Here are some thoughts on making dashboards more engaging to the users. The answer isn't to resort to making things “flashy” or “beautiful” but to create a personal connection between the data and the user.

Overview

A few years ago, a client was updating a collection of survey data dashboards and wanted to revisit the way demographic data was presented. They thought that the collection of bar charts comprising the demographics dashboard was boring and wanted to replace them with something that was more visually arresting. In particular, they wanted to take something that looked like what is in Figure 30.1 and replace it with something flashy.

Something with pizzazz! Something with treemaps and bubbles and pie charts! Something like what is in Figure 30.2.

Yikes. This is certainly a colorful montage, but it takes hard work to make sense of it. Instead of taking seconds to grasp, it will take several minutes to understand the demographics of the survey takers.

When asked why they wanted something “flashier,” they indicated a desire to draw the viewer into the dashboard, and they thought a dashboard with more than just bar charts would do the trick.

Why do they want to draw people into this dashboard? When pressed further, the client argued that because the data was boring, the dashboard needed to be cooler; otherwise people wouldn't engage with it.

But if the data is boring, why bother to visualize it? And why would you want users to spend time on this boring data when there was more important data to explore and understand?

There was, in fact, a very good reason to show the demographics of people who took the survey: to let interested parties see for themselves if there was enough overlap between the survey participants and the interested party. That is, the key to getting an individual involved with this dashboard and all the related dashboards was to show that individual how the data pertained to him or her.

So, how can we do that?

Personalized Dashboards

At the 2015 Tapestry conference on data storytelling, Chad Skelton, then of the Vancouver Sun, presented a great session making the case that people are ravenous for data about themselves.

Chad created an interactive dashboard that allows Canadians to see how much older or younger they are than other Canadians. Below is a similar dashboard using United States census data.

Figure 30.3 presents a histogram showing the distribution between age and U.S. population.

Not exactly thrilling stuff.

Now let's contrast the general-purpose graphic in Figure 30.3 with the personalized dashboard shown in Figure 30.4.

Every person seen using this dashboard immediately moves the slider left and right and applies the filters—first to compare his or her own age and gender, then to compare the age and gender of a spouse, friend, or child. Users find the allure of the dashboard changing based on an individual's input irresistible.

Indeed, data is much more interesting when it is about you. So, how can we apply this concept to our “boring” demographics dashboard?

Make the Demographics Dashboard Personal

With the goal of personalization in mind, let's see how we can make the dashboard in Figure 30.1 more interesting.

Let's start by gathering some information about the person viewing the dashboard; that is, let's present some parameters from which the viewer can apply personalized settings. (See Figure 30.5.)

Now we can take these parameter settings and highlight them in the dashboard (and all of the other dashboards, for that matter, see Figure 30.6).

We can then go one step further and invite the viewer to select the colored bars to see exactly how many people who took the survey have the same demographic background as him or her. That is, have viewers click to select their gender, their generation, and their location. (See Figure 30.7.)

Twenty-seven people fall into the identical demographic pool as the person viewing the dashboard.

But What If You Still Want to Make Something Beautiful?

There may be times when you feel the inexorable pull to make your dashboards more beautiful. You see stunning visualizations like the ones in Figures 30.8 through 30.10 and wonder, “Why can't I make something like that?”

Source: Used with permission of Pooja Gandhi

Source: Used with permission of Adam E. McCann

Source: Used with permission of Adam Crahen

Realize that these data visualizations were built for public consumption, not internal use. They compete with other visuals for readers' attention. They virtually scream “Look at me! Look at me!” Indeed, the treemap, pie chart, and bubble chart demographics dashboard in Figure 30.2 was from a client who was building a public-facing dashboard.

Personally, I think the bar charts in Figure 30.7 are beautiful, but I get the point. You see these come-hither dashboards and wonder if maybe you can borrow some design elements from them.

If you're the only person looking at the dashboard, I suppose you can do anything you want. Sure, go ahead and replace those bar charts with donut charts if you like. As long as doing so doesn't impede your understanding of the data, then what's the harm?

But if others need to understand the data, you should proceed with caution. Your goal is to make something that is accurate, informative, and enlightening. You need to make sure whatever you add does not compromise that goal.

And how do you do that? Ours is not a book about graphic design and we do not attempt to tackle issues related to sophisticated use of typography, layout, and shapes here. But I want to show you a chart type that combines the analytic integrity of a bar chart with the “ooh, circles” of a bubble chart.

An Alternative to the Bar Chart: The Lollipop

A lollipop chart is simply a dot plot chart superimposed on top of a bar chart that has very thin bars. Figure 30.11 shows an example using sales data.

Figure 30.12 presents the demographic dashboard from earlier, rendered using a lollipop chart.

I personally prefer the bar chart but would not protest if a client wanted to use the lollipop version instead.

Conclusion

If people aren't using your dashboards, it's because the information isn't meaningful to them, not because the dashboards aren't cool. Adding packed bubbles and pictograms in the hope of getting people engaged may attract attention at first but likely will hinder people's ability to understand the data. Then they will abandon the dashboards.

Although a lollipop chart might add some visual variety without sacrificing analytical clarity, if you really want to engage people, make the data meaningful. One of the best ways to do that is to make it personal.

For the record, I think personalized bar charts beat packed bubbles any day of the week.