The importance of Purpose in Data Visualization

by Matthew Armstrong

Last Friday was presentation day, and we were tasked with giving one of our first Vizzes a much-needed makeover using what we’d learned in our first week.
Most of the Vizzes chosen were some of our earlier work, meaning there were plenty of aspects to critique, mostly in terms of visual best practice. But I also noticed another common feature, namely, a lack of purpose in our original Vizzes.

Many of us came from an academic background, where we were given coursework to complete and dissertations to writing. For the most part, this involved being assigned (or choosing) a question, and going about answering that question. Similarly, if you’re working on a project assigned by a client, you’ll often find yourself being given a specification outlining the main questions and their target audience. In both of these scenarios, the question of purpose is mostly answered.
However, for those new to visualization, you’re most likely going to be working on personal projects, leaving the question of purpose in your own hands.

It can we tempting, when you’ve found a particularly good dataset, to try to visualize as much of the data as possible, and provide lots of insights. But, as we found the other day, in trying to maximize the scope and appeal of our Vizzes, we only served to complicate and confuse our narrative, and our audience. Instead of guiding our audience through a coherent story, our Vizzes overwhelmed and distracted from our central point.

So how do you give purpose to your Viz?

Try, at each stage of the process, to ask a few questions about your Viz:

  1. What am I doing/answering? – for most Vizzes, I try to answer/explore a central question and stick to it. That said, it’s perfectly fine for the answer to be “None”. You may want to create an entirely exploratory viz, letting the user ask the questions. Just make sure you’re clear on this.
  2. Who is this for? – knowing who you’re targeting will help you tailor a lot of the content for the viz (question, chart type, language used, etc.) and keep the tone consistent.
  3. Do I need this? – Every single element of the Viz needs to contribute to the viz as a whole, so it’s critical that you’re ruthless in asking this question and removing the elements you deem unnecessary.

Answering these will help simplify and clarify your narrative, and help your audience get the most out of your visualization.

Now, all this is much easier said than done, and I know personally how difficult it is to put yourself in the shoes of your prospective audience. So, my parting piece of advice is this: ask someone else for feedback (preferably someone outside of the creative process). Ask them what they think the central narrative is, and listen to them when they ask you questions about it. Theoretically, the perfect visualization is one where the purpose and narrative are clear, and the audience doesn’t need to ask any questions. So, when they ask you a question, don’t just answer it, think about why they had to ask it. Iterate, refine, and ask again.

I hope this helps you add a bit more purpose to your Vizzes, so we can all get the most out of them!  

Avatar

Matthew Armstrong