Stephen Few wrote an excellent book entitled Information Dashboard Design: The Effective Visual Communication of Data. He sets up many criteria for effective visual communication of data. For this post, I’m using his 13 Common Mistakes in Dashboard Design to evaluate the pictured dashboard that I found, fairly randomly, on the Dashboard Spy.

The below dashboard was created by McAfee in 2004. My purpose in evaluating this particular dashboard is not to be critical of McAfee, their design may have suited its purpose. I use it only as a sample on which to apply analysis that we should apply to our current dashboard designs.

Purpose of Dashboard:

When McAfee Inc. recently introduced its ProtectionPilot software—a “dashboard”-type management console for its Active VirusScan SMB Edition and Active Virus Defense SMB Edition suites—the trial downloads were fast and furious: In the first 10 weeks after release, more than 20,000 users went online to get a copy.

No surprise there; after all, McAfee is a leader in anti-virus technology, and the ProtectionPilot tool promised to relieve a lot of headaches for administrators of small to medium businesses.

In a nutshell, these sysadmins can use ProtectionPilot to manage antivirus settings on all the PCs in their domain, and to detect intrusions and make repairs quickly. These target users typically manage 50 to 500 machines with tiny, maxed-out IT departments; McAfee’s stated goal with ProtectionPilot is to reduce antivirus management time to one hour or less per week.

Description via SoftwareCEO

Evaluation Areas:

Display Honors the Boundary of a Single Screen

Fail

Display is fragmented in discrete screens to which one must navigate. The screen requires scrolling. As a result the user loses the ability to see meaningful relationships.

Supplying Adequate Context for the Data

Fail

Measures of what’s currently going on in the business rarely do well in isolation. It is more valuable to include comparison, whether the result is good or bad, how good or bad, are we on track, are we doing better than in the past, or better than the forecast

Displaying Appropriate Detail or Precision

Partial Pass

It appears that the appropriate precision is being used.

Choosing an Appropriate Measure

Partial Pass

For a measure to be meaningful, we must know what is being measured. A measure is deficient if it isn’t the one that most clearly and efficiently communicates the meaning that the dashboard viewer should discern. Units are not clearly described. I’m guessing that the upper data concerns the number of computers, and that there are 20 of them. Under detection, I am not sure what is being detected, but since the total is 165, I have to assume that computers can have more than one of them.

Choosing an Appropriate Display Media

Fail

Pie charts display quantitative information ineffectively. People are poor at comparing two-dimensional areas or angles very accurately. Bar graphs are almost always better way to display this information. It is interesting that for the top, a pie chart was used. The bottom uses a bar chart. I believe this was for visual diversity rather than some data display reason. Additionally, positioning the bar graph horizontally would place the labels right next to their results. This data may not require a graph at all. A table might be sufficient for this type of data. It is very easy to see that 80% is larger than 10%. The graph adds very little additional data, but takes up a lot of space.

Avoiding Meaningless Variety

Fail

As stated previously, there appears to be no reason for the different display methods between the top and bottom display.

Use of Effective Display Media

Fail

A legend was used to label and assign values to the slices of the pie. This forces our eyes to bounce back and forth between the graph and the legend to glean meaning. It is more effective to label the slices and corresponding labels in size order. The color choices of the graphs is effective. It does not produce sensory overkill.

Encoding Quantitative Data Accurately

Pass

The scale of the graph components seem to accurately represent the data.

Effective Arrangement of the Data

Fail

Most examples of dashboards on the Web are composed of a small amount of data to avoid the need for skilled visual design. There is too little useful data presented here.

Important Data is Highlighted

Fail

When you look at a dashboard, your eyes should immediately be drawn to the information that is most important. In this design, everything is visually prominant, and consequently nothing stands out.

Avoiding the Use of Useless Decoration

Pass

There is not a lot of distracting and useless decoration. There is a lot of screen real estate that is not used for conveying information.

Appropriate Use of Color

Partial Pass

The top graph uses color well as the color can be interpreted as stop light colors (up-to-date is green and not up-to-date is red). Numerical values are provided for people who do not see color. When colors in two different sections of a dashboard are the same, we are tempted to see them as related to one another. I do not know enough about the functional area to know whether this would be an appropriate assumption or not.

Designing an Attractive Visual Display

Pass

I think the display is clean and attractive.

Conclusion

There is much that could be fixed in this dashboard if our goal is the communication of information from which people could make decisions. In later posts, I will show an alternative design that would be more effective. [Note: A reader pointed out that I haven't gotten around to an alternate design yet. I will post a link to that as soon as I get this done.]

Please let me know your comments.

Comments