It isn’t a new thing. We’ve been storytellers since the dawn of man (literally and figuratively). Storytelling is a part of our DNA — and no one embraces that more than David Ciommo, Data Visualization Principal at Humana (+ our first recipient of the Best Data Storyteller Award).
A trained illustrator, David began his career in graphic design at Discovery Communications where he played a key role in the branding and creative development for the Discovery Kids Channel and Animal Planet Channel. Many years later, his career pivoted to data visualization & storytelling despite having no previous data & analytics background.
At Humana, David founded the Visualization Center of Excellence, an internal site created to share style guides, tip sheets, and other resources around data storytelling and data literacy. His overarching goal is to combine purposeful human-centered design principles with first class data visualizations in an effort to provide meaningful insights and tangible actionable opportunities to the leadership at Humana.
Drawing on his diverse experience, David has crafted his own approach to telling meaningful data stories and teaching others to do the same. Here are 5 tips from him that you can take back to your own team:
1. Start With The Story – Then Find The Data.
There are a lot of bad dashboards out there.
And they’re typically created using one of these two approaches:
1. You have data. You bring it into a BI platform. You throw some charts and graphs on it. You make changes throughout and then finally call it a data story or a dashboard.
2. You have a tool and you’re really only specialized in one tool or platform. You go get the data but struggle trying to get the data into the platform because it’s limiting for one reason or another. Or, you don’t really know what you’re doing.
And after all of that work, you never actually come up with a story. It’s just a statistical report. It’s a glorified Excel spreadsheet. We’re not slowing down enough and thinking about what we are actually doing or why we’re doing it.
That’s why David has always coached folks to start with the story.
“Forget about the data for a moment. Forget about the tool that you’re using. Just start with the story. What are you trying to say? It’s no different than writing a screenplay or a book. Who is my audience? You ask the important who, what, why, when, where, and how questions around the information. Then, you go look at your data to see if it supports that story. If you don’t have enough data to tell the story — go get that data.”
2. Do The Upfront Work
Creating BI product is no different than creating in marketing and advertising. You’re not going to just pump out design products. You’re not going to spend a lot of money on printing and publishing and ship out ads if they’re not going to work.
We need to slow down and think about the purpose of our data stories, dashboards, etc. A lot of people think this will slow down the project. It’s a common mindset and it stems from traditional business thinking — speed to value.
However, it’s not going to slow down the project. If anything, that extra time you spend in the beginning will help speed up the development process.
“I tend to do all of my upfront work and then just let the development team, or the programmers do what they’re really great at and just build at speed. We’ve delivered our product at least half of the time that our traditional product would be delivered. And it is a lot more efficient and the business partner appreciates it because they can weigh in earlier on what they want or what they think they want”.
3. Become a Better Interrogator
As data storytellers, we need to become better interrogators.
If you saw a dashboard for the very first time, what would you ask?
Chances are, you’re probably going to ask very logical and simple questions: Have the numbers changed over time? Have they gone up or down? Is that good or bad?
Get ahead of the questions. Help your audience understand the “so what,” not just the numbers. Ask yourself: Why should someone care about my findings? How does this information impact them?
Most people actually don’t want to look at data. They just want to know what it means for them and what they should do next.
“I always tell people, I’m like, you know, to some degree for me, it’s like throwing a dart to a dartboard blindly. You can sit there and tell me all day long about your data. I’m going to gloss over. It’s going to go right over my head. I want to understand what you want to do with it.”
4. Understand the Difference Between Wants & Needs
The truth is, most of the time people don’t know what they need.
They think they do — but they don’t.
“My dad taught me at a real young age the difference between needs and wants. And a lot of times we illustrate and tell stories around what leaders want to see, not what they necessarily need to see.”
It’s likely you’ll come across leaders who just want their Excel spreadsheet even if it’s not the right visual vehicle to communicate a story.
But as data storytellers, it’s our job to give people what they need, not necessarily what they want.
So, what’s the best approach? How do we uncover those needs?
If we look back in history, Edward Tufte, a pioneer in the field of data visualization, would constantly push people to ask the question, “Why?” Why are we doing this? Why is it important?
Challenge your leadership by asking them these questions. This will encourage them to open up their minds, to think about the intent of the story, and the value that it should bring.
5. Don’t Be Afraid to Prescribe a Story
Subjectivity is a natural human instinct. And it’s ever present in data storytelling.
People will interpret something differently than you, if you allow them to.
We see this a lot when people that have a heavy analytical or data science background (think developers, programmers, BI practitioners, etc) are creating products for non-analytical people.
They put together these products and it makes sense to them. However, they don’t always account for how somebody is going to interact with it or whether there’s clarity in the story.
David believes that this is where a lot of these visualization redundancies stem from — like having a pie chart directly below a table of the exact same data.
“It’s almost like people are afraid to some degree to prescribe a story and they would rather allow the individual to interpret what they need to or want to. ”
Through his work, David has started to address why people are so hesitant to really understand the data enough to tell a better story — to tell the correct story. And more importantly address why they are so fearful of telling that story.
“You need to be confident in your storytelling. If you take a direct approach to understanding the date and apply the right technique and rigor to telling a good data story, there should be no problem. You have to believe that it’s going to be interpreted the way you intended it to be interpreted.”
We have to do better with our stories. Not just making them clear to all, but making sure that they are the right stories.
There isn’t a simple path forward for teaching non-analytical team members to interpret dashboards and data visualizations.
Let data storytelling be your path forward and start coaching your team to:
- Start with the Story – Then Find the Data
- Do the Upfront Work
- Become a Better Interrogator
- Understand the Difference Between Wants & Needs
- Don’t Be Afraid to Prescribe a Story