visual-literacy.org: Business

Visualizing Indicators

Example: Situation: A financial controller in a globally acting organization in the production industry has to regularly control and report about the IT budgets. To do so, he has a powerful information system which allows to create various management reports and the export of data to excel. The problem of the controller was the missing overview (due to the large amount of IT-budgets) and missing filtering techniques.

Solution: We have chosen the tree map approach to visualize the data due to its wide acceptance and because it supports the principles of human perception. To do so, a customized software was implemented in Java. A screenshot of this software can be seen in Figure 6. In this tree map each rectangle represents one IT-budget (e.g., licences, depreciation). The hierarchy represents the organizational structure of the company. Each section represents a country, which is then subdivided into cost centers and individual deaprtements. Each rectangle varies in the size of the area and the color of this area. These two parameters can be linked to the controlling data. One possibility is that the size of the rectangle correlates with the size of the IT budget, and the color of the rectangle with the current status of the budget, where a green color means below and red above the budget, while the color gradients between green and colors mean that the used money is in track with the budget. For ten budgets a traditional Excel list would be much more easy to compare the ten budgets, but when we think of comparing around 1000 budgets or more, as in this example, the tree map visualization can be a very helpful tool. The next step in this tool are filtering techniques were the data can be filtered with sliders. All budgets that do not fulfill a certain filtering criterion will change there color to white which allows to rapidly analyze all the budgets visually. Additionally a keyword filter allows the filtering of all information with keywords. An example is the filtering of all budgets with a name, such as the budget owner.

Evaluation: The software is an adapted version of an existing tree map tool which was previously implemented as an additional tool for Excel users. Therefore the modifications where easy to implement and consisted of adjustments and tailoring to the structure of the data and the filtering techniques. The part which was a bit more tricky was the automatic linkage to the data warehouse. However, after two weeks of time the controller had in his management information systems a new button with the label “Start Tree Map”. Clicking on this button started the application and loaded the actual data for explorations. The controller reported that the tool was very easy to understand and use. However he also reported that it took him about a week to find out in which way he could integrate the new approach in his reports and communication processes. He basically used it personally for exploratory tasks and presented the findings with traditional charts. However, he regularly also explained in meetings for interesting persons in the projecte treemap that he found out specific patterns which should be taken serious.

Conclusion: The treemap tool is a simple and powerful tool that can be used by every person that is familiar with Excel and import and export techniques. It can be used by controllers, project managers, researchers, who have datasets with (1) around 50 to 3000 elements (as the IT-budget), (2) a tree structure (as the regional structure), (3) with large numeric difference (e.g., the small and very large budgets), (4) few elements that can be represented with the colors (e.g., above or below budget), and (5) additional data that makes sense for the numeric or textual filtering techniques (e.g., budget owner, year).

(Shneiderman, 1992, Johnson and Shneiderman, 1991)