Brent Dykes’ “Effective Data Storytelling” and Why Being a Data Visualization Expert is Pointless


Doc Brown from BBTF reading Brent Dykes' book!

“My urgency to write this book increased when I realized how poorly understood the concept of data storytelling was and how the term was in danger of becoming just another empty buzzword. Despite its immense potential, it was frequently positioned as just an extension of data visualization.”

Brent Dykes, Effective Data Storytelling: How to drive change with data, narrative, and visuals

With our Custom Charts for Jira and Confluence apps, we have often positioned ourselves as data visualization experts. In that we can help you visualize your data in Jira and Confluence more effectively using customized charts and reports.

That wasn’t the right thing to do. Because, as my colleague Jacek said in a LinkedIn post a little while back, that’s like helping you find the right ingredients for a cake but not baking it. Data visualization’s only half the job. Really, our focus should have been on data storytelling. A cake is only edible once it’s baked. And data is only valuable if you tell a story with it.

As Brent Dykes says in his book, the fact that data storytelling gets positioned as an extension of data visualization comes from a fundamental misunderstanding of both, as if data storytelling is a type of data visualization. Well, going back to our cake metaphor (I promise I’ll stop soon before I make you dash for the kitchen), that’s like saying an egg is a type of cake. Rather, an egg is an essential and necessary component of a cake, just like data visualization is an essential and necessary component of data storytelling.

Informing versus communicating

Dykes illustrates the difference by drawing a distinction between informing and communicating, quoting journalist Sydney J. Harris: “Information is giving out; communication is getting through”. He says that storytelling is crucial to “getting through”, citing, as an example, two ways of conveying the details of your recent vacation to someone. If you’re informing, you’re reeling off a list of facts: where you went, who you went with, how long for, and what you did. But if you’re communicating, you’re explaining what was interesting about the experience, what you enjoyed, what you didn’t, why you chose to do it, how it made you feel, etc. As Dykes says, informing connects with the head, communicating touches the heart.

Dykes’ distinctions aren’t that different to the old writing chestnut I hear over and over again as an author of fiction: show, don’t tell. The following quote sums up nicely what show, don’t tell, means:

“Don’t tell me the moon is shining; show me the glint of light on broken glass.”

Anton Chekov

Showing is akin to communicating, whereas telling is akin to informing. Showing, like communicating, enables the listener or reader to become fully immersed in the details. Telling doesn’t. It keeps the reader or listener at a distance.

In effect, data visualization is telling. It’s better telling than just giving your audience the raw data in a grid, but it’s still telling. It’s pulling up a bar chart and expecting your audience to interpret and understand the chart for themselves. Sure, your bar chart might be the clearest, most beautiful bar chart ever built. But if there is something in that bar chart that you really want to “get through” to your audience, you need to craft it into a story.

The anatomy of a data story

Dykes goes on to explain the six essential elements of a data story: data foundation, main point, explanatory focus, linear sequence, dramatic elements, and visual anchors.

  • Data foundation just means that the insight you’re sharing as part of your data story has come from facts you’ve gathered.
  • Main point is focusing on an overarching insight you’ve gleaned from those facts and not confusing the audience with too many extraneous details.
  • Explanatory focus means explaining the insight so that it is understandable to the listener, as opposed to just describing it. In other words, the how and the why rather than just the who, what, when, and where.
  • Linear sequence is about introducing data points in a sequential fashion and letting them build support for your overarching insight.
  • Dramatic elements means providing sufficient context for your insight, i.e. what happened before and why it matters to what’s happening now. It’s the setup for your data story.
  • Visual anchors is as it sounds: accompanying your narrative with visual depictions of your data in the form of charts and diagrams.

To be honest, it’s easy to see how this last element so frequently overshadows the rest. We humans are visual creatures and the oft-repeated adage, pictures paint a thousand words, can lead us to think that words aren’t necessary. However, in the context of a data story, leaving out the words can lead us to leave out the story, too. It’s why Dykes calls them visual “anchors”. The beautiful charts and diagrams you make are meant to illustrate your story. They’re not meant to tell it for you.

Beginning, middle, and end

Dykes’ essential elements of a data story aren’t that different to how I might craft a novel. Dramatic elements, linear sequence, and main point are, broadly speaking, the same as requiring a beginning, middle, and end in a novel. The beginning is setup, where the reader is provided with context to be able to understand the characters and their motivations, the setting, and what’s coming plot-wise. The middle contains a sequence of events, each one building on the previous, to lead the characters to the end, the climax, when the plot comes to a head and the main conflict is resolved.

With a data story, your beginning might be: these were our sales in the previous quarter. Your middle might be: these are our sales for the current quarter. And your end might be: here is a major dip in our sales in the current quarter.

Then, explanatory focus is back to show, don’t tell. With a novel, I have to make sure I’m immersing the reader in my plot and characters so that they can imagine they’re there in the story and feeling what my characters feel. It’s much the same with a data story. When you get to your main point, such as a dip in sales in the current quarter, that’s when you explain that certain internal and/or external pressures have contributed to the dip. This helps your audience understand and appreciate why the dip is happening. If they do that, they’re more likely to retain the message and, most importantly, do something about it.

Brent Dykes’ Effective Data Storytelling: How to drive change with data, narrative, and visuals is available from Amazon as a hardcover and as an ebook and comes highly recommended!

Christopher is a self-confessed nerd who’d probably take the cake on Mastermind if Star Trek: Voyager was his specialist subject. He writes fiction about time travel, conspiracies and aliens; loves roller coasters, hiking and Christmas; and hates carpet, rom-coms and anything with chilli in it. He’s written extensively for technology companies and Atlassian partners and specializes in translating complicated technical concepts, specs and jargon into readable, benefits-driven copy that casual readers will understand.