TopN with Multiple Dimensions

Hi there clever people!

I have a ranking situation that I am trying to solve and have approached it from 5 different ways and reach the same result every time, so I may be about to go insane. Hopefully, someone can help!

I am trying to Rank the Top 5 retail stores by WOS (weeks of Supply) using TopN (but I have tried Rankx as well without success) that is a simple calculation of average stock on hand divided by the average sales over the past 4-weeks.

The initial ranking works well (I can see the top 5 stores) but it falls apart when I bring in a second dimension.

I need to see the weeks of supply for these top 5 stores with the associated product price tier, however, as soon as I add the price tier to the visual I get other stores too because it pulls in the top 5 price tiers.

The model can be found here: https://drive.google.com/file/d/1vnNplHUeSV2o19D08bFDou6EBoPNxyjO/view?usp=sharing

Please let me know if you need more information to help.

Many thanks

Phil

Hi @PhilBowen,

Welcome to the forum!

Thank you for providing so much detail including a sample PBIX :+1:
So this is quite tricky indeed… can you let me know if this is what you were expecting?

Rank Stores = 
VAR Top5Stores =
    CALCULATETABLE (
        TOPN (
            5,
            CALCULATETABLE ( VALUES ( Locations[Store Name] ) ),
            [Weeks of Sales]
        ),
        ALLSELECTED()
    )

RETURN
    CALCULATE (
        [Weeks of Sales],
        KEEPFILTERS ( Top5Stores )
    )

With this result.

.
I hope this is helpful.
eDNA - Top 5 Problem.pbix (528.3 KB)

3 Likes

Thank you very much Melissa,

I was so close but oh so far!

You rock!!!

Cheers,

Phil

A response on this post has been tagged as “Solution”. If you have a follow question or concern related to this topic, please remove the Solution tag first by clicking the three dots beside Reply and then untick the check box. Thanks!"