4 Replies Latest reply on Sep 14, 2016 3:27 PM by Nick Foley

    Ive been stuck on creating this Viz for a week.

    Nick Foley

      So, its a more or less a a dual axis chart.

       

      The goal:

      it shows all of our customers relative position compared to other customers. 

      The circle would be the customer, and their location along the bar chart shows where they rank relative to other customers.

       

      The Details:

      -It shows our product lines revenue by customer. Products A-E. 

      -These products will have varying numbers of customers. Product A could have 100 customers, product E could have 1000.

       

      The Issue:

      The end user wants all of the bars to be the same length. This means i think i would have to compute this chart as some type of percentile.

       

      So, i would need to create a bar that is always 100%. easy enough.

       

      But, i need to create a calculated field or something that shows only that customers Percentile relative to the other customers.  This would have to be a table calc i imagine, but i couldnt show any of the other customers.  I have no idea how to do this. make sense?

       

      Imgur: The most awesome images on the Internet

        • 1. Re: Ive been stuck on creating this Viz for a week.
          Alastair Young

          Hi Nick,

           

          The easiest way to run this is by allowing a parameter to run the selection of the customer.   Once in place you can create your bar chart as expected and then look at creating the dual axis.

           

          Firstly. We need to use the parameter to retain the Customer Value - an example would be this.

          if [Customer Name] = [Parameter 1] then [Sales] END

           

          This essentially gets you to display only one customers value.  We cannot apply a filter here on the worksheet, otherwise we wont be able to reference the total.

          Now we need to create a calculated field to compare against the overall sales value.

          Like so -

          sum([Customer Sales Calc])/sum([Sales])

           

          We can now drop this onto the worksheet (beside the already created bar chart) to create the dual axis, change the shape to which ever mark you wish and dual these across.

           

          This is simply only using the % Sales across a given dimension, but may give you some guidance.

           

           

          I have a workbook example from Superstore if you wish to look at that.


          Al

          • 2. Re: Ive been stuck on creating this Viz for a week.
            chris.moore.11

            I don't understand the problem. What is the issue with the solution from this thread? Dual Axis graph benchmarking help

             

            Attached is the example. (similar to what was described by Alastair)

             

             

            • 3. Re: Ive been stuck on creating this Viz for a week.
              Nick Foley

              Hi chris, thanks for the reply.  Ive had the flu and forgot i posted this before.

               

              The issue is that your view will show what % a customer is of the total revenue of all our customers. So, they could our best customer and be at the  bottom of the graph with 8%.

               

              It has to show where they stand relative to avoid this.  If you use max(revenue) instead, it shows how they compare to the best customer. Which also isnt that useful, because if we have one or two reallys trong customers, it will just show every one else as bad customers

              • 4. Re: Ive been stuck on creating this Viz for a week.
                Nick Foley

                Tghanks for the reply Alastair, sorry ive been sick for the past week and havent gotten around to it now.

                 

                The issue is  your graph shows what % the customer is of our total revenue. If you you max(revenue), this will show you how they compare to our best customer.  This isnt what is needed.

                 

                We need to see how essentially where a customer ranks among our customers.  This way we can see how strong of a customer they are in various products, compared to other customers.