Returning the Highest Rank Number from the Rank Measure

[cbolling]

Hi There,

I have a list of customers that is fluid, and there is a RankX ranking them based on their revenue. Lets say that there are 1 through n customers, I want to take n and multiply it by .2 and then only display customers with rank <= n*0.2. In other words I want to only display the top 20% of my customers based on revenue. What is the best way to do this?

[wdx223_Daniel]

MxRankN :=
VAR pct = 0.4
VAR MxRx =
    RANKX (
        ALL ( Sales[Customer] ),
        CALCULATE ( SUM ( Sales[Revenue] ) ),
        MAXX (
            TOPN ( 1, ALL ( Sales[Customer] ), CALCULATE ( SUM ( Sales[Revenue] ) ) ),
            CALCULATE ( SUM ( Sales[Revenue] ) )
        ),
        ASC
    )
VAR n =
    INT ( pct * MxRx )
RETURN
    SUMX (
        VALUES ( Sales[Customer] ),
        IF (
            RANKX ( ALL ( Sales[Customer] ), CALCULATE ( SUM ( Sales[Revenue] ) ),, DESC ) <= n,
            CALCULATE ( SUM ( Sales[Revenue] ) )
        )
    )

[amitchandak · Accepted solution]

@cbolling ,Not very clear. But I think this case of top percentile.

Refer if these can help

https://blog.enterprisedna.co/implementing-80-20-logic-in-your-power-bi-analysis/
https://forum.enterprisedna.co/t/testing-the-pareto-principle-80-20-rule-in-power-bi-w-dax/459
https://finance-bi.com/power-bi-pareto-analysis/
https://community.powerbi.com/t5/DAX-Commands-and-Tips/Calculate-the-sum-of-the-top-80/td-p/763156

Please provide your feedback comments and advice for new videos
Tutorial Series Dax Vs SQL Direct Query PBI Tips
Appreciate your Kudos.