As data has permeated our offices, our homes, and just about everywhere else, it continues to shape expectations about how our lives work. Nearly every aspect of what we do with devices is mediated through a data-driven display. As a result, we naturally expect that our lives will run faster and more efficiently. But will they? And when will we hit a point of diminishing returns? 

As data managers, we grapple with this challenge firsthand. 

Should we convey more…or less? 

Recently we had a hackathon at my company. One of the themes was how to use good data analytics to stay on top of all the news being written about the telecommunications industry. One of our coders took an approach along these lines: “There is all this great information packed in these articles, but I don’t have time to read them, so maybe I can write an algorithm to sum each one up in two sentences and then tell me the sentiment of the article.”

This person’s attitude speaks to our conundrum well. On the one hand, we want to have interesting, engaging information at our fingertips. On the other, we have almost no time to read or engage with any of the content. At some point I feel that we’ll need a way to ingest that data faster, but I worry that we’ll lose some of the context and perspective that gives data its richness and ability to drive change.

Enter better data storytelling

I believe data storytelling is going to be our most critical capability for success going forward. Being able to decipher information in only one or two ways is insufficient for the modern data organization, so building data literacy should be at the top of our agendas. 

Here’s how I define the three principles of good data storytelling:  

  1. Good data storytelling has to center on a story. This may seem obvious, but if your team’s visualizations aren’t connecting on an emotional, personal or practical level, they will be far less likely to engage the “readers” or change their minds. 
  2. Visualizations must lead you past a point of admiring their beauty. If your staff can’t describe the insight they’re highlighting, they need to keep pushing for a simple, powerful story to come to the surface. 
  3. We have to tell stories that non-technical users will understand. Good storytelling is a function of good visualization, regardless of your technical background. If a user can’t pull a story out of a visualization in less than ten seconds, it’s probably not serving your purposes.

There’s no question that data has more societal impact every day. As data leaders, we need to scale the belief in and ability to distill what our users need to take away from the rising tide of information so they make better decisions. 

We live inside data now for many hours of the day, so it’s up to us to tell stories that reflect the facts, that create memorable hooks in users’ minds, and that illustrate the implications of what we display. Every visualization may not have characters and conflict, but each one should take place in a setting, share a plot of what is happening, and share a theme that clarifies what might have been unclear before.