Data Storytelling in Flow

What is a data story?

Data, plus story? Can the whole be greater than the sum of the parts?

Is Cool meaningful; is it a scientific term?



Could you define 'data story' for me? What do we mean by 'data story'?


Sure, data storytelling is really about data communication. There are analysis use cases, which are really important. But data communication is, to me, a pretty big unmet need. Think about it, we've got the world's data, I mean, it's expanding it every day, so we're just inundated by data. But we have this problem of getting the right information into the minds of the right decision-makers at the right time. I call it the last mile problem. You've got fiber throughout the whole internet, but the challenge is to get that fiber the last mile into someone's home. Well, in this case, it's data. And we want to get it not just to someone's screen, but we want to get it into their heads - present it in the right way so it really makes a deep impression for them. They should get a good mental model of that information, and be able to drill in and explore it. So let me just talk you through this a little bit more rather than just talking about it.

If you're trying to communicate effectively, you need to engage your audience, nothing new there. How do we help an audience really understand the complexity, see the forest and the trees, the big picture, and the details? And then if you can get that audience to interact with the data, then it's going to make a much stronger impression. 


How would you say this is different than traditional data viz?


Good question. Traditional data viz often is focused on the analysis use case. Working with Tableau or Power BI, analysts are trying to discover the insights in the data. And it's a great use case, and we're not trying to compete with the data analysis space at all. But there's a certain gap between what happens on a dashboard, and the story that gets told to really make an impact.

On the other side of the coin, a lot of people are using presentation software to tell their stories. And sometimes they're copying and pasting screenshots of their Tableau dashboards into PowerPoint and then they add some lines and draw some circles. But when you do that, you lose some core aspects of the data: you lose the interactivity, you lose the discoverability, and you lose the energy. They don't let you walk through it and turn it around in your mind and build those mental models.

In between Tableau and PowerPoint is data stories, which we are focused on. It's a data communication use case, as opposed to data analysis. But think about it, if you can get your audience to understand deeply about a problem and your solution, that's where you can really start to have an impact on the decisions that are getting made. Maybe it's at an executive level, you're trying to communicate to your board, or you're trying to communicate to an executive in another company in a sales situation. Or you just want to communicate broadly, across the internet. And you want to get people's attention, in all of those cases, marrying both the intellectual aspects of the data and the emotional side of the story, that marriage is powerful.


So could you give me an example of a data story and what that looks like in Flow?


Sure. I'll drop it over my shoulder here.

So here, what I'm looking at is a map of the United States for the 2020 election, and this data is per county, Republican majority counties are marked red, Democratic counties marked blue. At first glance you could say it looks like the whole country voted red, but land doesn't vote people vote. And once you see these pillars pop up to show you the actual percentage of the vote results. Now you can really see that the election results felt like. First of all, all around the coasts and in certainly all the major population centers counties voted for Biden in this election. I can go ahead and click and interact with these dots.

I can see Sierra, who is on a simultaneous multi-user Flow meeting with me, so I can see exactly what she's clicking on. So Sierra, go ahead and use your laser pointer to point out some things, it's actually really nice to be able to see someone else in your space and see what they're looking at. 

I can see she's taking a look at Chicago. And as she does that, that helps me to tune my message. Maybe she is not an expert and doesn't know the jargon of a particular discipline. And for her to be able to say, "tell me about this thing", this becomes a technical term. And that's actually a really powerful mechanism to communicate to non-expert users.

Just to take you another step further through the story, you can actually see that we're going to change the color of the dots, making them shades of purple, which better reflects the reality on the ground. You can see that Joe Biden won Maricopa County by only a hair, and you can see the purple color of the dot.

To give you a sense of how these various kinds of visualizations and ways that you can communicate, this is an example of COVID reproducing throughout a population.

Here's an example of foot traffic from 2020. And this tells a really nice story. Because just by looking at these five brands, we can tell a story about what happened to us last year. So we start the year and we're off to the races. And whoo, there's a huge plunge in March! Gee, that's everybody staying home and not going to the stores. But then, as it progresses, you see the ups and downs of the economy. One of the stories in here is Home Depot: looks like an awful lot of people just got tired of looking at their old curtains, and went and started to remodel their homes, which you can see that by this spike in Home Depot traffic. Starbucks (by the end of 2020) still had not recovered well at all. And you can actually see the start the results of a second wave, people heading to target to stock up, and people avoiding Starbucks. Just a quick example of another set of visualizations that help you tell a data story.


This is pretty cool. But does cool really help me communicate better?


A lot of potential customers asked this question and I actually have a definition for the word cool. The Jason Marsh definition, which is mastery and control. So think about that for a minute. So you've got a skateboard trick that somebody just executed beautifully. They have such control over the skateboard and over gravity. We see that we go "Oh, that's cool!" Or maybe it's Tony Stark flicking his atoms around as he's trying to create a new element to power his Iron Man suit? That feels cool, because he's got complete control over his information: he can just manipulate it exactly the way he wants.

So cool is mastery and control, and what you're able to do is have that control over your information and have that ability to communicate those ideas more effectively to your audience. That's not just cool. That is business efficiency. That is shorter sales cycles. That is helping an audience to understand a new point of view, and take an action and move forward. That's what we mean when we talk about cool. I define it as mastery and control. And it's not just cool. It's gold. But yeah, it is cool.