2 Questions Designers Should Ask Data Nerds
An engaging data visualization is more than making a graph look ‘pretty’
If you’re a designer who works with data nerds a lot, you might have found yourself getting requests like: “Just do your magic and make this look good.” But of course, what they really want is for you to take the data and create visuals that allow the user to fully understand it.
When the data nerds bring you something like this (actual Federal Communications Commission data from 2016, reported here):
You will, of course, make some smart choices like removing the 3D from the columns, changing those columns to an area graph, and abbreviating each month so the text isn’t diagonal.
Then you’ll apply your branding colors and fonts.
And add the logo, as requested by the department style guide.
But this needs to go even further to really engage the audience with the data. You’ll know where else to take this when you have more information to work with, so take advantage of initial project meetings and discussions to get what you need in order to generate high-impact visuals that tell the data story with dignity. Frame those early discussions with these two questions.
Question 1: What’s your point?
This is the first question your audience is going to ask, so it should be the first question you ask the data nerds. It is a pretty genius move to start with “What’s your point?” because it means that if the data nerds don’t have a point, you don’t make a graph. This will save you so much time.
Our audiences are primarily driven to engage through pictures. Our brains are all working on a “pictures first” basis, and this includes graphs. The graphs we make have to bring folks in or they’ll drop us and go check Instagram. A graphic designer I know says we have about three seconds to grab attention. This means that our visuals must be strong, and able to stand on their own.
But when we show audiences graphs, even pretty ones, that don’t have a clear point, we waste that power of the visual and teach them to generally ignore our entire organizational body of work because it’ll take a ton of effort to pull out meaning from what we deliver. This is not the path to success.
So when the data nerds ask you to make the graph pretty, reply with your own question: What’s your point?
Designers can be great partners in helping the data nerds craft a reader-appropriate, tweet-length insight to serve as the graph’s headline.
Some helpful hints: The point cannot be “to show the data.” In a three-second, distraction-driven environment, our audiences do not want to just be shown some data.
I was once working with a group of data nerds who told me just this: The point was to show the data to the public. I replied that the public doesn’t want to just see data, they want to know the meaning, the insight, the point. They huddled for a few moments and then clarified, “The point is to illustrate the data.” LOL. Sorry, folks: The point shouldn’t start with “to” and a verb.
Audiences want insights, so the point of the graph should be short and digestible enough to be the title or perhaps the subtitle of the visual. The point should be right up front, hitting the audience in the eyes.
Now, the data nerds have a burden of knowledge. So it may be difficult for them to provide a quick, easily digestible insight, since traditional academic backgrounds discourage this and data nerds spend a long time in academia. When they are able to provide an insightful point, it may be long and cumbersome and laden with technical jargon. Designers can be great partners in helping the data nerds craft a reader-appropriate, tweet-length insight to serve as the graph’s headline.
When you ask the data nerd about the point of the graph, she may reply with something like “Our average consumer complaint volume is 180 complaints per month, but in September the call volume to the FCC increased 10.55 times over the pre-existing monthly average.” So you translate that into a title that reflects the point in a more concise way:
Question 2: Who is the audience?
This question is packed with so many more: What will be the engagement scenario? Is the visual going into a slideshow where someone will be there to elaborate the story? Will it be used on a website where people might expect interactivity? (It isn’t even clear what software program to open unless you know the engagement scenario.)
What are their key questions and worries? What keeps this group up at night? If you can tap into your audience’s main concerns and frame your insights to speak to those concerns, you’ll be seen as trusted experts and leaders in your field. You’ll get gratitude and maybe even a high five.
Is the audience made up of nerds? If so, maybe it is okay to retain some of the more jargon-heavy insights. That type of audience might expect to see concepts like statistical significance make an appearance. If the audience is not so nerdy, this is the point in the discussion where the designers can gently remind the nerds that reporting that level of detail will scare off many folks.
A nerdy audience would expect to see certain traditional chart types, like bars and lines, and they would likely prefer that those charts include error bars and confidence intervals. If the audience is less nerdy, designers would make different choices, perhaps skewing a bit more toward an infographic.
If you can tap into your audience’s main concerns and frame your insights to speak to those concerns, you’ll be seen as trusted experts and leaders in your field.
We also want to determine how much detail the audience needs to do the thing we want them to do. Let’s give them that much detail and no more. If the audience doesn’t need to see the entire dataset in order to get motivated to make a donation or call an elected official, remove what isn’t necessary. If they are going to make decisions based on differences to the hundredth degree, include two decimal places. The level of detail designers retain from the data should be audience-dependent.
In addition, the way you word that headline framing the point of the visual could change depending on the audience you are making the point to.
So when you ask the design nerd about the audience she’ll present the graph to, she tells you it’s for the board of directors, in an attempt to assure them that the billing glitch that caused the spike in complaints was remediated and complaints are now on the decline. Given this audience and scenario, you will make some refinements to the graph.
You take out the logo (the board doesn’t need to see that), add in a calming blue (even though red is your primary branding color), refine the title to be more assuring, and add in a projection toward November.
As you can see, some of the work that took place at the start of this story — a designer’s basic work on a graph — was revised after discussion with the data nerd. Indeed, the entire design workflow, from choosing the right chart type to applying a story-telling color palette, stems from the information gathered when designers ask the data nerds these two questions. Asking these questions upfront, before engaging in the typical designer workflow, can make for a more efficient process.
Data nerds are likely not thinking about these questions when they bring you the data. Their minds are usually in the numbers, not in how those numbers are communicated. So data nerds need designers to press them for information about insight and audience just as much as designers need to press data nerds for this information to do their jobs effectively.