I have encountered some difficulty in getting the behavior I seek from the QlikView I'm working on.
The exercise is to analyse the workflow concerning the approval of invoices in a large corporation. The most important element they want to focus on is how many invoices that are not paid on time and how they can improve this.
I have included a QVW file with a simplified setup and only very few records and fields. The main part is the Invoice straight table. It contains the following columns: Unit (the business unit), Invoice (count of how many invoices that unit has handled), Paid on time (how many of these are paid on time, listed in the data file as a simple boolean Y/N, found via regular Set Analysis) and an On time % (calculated percentage) column.
Now for the issue:
The client would like to have three categories created that are Invoices with an On time % of less than 50%, Between 50% and 70% and above 70%. That chart should be a pie chart and show the amount of Invoices (Count) that belong to those categories.
Please note that in the real dataset there are more than 80,000 Units (and it's growing) and more than 30 million records a year.
There is also a lot of Approval information, organization hierarchies and so on but I would think that the included file should give an overview of what I'm trying to do. I have made a mock-up of the chart I'm going to need (labelled Fake Pie Chart).