Within a given category, we can compare the blue bar to the grey bars to see how our business is faring relative to competitors: winning compared to the competition on Price, losing on Service and Selection.
Competitors are distinguished from each other based on the order in which they appear (Competitor A always appears directly after the blue bar, Competitor B after that, and so on), which is outlined in the legend at the left. If it were important to be able to quickly identify each competitor, this design doesn’t immediately allow for that. But if that is a second- or third-order comparison in terms of priority and isn’t the most critical thing, this approach can work well. In the makeover, I’ve also organized the categories in order of decreasing weighted performance index for “Our business,” which provides a construct for our audience to use as they take in the information, and added a summary metric (relative rank) so it’s easy to know quickly how “Our business” ranks in each category in relation to our competition.
Note here how the effective use of contrast (and some other thoughtful design choices) makes it a much faster, easier, and just more comfortable-feeling process to get the information we’re after than it was in the original graph.
When redundant details shouldn’t be considered clutter
I’ve seen cases where the title of the visual indicates the values are dollars but the dollar signs aren’t included with the actual numbers in the table or graph. For example, a graph titled “Monthly Sales ($USD Millions)” with y-axis labels of 10, 20, 30, 40, 50. I find this confusing. Including the “$” sign with each number eases the interpretation of the figures. Your audience doesn’t have to remember they are looking at dollars because they are labeled explicitly. There are some elements that should always be retained with numbers, including dollar signs, percent signs, and commas in large numbers.
Decluttering: step-by-step
Now that we have discussed what clutter is, why it is important to eliminate it from our visual communications, and how to recognize it, let’s look at a real-world example and examine how the process of identifying and removing clutter improves our visual and the clarity of the story that we’re ultimately trying to tell.
Scenario: Imagine that you manage an information technology (IT) team. Your team receives tickets, or technical issues, from employees. In the past year, you’ve had a couple of people leave and decided at the time not to replace them. You have heard a rumbling of complaints from the remaining employees about having to “pick up the slack.” You’ve just been asked about your hiring needs for the coming year and are wondering if you should hire a couple more people. First, you want to understand what impact the departure of individuals over the past year has had on your team’s overall productivity. You plot the monthly trend of incoming tickets and those processed over the past calendar year. You see that there is some evidence your team’s productivity is suffering from being short-staffed and now want to turn the quick-and-dirty visual you created into the basis for your hiring request.
Figure 3.17 shows your original graph.
Figure 3.17 Original graph
Take another look at this visual with an eye toward clutter. Consider the lessons we’ve covered on Gestalt principles, alignment, white space, and contrast. What things can we get rid of or change? How many issues can you identify?
I identified six major changes to reduce clutter. Let’s discuss each.
1. Remove chart border
Chart borders are usually unnecessary, as we covered in our discussion of the Gestalt principle of closure. Instead, think about using white space to differentiate the visual from other elements on the page as needed.
2. Remove gridlines
If you think it will be helpful for your audience to trace their finger from the data to the axis, or you feel that your data will be more effectively processed, you can leave the gridlines. But make them thin and use a light color like grey. Do not let them compete visually with your data. When you can, get rid of them altogether: this allows for greater contrast, and your data will stand out more.
3. Remove data markers
Remember, every single element adds cognitive load on the part of your audience. Here, we’re adding cognitive load to process data that is already depicted visually with the lines. This isn’t to say that you should never use data markers, but rather use them on purpose and with a purpose, rather than because their inclusion is your graphing application’s default.
4. Clean up axis labels
One of my biggest pet peeves is trailing zeros on y-axis labels: they carry no informative value, and yet make the numbers look more complicated than they are! Get rid of them, reducing their unnecessary burden on the audience’s cognitive load. We can also abbreviate the months of the year so that they will fit horizontally on the x-axis, eliminating the diagonal text.
5. Label data directly
Now that we have eliminated much of the extraneous cognitive load, the work of going back and forth between the legend and the data is even more evident. Remember, we want to try to identify anything that will feel like effort to our audience and take that work upon ourselves as the designers of the information. In this case, we can leverage the Gestalt principle of proximity and put the data labels right next to the data they describe.
6. Leverage consistent color
While we leveraged the Gestalt principle of proximity in the prior step, let’s also think about leveraging the Gestalt principle of similarity and make the data labels the same color as the data they describe. This is another visual cue to our audience that says, “these two pieces of information are related.”
This visual is not yet complete. But identifying and eliminating the clutter has brought us a long way in terms of reducing cognitive load and improving accessibility. Take a look at the before-and-after shown in Figure 3.24.
Figure 3.18 Remove chart border
Figure 3.19 Remove gridlines
Figure 3.20 Remove data markers
Figure 3.21 Clean up axis labels
Figure 3.22 Label data directly
Figure 3.23 Leverage consistent color
Figure 3.24 Before-and-after
In closing
Any time you put information in front of your audience, you are creating cognitive load and asking them to use their brain power to process that information. Visual clutter creates excessive cognitive load that can hinder the transmission of our message. The Gestalt Principles of Visual Perception can help you understand how your audience sees and allow you to identify and remove unnecessary visual elements. Leverage alignment of elements and maintain white space to help make the interpretation of your visuals a more comfortable experience for your audience. Use contrast strategically. Clutter is your enemy: ban it from your visuals!
You now know how to identify and eliminate clutter.
chapter 4
focus your audience’s attention
In the previous chapter, we learned about clutter and the importance of identifying and removing it from our visuals. While we work to eliminate distractions, we also want to look at what remains and consider how we want our audience to interact with our visual communications.
In this chapter, we further examine how people see and how you can use that to your advantage when crafting visuals. We will talk briefly about sight and memory in order to highlight the importance of some specific, powerful tools: preattentive attributes. We will explore how preattentive attributes like size, color, and position on page can be used strategically in two ways. First, preattentive attributes can be leveraged to help direct your audience’s attention to where you want them to focus it. Second, they can be used to create a visual hierarchy of elements to lead your audience through the information you want to communicate in the way you want them to process it.
By understanding how our audience sees and processes information, we put ourselves in a better position to be able to communicate effectively.
You see with your brain
Let
’s look at a simplified picture of how people see, depicted in Figure 4.1. The process goes something like this: light reflects off of a stimulus. This gets captured by our eyes. We don’t fully see with our eyes; there is some processing that happens there, but mostly it is what happens in our brain that we think of as visual perception.
Figure 4.1 A simplified picture of how you see
A brief lesson on memory
Within the brain, there are three types of memory that are important to understand as we design visual communications: iconic memory, short-term memory, and long-term memory. Each plays an important and distinct role. What follows are basic explanations of highly complex processes, covered simply to set the stage for what you need to know when designing visual communications.
Iconic memory
Iconic memory is super fast. It happens without you consciously realizing it and is piqued when we look at the world around us. Why? Long ago in the evolutionary chain, predators helped our brains develop in ways that allowed for great efficiency of sight and speed of response. In particular, the ability to quickly pick up differences in our environment—for example, the motion of a predator in the distance—became ingrained in our visual process. These were survival mechanisms then; they can be leveraged for effective visual communication today.
Information stays in your iconic memory for a fraction of a second before it gets forwarded on to your short-term memory. The important thing about iconic memory is that it is tuned to a set of preattentive attributes. Preattentive attributes are critical tools in your visual design tool belt, so we’ll come back to those in a moment. In the meantime, let’s continue our discussion on memory.
Short-term memory
Short-term memory has limitations. Specifically, people can keep about four chunks of visual information in their short-term memory at a given time. This means that if we create a graph with ten different data series that are ten different colors with ten different shapes of data markers and a legend off to the side, we’re making our audience work very hard going back and forth between the legend and the data to decipher what they are looking at. As we’ve discussed previously, to the extent possible, we want to limit this sort of cognitive burden on our audience. We don’t want to make our audience work to get at the information, because in doing so, we run the risk of losing their attention. With that, we lose our ability to communicate.
In this specific situation, one solution is to label the various data series directly (reducing that work of going back and forth between the legend and the data by leveraging the Gestalt principle of proximity that we covered in Chapter 3). More generally, we want to form larger, coherent chunks of information so that we can fit them into the finite space in our audience’s working memory.
Long-term memory
When something leaves short-term memory, it either goes into oblivion and is likely lost forever, or is passed into long-term memory. Long-term memory is built up over a lifetime and is vitally important for pattern recognition and general cognitive processing. Long-term memory is the aggregate of visual and verbal memory, which act differently. Verbal memory is accessed by a neural net, where the path becomes important for being able to recognize or recall. Visual memory, on the other hand, functions with specialized structures.
There are aspects of long-term memory that we want to make use of when it comes to having our message stick with our audience. Of particular importance to our conversation is that images can help us more quickly recall things stored in our long-term verbal memory. For example, if you see a picture of the Eiffel Tower, a flood of concepts you know about, feelings you have toward, or experiences you’ve had in Paris may be triggered. By combining the visual and verbal, we set ourselves up for success when it comes to triggering the formation of long-term memories in our audience. We’ll discuss some specific tactics for this in Chapter 7 in the context of storytelling.
Preattentive attributes signal where to look
In the previous section, I introduced iconic memory and mentioned that it is tuned to preattentive attributes. The best way to prove the power of preattentive attributes is to demonstrate it. Figure 4.2 shows a block of numbers. Taking note of how you process the information and how long it takes, quickly count the number of 3s that appear in the sequence.
Figure 4.2 Count the 3s example
The correct answer is six. In Figure 4.2, there were no visual cues to help you reach this conclusion. This makes for a challenging exercise, during which you have to hunt through four lines of text, looking for the number 3 (a kind of complicated shape).
Check out what happens when we make a single change to the block of numbers. Turn the page and repeat the exercise of counting the 3s using Figure 4.3.
Figure 4.3 Count the 3s example with preattentive attributes
Note how much easier and faster the same exercise is using Figure 4.3. You don’t have time to blink, don’t really have time to think, and suddenly there are six 3s in front of you. This is so apparent so quickly because in this second iteration, your iconic memory is being leveraged. The preattentive attribute of intensity of color, in this case, makes the 3s the one thing that stands out as distinct from the rest. Our brain is quick to pick up on this without our having to dedicate any conscious thought to it.
This is remarkable. And profoundly powerful. It means that, if we use preattentive attributes strategically, they can help us enable our audience to see what we want them to see before they even know they’re seeing it!
Note the multiple preattentive attributes I’ve used in the preceding text to underscore its importance!
Figure 4.4 shows the various preattentive attributes.
Figure 4.4 Preattentive attributes
Source: Adapted from Stephen Few’s Show Me the Numbers, 2004.
Note as you scan across the attributes in Figure 4.4, your eye is drawn to the one element within each group that is different from the rest: you don’t have to look for it. That’s because our brains are hardwired to quickly pick up differences we see in our environment.
One thing to be aware of is that people tend to associate quantitative values with some (but not all) of the preattentive attributes. For example, most people will consider a long line to represent a greater value than a short line. That is one of the reasons bar charts are straightforward for us to read. But we don’t think of color in the same way. If I ask you which is greater—red or blue?—this isn’t a meaningful question. This is important because it tells us which of the attributes can be used to encode quantitative information (line length, spatial position, or to a more limited extent, line width, size, and intensity can be used to reflect relative value), and which should be used as categorical differentiators.
When used sparingly, preattentive attributes can be extremely useful for doing two things: (1) drawing your audience’s attention quickly to where you want them to look, and (2) creating a visual hierarchy of information. Let’s look at examples of each of these, first with text and then in the context of data visualization.
Preattentive attributes in text
Without any visual cues, when we’re confronted with a block of text, our only option is to read it. But preattentive attributes employed sparingly can quickly change this. Figure 4.5 shows how you can utilize some of the preattentive attributes introduced previously with text. The first block of text doesn’t employ any preattentive attributes. This renders it similar to the count the 3s example: you have to read it, put on the lens of what’s important or interesting, then possibly read it again to put the interesting parts back into the context of the rest.
Figure 4.5 Preattentive attributes in text
Observe how leveraging preattentive attributes changes the way you process the information. The subsequent blocks of text employ a single preattentive attribute each. Note how, within each, the preattentive attribute grabs your attention, and how some attributes draw your eyes with greater or weaker force than others (for example, color and size are attentio
n grabbing, whereas italics achieve a milder emphasis).
Beyond drawing our audience’s attention to where we want them to focus it, we can employ preattentive attributes to create visual hierarchy in our communications. As we saw in Figure 4.5, the various attributes draw our attention with differing strength. In addition, there are variances within a given preattentive attribute that will draw attention with more or less strength. For example, with the preattentive attribute of color, a bright blue will typically draw attention more than a muted blue. Both will draw more attention than a light grey. We can leverage this variance and use multiple preattentive attributes together to make our visuals scannable, by emphasizing some components and de-emphasizing others.
Figure 4.6 illustrates how this can be done with the block of text from the previous example.
Storytelling with Data Page 8