Cognos Analytics

Do you display the data so the insight is consumable? The goal is to provide clear understanding of the information and what it is trying to achieve. Over the years I have seen many people create BI reports that generate 100s or more pages. I have no doubt that someone can justify that kind of data valuable but let’s agree that its more likely to just be an extract. What I would like to focus on are visualizations that are compelling and easily consumed. I have seen two common threads when it comes to building charts that become contentious.  The first group insists on creating a chart that may not be “in the box” and insist it’s a lacking feature of the product. That is a challenging argument to contend. An answer may be to build custom charts when the product being used does not have such a chart and supports custom charts, but you should evaluate the complexity of the chart and is it displaying the right information.

What exactly is this chart showing?

Alternatively, you could identify a way to show the same data that is easy to interpret yet use in-the-box capability so that upgrades are seamless. 

The second group are those who use charts that may not impart the correct message based on the data. Here is a simple example of a Pie chart that may be misleading if you are comparing numbers where a Bar chart would accurately portray the differences.

A consideration for either of these cases is to understand how charts should be used to display the information accurately. There are many chart options and knowing which should be used may be broken down into a few categories based on what you are trying to achieve.

1. Comparison
2. Relationship
3. Distribution
4. Composition

Here is a chart selection diagram created by Dr. Andrew Abela that could help you pick the right chart for your data. PDF version here: Chart Selection diagram.

Information can be shown in many ways and a dashboard can present many variations of the data.  The requirement is to make sure there is clear understanding of what the message is. What are we trying to highlight and what is the story telling your users. Ask yourself, are you using the right charts?