The Information Cannot Speak for Itself

Massive IT investment won't matter if users are unable to make sense of the data. With businesses reliant on information, presentation can no longer be an afterthought. Here's how you can improve that first impression.

Figures 3 and 4 illustrate the difference between uninformed visual design and that which is firmly rooted in an understanding of how people visually perceive information. Despite its familiar nature, the first graph (Figure 3) misses the mark. Imagine that an analyst is using this graph to communicate to Company G's executive management team how it's doing compared with its competitors. Does the graph in Figure 3 tell the story effectively and efficiently?

If the executives studied Figure 3 long enough, they could approximate how well Company G is doing, but that's time they could be doing something about the information, rather than trying to figure it out. The analyst, sensing the inadequacy of this pie chart, might spruce it up by adding 3D. Figure 4 displays the result, which is even worse. Everyone may be dazzled by 3D visual effects, but these effects make the graphs even more difficult to read.

If you were the analyst, how might you approach this task? Remember, your purpose is to communicate. How might you effectively present the performance of your company relative to the competition? Figure 5 shows a simple solution that works quite well. There's nothing fancy about this design, yet it's quite effective. It communicates clearly and efficiently. No more than a few seconds would pass from the time executives saw this graph to the time they could begin knowledgeably discussing plans to beat the competition.

FIGURE 3 - The graph might look good, but it takes too long to figure out the information.

FIGURE 4 - The addition of 3D to Figure 3 makes the information even more difficult to read.

FIGURE 5 - Nothing fancy, but a simple and effective design.

Effective data presentation is rooted in an understanding of visual perception: What works, what doesn't, and why. Just as a good business analyst understands business and a good manager understands people, so a good BI professional must understand visual perception and the nature of meaningful information.

Visual Clutter

Let's take a look at one more example, in Figure 6, of the graphical labyrinth that we navigate daily in our frustrating quest for the data. Without this graph's title, would you have any idea that its specific purpose is to compare the sales performance of the product named SlicersDicers to the performance of each of the other products? In the general field of design, we speak of things as having "affordances": that is, characteristics that reveal how they should be used. A teapot has a handle. A door that you need to push has a pushplate. The design of something should suggest — in and of itself — how it should be used. This graph, however, relies entirely on its title to declare its purpose. Not only does its design fail to suggest its use, it undermines it.

FIGURE 6 - The graph offers visual clutter, full of details that don't communicate important data more effectively.

What's wrong with the graph in Figure 6? For one thing, it's visually cluttered. Too much is going on and much of it has nothing to do with the data. Notice that the heavy grid lines in this graph get in the way of seeing the shape of the data itself. The grid lines serve no useful purpose. Now consider the legend, which contains nine labels for the differently colored bars in the graph. To interpret which bar represents which product, you must shift your attention back and forth between the legend and the bars again and again. This is time-consuming and frustrating.

Now, take a look at Figure 7, which supports the message directly, clearly, and efficiently. The creator of this presentation has a clear grasp of the message and, through an understanding of visual perception, designed a display that communicates that message vividly.

FIGURE 7 - A better version of the information in Figure 6: This graph exhibits a clearer grasp of the message being communicated.

What You Can Do

Does the level of expertise in report design that I've illustrated require many years of training and experience? If it did, you might argue that the cost of the remedy would exceed the cost of the problem, but it doesn't. The practices that produce consistently effective data presentations aren't all intuitive, but they're all easy to learn.

As a professional concerned about BI, you're in a unique position to recognize the need for better data presentation and, most importantly, do something about it. You understand the importance of business information and of presenting it to decision makers in a way that tells them what they need to know, quickly and accurately. Your perspective makes you an ideal advocate for this cause. You care about the health and success of your business and want your contributions to effectively promote strategic goals. Don't let your company get caught up in the technology race and forget the fundamental skills required to make effective use of the information delivered by that technology. Enlightening data presentation is a skill that you can learn, but only if you first recognize the need and then give it the attention it deserves.

Stephen Few is the founder of Perceptual Edge, a consulting firm that specializes in information design for communication. His new book, Show Me the Numbers: Designing Tables and Graphs to Enlighten, will be available from Analytics Press in July 2004.

Editor's Note: With this article, we begin a practical series of articles by Stephen Few devoted to data presentation. We look forward to readers' feedback, including good and bad examples of data presentation that you've encountered. (We can change the names to protect the guilty.)

Editor's Choice
Brian T. Horowitz, Contributing Reporter
Samuel Greengard, Contributing Reporter
Nathan Eddy, Freelance Writer
Brandon Taylor, Digital Editorial Program Manager
Jessica Davis, Senior Editor
Cynthia Harvey, Freelance Journalist, InformationWeek
Sara Peters, Editor-in-Chief, InformationWeek / Network Computing