I am still facing some basic issues.

Part-1 :

I am trying to extract the userID(MUID) but everything is coming under last seg-D as the _sum is 1 for all the MUID.

ExtractSegments(Measure) :

VAR _sum =
    CALCULATE (
        SUMX(Fact_GA_DATA,[% GT_Frequency] ),
        FILTER (
            ALL ( 'Fact_GA_DATA' ),
            [FreqeuncyRanking] <= MAXX(Fact_GA_DATA,[FreqeuncyRanking] )
        )
    )
RETURN
    IF (
        _sum <= 0.25,
        "Seg-A",
        IF ( _sum < 0.5, "Seg-B", IF ( _sum < 0.75, "Seg-C", "Seg-D" ) )
    )

Screenshot:

----------------------------------------------------------------------------------------------------------

Part-2: How can I make Segmentations, based on Custom columns? or by using Measures?

As I have large dataset(16M rows) per year, Currently I am trying to do it on small data sample. can you guide me on how to do this task proplery. my task is to extract the users per segmentation for the selected timeframe.

Link for sample data : SegmentationWorkOct2020.pbix

https://drive.google.com/drive/folders/1fVrKuN044YP1YoBVPf7goB8xcFHW8Omo?usp=sharing

[v-eqin-msft]

Hi @Anonymous ,

According to my understand, you want to set a flag :Seg-A, Seg-B, Seg-C, Seg-D for the Top X of %GT_Frequency when it is less than 25%,50%,75%,100%, right?

You could use the following formula:

Measure =
VAR _sum =
    CALCULATE (
        SUM ( Fact_GA_DATA[% GT_Frequency] ),
        FILTER (
            ALL ( 'Fact_GA_DATA' ),
            'Fact_GA_DATA'[FreqeuncyRanking] <= MAX ( 'Fact_GA_DATA'[FreqeuncyRanking] )
        )
    )
RETURN
    IF (
        _sum <= 0.25,
        "Seg-A",
        IF ( _sum < 0.5, "Seg-B", IF ( _sum < 0.75, "Seg-C", "Seg-D" ) )
    )

My visualization looks like this:

Here is the pbix file.

Did I answer your question ? Please mark my reply as solution. Thank you very much.
If not, please upload some insensitive data samples and expected output.

Best Regards,
Eyelyn Qin

[Anonymous]

@Eyelyn9 , Thanks for helping me, but I couldn't use your logic as it can't work on measure for this part.

SUM ( Fact_GA_DATA[% GT_Frequency] )

Fact_GA_DATA[% GT_Frequency] is a Measure ,its not a column. so sum is not working here.

My Measure formula is :

% GT_Frequency = 
VAR CurrentFrequency = [M_Frequency]
VAR TotalFrequency =
    CALCULATE(
        [M_Frequency],
        REMOVEFILTERS(Fact_GA_DATA)
    )
RETURN 
    DIVIDE(
        CurrentFrequency,
        TotalFrequency
    )

Can you help me here?

[amitchandak]

@Anonymous , Try like

FreqeuncyRanking =
return RANKX(allselected(Fact_GA_DATA[MUID]),[% GT_Frequency],,DESC,Skip)

or

FreqeuncyRanking =
return RANKX(allselected(Fact_GA_DATA[MUID]),[% GT_Frequency],,DESC,Skip) +rand()/100

[Anonymous]

Thanks alot @amitchandak , can you explain me how to group users for the 1st 25% of ' %GT_Frequency'

Steps can be :

1. I want to sum(%GT_Frequency) till it reaches 25% of the total Frequency.

2. Select above users and assign them a Segment name like Seg-A