4 Replies Latest reply on Apr 28, 2018 12:00 PM by Chad Bateman

    Sequential Filters/ Heat Map Question

    Chad Bateman


      This is my first time posting here- I hope what I’m writing here makes sense, and that my question posting etiquette is OK


      I have a heat map that shows 2 dimensions- “Defect” (rows) and “Team” (columns). There are ~21 types of defects and 11 teams.


      On this data there are is a filter named “Process”; only one “Process” should ever be selected at once when this report is in production.


      I want to add this chart to a dashboard (it will take up 1/4th of a view), but it is too large/dense if I add it unfiltered. I want to put and emphasis on the most impactful combinations of items. I’m open to other ideas, but generally I was thinking of one of the following approaches:

      • Filter by the top 7 teams with the most defects AND by the top 7 Defects that occur the frequently (individual filters); filter out everything else
        • OR
      • Filter by the highest volume LOB/Defect combinations so that only the highest numbers that make up the top 75% or top 50% appear,  and everything else is filtered out


      When trying to implement this myself I ran into problems where the correct numbers of rows and columns would appear UNTIL I actually apply to “Process” filter. After applying the "Process" filter fewer than 5 defects or teams would appear in the chart. 


      I know that this is being caused by the filter order of operations, but I can’t figure out how to fix it. I’ve been playing with LOD/ Fixed calculations, Index calculations, and adding different fields to context. I’ve been working on this for hours and have done tons of research the forums and elsewhere. I just can’t figure it out.


      Attached is a workbook of desensitized/ generic data- I’m trying to accomplish what you see on the “Top 7” tab, but with the “Process” filter applied.


      Thanks in advance for any assistance that the forum is able to provide.