10 Tips for Better Data Storytelling


1) Accept the fact that communication is your number one job

Seriously. Nothing else matters as much as communications. Do it well and you will succeed in life, in your career, in your relationships… Be more clear, more easily understood, more creative and you will be head and shoulders above everyone you’re competing with.

2) Use the 4X4 model for knowledge content

The 4X4 Model is a very simple idea. You have to create content for people at varying levels to get the right people to the right content at the right time. You have to create what I call “water cooler moments” (headlines, tweets, images, 30-second videos) to entice people.

Those who are enticed want “café content” (3–5-minute videos, blog posts, etc.) And some people will even go to “the research library” and read a 100-page PDF or the “lab” and play with the data directly. The point is, no matter how deep your content is, you have to create the shorter form content to entice people. Even your most committed audience doesn’t want your long form content without being seduced by shorter form content first.

3) Maximize the number of senses you activate

Research has shown that when people hear or read certain words, the portions of their brains that are devoted to various sensory systems are activated. For instance, when you read the word “coffee”, your brain’s olfactory receptors fire, meaning your brain essentially thinks it is smelling coffee. This is very powerful in storytelling because the more senses you can stimulate, the more likely your audience will remember your story.

4) Make it human

Statistics is boring because I don’t care about numbers. And I only mildly care about categories of things, groups of people, collections. But tell me about one person with a name, a face and characteristics I can relate to and I’ll be much more likely to be empathetic to their plight. This is why so many great stories — even journalistic stories — are about specific people, not just “the numbers”. Make your stories accessible and memorable by making them about real people.

5) Talk to your audience

You can’t communicate nuclear physics research findings to an audience of turtles. (That’s a pretty big stretch, metaphor-wise, huh?) You will struggle to share that same information with a room full of two-year-olds. Conversely, communicating with a roomful of nuclear physicists about the relative pros and cons of strained peas and pureed bananas might also not be so thrilling for them. Your audience should drive what you say and how you say it. So get to know your audience — talk to them — and your targeted communication to them will be more successful.

6) Make it emotional

Emotions are at the heart of every great story. Whether it’s fear or joy, emotions are what grab your readers attention and burn your story deep into their brains (and hearts), making them memorable and impactful. Emotions aren’t inherent in numbers themselves, but when you can extrapolate from the data and think about the human aspects of it, you can almost always find an emotional impact that can be explored in your communications.

7) Storyboard everything

My teachers used to always say “make an outline before you start writing your paper,” and I would always say “pfffffttttt.” Who has time to outline things? You just want to dive in — you have great ideas that must be captured! When I sat down to write this very piece you’re reading, I wrote the intro and the conclusion and then wrote the 10 tips. Then I went back in and filled them in.

I am a firm believer in a strong outline because it helps you see the forest for the trees while developing your ideas. Sometimes I have such a strong detailed idea that I have to step out of outlining to capture it (before my terrible memory lets it slip away) but my default is to outline and make sure the structure works before getting to the individual words. Same goes for data storytelling — you have to think about the key points and flow of your ideas, and then how the visuals and functionality (if interactive) will work. This is all before you start doing visual design, chart making or anything else.

8) Constantly remember your audience’s pre-attentive predilections

If you’ve read any data visualization books, you’ve undoubtedly heard of pre-attentive processing — the idea that humans process visuals in a pre-attentive way. We are making sense of our world before we’re even aware of it — within 1/10 of a second. Visual processing is what visual communications is all about. Knowing this allows us to design data visualizations in a way that leverages what we understand about humans. We know that with the strategic use of color, contrast, shape and size, we can be sure that we are triggering a pre-attentive response to reveal patterns and trends. So think about what your audience will see at a glance when they look at your data.

9) Try different charts

When it gets to designing the data itself — picking how you’ll visualize it — try different things. Try traditional charts, try unique and novel visual ideas, try to look at your data from many different directions. There is no ‘perfect’ chart, so just find the right visual that best communicates what you’re trying to say with your data. This implies you know what you’re trying to say, which is obviously important!

10) Be inspired

You’re not alone. There are so many organizations and individuals doing great work in this field. Consume that content and you will be constantly inspired to do great work yourself. Keep your eye on the best of the best: the NY TimesWashington PostBloombergthe GuardianFiveThirtyEight and many more. Follow the hashtag #dataviz or check out the subreddit dataisbeautiful. You can also follow me on Twitter (@billshander) where I’m always sharing interesting work.

I hope these tips are helpful. They are truly the 10 most important things you need to improve your data communications. For more specific ideas about data storytelling,

Learn Data Viz n San Diego May 1, 2017

Learn Data Viz in Los Angeles May 4, 2017