Making Data Design Work

Jim McManus, January 10, 2017

The best design in the world won’t rescue you from bad data. And certainly won’t tell a story.

We’ve recently completed a number of presentation projects (including this recent one for JetBlue) that got me thinking about some common best practices related to the use and design of data in presentations. While we create stories in many formats — videos, presentations, online and interactive pieces, reports, collateral, and more — most involve some kind of data design. So here are some thoughts, inspired by recent conversations and client needs, about data, design, and what works for your audience.

I offer these in no particular order:

Start with good data
At one level this doesn’t sound hard to do, but you’d be surprised at number of times clients have started a conversation with us by asking us to make what they thought was boring data interesting by making it visually dynamic on some way (this request sometimes comes with a hand wave). Now, we’re all for making data visually dynamic, but if the data is boring to our client (and likely to their audience), no amount of design energy can transform it into an interesting graphic. In those cases we always suggest going back to the source to find more compelling, surprising, or interesting data. It’s like writing the lede of a good news story: start by grabbing audience attention.

Trust your audience
This can be tricky. It’s always best to assume that your audience is going to be more sophisticated and engaged than you might think. When I worked at the New York Times, we were often surprised by readers who would spend significant amounts of time pouring over our information graphics, and then giving us their feedback. They either appreciated the detailed work we presented or told us if we’d missed something they cared about. Yes, you want to spell things out, avoid jargon, and make graphics that are easy to enter; but you also want to respect the intelligence and sophistication of your audience by providing them with rich data. Doing so both reflects well on you and demonstrates your respect for them.

Turn your data into information
In the context of a presentation, you don’t want to stop at the simple presentation of good, rich data. Since you are trying to inform and/or persuade, you need to inject explicit insight into the data, and carefully guide the audience to the conclusion you feel is so important/compelling. You might add text that highlights conclusions to draw from the data, or introduce new design elements that visually emphasize such conclusions.

Make it visually interesting (but clear)
Because presentations are inherently visual (especially when you think of a big-tent keynote address or some other large audience event) you want to make your data, particularly more complex data, visually dynamic and compelling. It could be that the shape of the data (and the relationships within) are inherently interesting. It could be that their are interesting forms for conveying the data. Or it could be that the data should be conveyed via metaphor. Nigel Holmes made this technique popular many years ago. More recently, many people have made bad versions of these kinds of graphics, where they dumbed down the information and diminished their audience in hopes of gaining wide appeal. But when done creatively, thoughtfully, and with an appropriate level of sophistication, metaphor-based data graphics can be quite effective.

Tell a story
That’s a final point I want to close on… When possible, your data will make the most lasting impact when you make it tell a story. There may not be more than a simple story to tell when you’re presenting a simple bar chart or pie chart. But by combining simple charts in a sequence, or by starting with a more complex data set and layering on that data across multiple slides or with an interesting animation sequence, you can lead your audience through one or more data sets — setting context, building a case, and concluding with a strong end that makes an impact.

In a future post I’ll dive a little deeper into the range of visual approaches that can be appropriate for data presentation —from simple and straightforward, to image-based, to the most elaborate forms of computer aided data visualization.