A BI Connector interview on data storytelling with Dave Wells of the Eckerson Group.

Why do you think that the concept of data storytelling is important, particularly now with so many available data visualization tools?

Data storytelling is absolutely critical because many people create visuals that are a bit ambiguous and subject to interpretation.

It is so easy to create visuals themselves but what matters is how they’re ultimately interpreted by others. I’ve in fact done exercises where three different people are shown a visual and they deliver different interpretations of what the visual says.

My take is that you, as a creator of these visuals, want people to understand the message that you’re trying to communicate. But it’s easy to forget that these visuals may be subject to a variety of misinterpretations. Ultimately, a clear, defined narrative helps create a more durable understanding of the story that your visual is trying to tell.

What high-level advice would you give to BI analysts who are looking to add a storytelling component to their reports?

There are several things that are really important. The first thing is that you must know your audience in order to capture your audience. You can’t capture their attention if you don’t know to whom you’re speaking. Equally important, remember that you have to be true to the data. If you skew the data people will feel manipulated and reject the entire message.

Hans Rosling, a master of data storytelling, said that his job is to communicate with data in a way that is both understandable and enjoyable. Remember that good storytelling is not just about telling but about listening. Be sure to read the audience and invite their responses. With interaction you’ll do a better job of communicating than with a monologue.

There’s a tendency in the industry to create visualizations first and worry about the narrative later. My experience is that it works equally well, and perhaps better, to build the narrative first and then ask yourself: “how can I use data visualization to best support this story?”

There is a lot of hype about making data-driven decisions. Do you feel that information should be data-driven or story-driven or both?

I’m not a fan of data-driven decisions. I’m a fan of data informed decisions. I think data should help to guide decisions, but it’s necessary to respect the judgement and experience of the decision maker as well as to consider what the data has to say.

What data visualization tools do you feel are more suitable to performing data storytelling?

The right tools depend on the storyteller and the audience. Data storytelling will have a visual and a narrative component. Visual tools that are appropriate to the story, such as Tableau, Domo and Qlik, must be well suited for the narrative.

The least effective method is to support static visualizations with written text. If you need to do it with written text and static images, then your visualization is likely to be less effective than if you add elements of visual animation and spoken narrative that is either live or recorded.

Learn more about BI Connector and read more posts on data visualization in our blog.