When explaining the results of data analysis, it's easy to say, "80 percent of the losses came from 20 percent of the causes." But reinforcing this narrative with a data visualization can really send home that message.
The 80/20 rule or Pareto's principle developed by Vilfredo Pareto is often applied to represent the significant few versus the trivial many. Graphically, a Pareto diagram simply displays a bar chart distribution of a metric by a dimension sorted in descending order of the value, overlaid with a line chart plotting the growing cumulative percentage. It helps to identify the significant few. A Pareto chart can often help in identifying top performers for a metric and their overall contribution across the categories.
Subscribe to the Oracle Analytics Advantage blog and get the latest posts sent to your inbox
Our team recently added a custom data visualization plugin for anyone wanting to take advantage of Pareto's principle in Oracle Analytics. We're going to show you a few key steps in this article to help you get up and running quickly.
A Pareto Chart requires two mandatory input parameters: A Category dimension (the X axis) and a Values object (the Y axis). Color is an optional parameter. The graphic below shows how Values (Y-Axis) accepts a single measure. Category (X-Axis) accepts one or more attributes.
The plugin automatically sorts the metric value in descending order, by dimension members, and plots it as a bar chart for each member. The metric value is computed as a cumulative percentage and plotted as a line chart. The axis of cumulative percentage is placed on the right-hand side. Here's an example of a Pareto principle depicting Sales by City:
Adding a color parameter changes the bar chart to a stacked bar chart with each bar divided into different color categories. However, the line chart is not altered.
As you hover over the bars, you can see a tool-tip with X-Axis and Y-Axis values. When you hover the mouse on the line chart, along with the tool-tip, you see two dotted lines each pointing towards X-Axis and Cumulative percent. These lines are helpful in dividing the category with Cumulative percent. You can see the stacked bar and the dividing lines in the graphic below.
The Oracle Analytics Pareto Chart Plugin can plot negative values as well. Here's an example of a Profit metric with positive and negative values.
The plugin also supports typical mouse right-click interactions in a data visualization such as Keep/Remove Selected, Use as Filter, Brushing, and so on.
If you are an Oracle Analytics customer or user, we'd love to hear from you. Leave a comment below and let's get that conversation started. Remember, this and other custom data visualization plugins can be downloaded for free from the Oracle Analytics Library.