Part 2: Combining Dynamic Attribute Slicer, Dynamic Measure Slicer, and Dynamic TopN /RANKX Slicer

Hey All.

Been working on this awhile, still can’t get it. Want to combine Dynamic Attribute Slicer, Dynamic Measure Slicer, and Dynamic TopN /RANKX Slicer. The goal here is to allow end users to toggle within a slicer between TopN choices for a given attribute they select and given measure they select. I mostly followed and tried to reverse engineer the logic of @sam.mckay with the Showcase Power BI Viz, specifically the Global Statistics dashboard. I also used tutorials on Disconnected Slicers from PowerPivotPro and FourMoo on disconnected slicers and such.

Currently the TopN Slicer is giving the Top10 values for all values within the AttributeSlicer. It needs to give the TopN within the specific group in that column. Look at the global stats worksheet.

I have an idea of what needs to happen. I imagine the Rank Selection isn’t able to drill down into the Top 5 for the specific attribute so some sort of a groupby or virtual calculated table needs to occur. Solving this problem would really help Power BI Datanauts create stellar dashboards. Any help? Example Power BI click here or download below.

https://drive.google.com/open?id=1pAn4Wv85beKtnyMWJ6gSmT9O9ToxRHdQ

Having a look at this again.

One thing I really recommend here is to simplify it down. Try to get the core concept for the ranking before moving ahead with the dynamic measures…it’s even confusing me just going through your model.

As mentioned in the previous post there’s a bit you can do to make this easier on yourself. Cleaning up the model for one, then create measure groups is another simple thing to do.

image

So, I got it into a table so I could see the results and then just tested out a simple formula like this using TOPN

Now I’m going to try to integrate the ranking selection

Seems to work well.

See if this is what you needed.

dynamic_attribute_measure_topn_slicer_09_17_19_example.pbix (502.1 KB)

Once you’ve simplified it like this, then you should be able to build it back up again into your more complex logic you need.

Thanks
Sam

@sam.mckay, Sorry for the late reply. I appreciate the effort here. I thought this worked originally, but looking back at it. The Top N ranking actually gets a rank of ALL AttributeSlicer[Value]'s and so when we choose a [Value] w/i the specific Attribute, this measurement gives us the Values that fall into the Top 5 of all Values. I understand not crossing the consulting barrier, so I actually really appreciate the efforts here. Working at one of the largest corporations in the world, I’ve gotten a lot of compliments on the set up I have going here, this TopN slicer would really be the cream of the crop!