The Secret to Data Storytelling

Hello, Data Radicals. Let’s start today with a spoiler alert.

If you’ve never seen the movie The Sixth Sense, then stop what you’re doing right now and go watch it.

If you haven’t seen the movie, I’m going to play the alert, and when it’s over, hit pause.

So, now that you’ve seen the movie, you should know The Sixth Sense was released in August of 1999, and it became one of the biggest movies of all time.

M. Night Shyamalan wrote and directed the film to create a terrifying, chilling reality.

But what ultimately made the movie so successful was that twist ending — where they reveal that Bruce Willis’s character had been a ghost the entire time.

That’s the genius of The Sixth Sense.

Yes, the movie is spooky and well-crafted.

But it’s how the story was told that hooked audiences.

Upon revisiting the film, you notice the color red is associated with death.

You realize that Bruce Willis is not having marriage problems.

His wife just has no idea he’s even there.

But it’s that final twist that made the movie so memorable.

Many people see the movie again — just so they can figure out how the twist was set up.

Behavioral economist George Lowenstein calls this the “gap theory” of curiosity.

He argues that “gaps cause pain. When we want to know something, but don’t, it’s like having an itch that we need to scratch. To take away the pain, we need to fill the knowledge gap.”

If you’re a data analyst, another way of looking at your job is to fill the gap, to tell the story that allows the data and world around us to make sense.

Principles of Storytelling

Storytelling is an essential part of our field.

And the principles that create a great story, like The Sixth Sense, are the same principles that can help us tell stories with data.

We can create intriguing riddles for our audience that seemingly cannot be explained.

We can use color in our presentations to draw the eye towards certain elements.

We can set up data mysteries that can only be explained by a compelling conclusion.

And above all else, we have to keep our audience engaged — whether or not we have a surprise twist ending.

And, NO, this is not about telling the story of how you got to the insight.

How you first sourced the data, then profiled it, then appended the dataset, then built a scatter plot, and then applied a regression.

No one but your manager cares about that story.

This is about making your data tell the story about the world that produced it.

Today, we’re speaking with Cole Nussbaumer Knaflic, founder and CEO of Storytelling with Data, which shockingly enough is about… storytelling with data.

After you listen to this episode, I can’t guarantee your storytelling will pack the same punch as The Sixth Sense.

But I can guarantee you’re going to leave this episode ready to make your next presentation a heck of a lot stronger.And you’re never going to look at tables the same way again.

Producer Read: Welcome to Data Radicals, a show about the people who use data to see things that nobody else can. This episode features an interview with Cole Nussbaumer Knaflic, founder and CEO of Storytelling with Data.

In this episode, she and Satyen discuss the fundamental principles of storytelling, the importance of public speaking, and how feedback can help transform organizations.

Data Radicals is brought to you by the generous support of Alation, the data catalog and data governance platform that combines data intelligence with human brilliance. Learn more at Alation.com.

The Most Important Factor When Communicating With Data

Satyen Sangani: Ready to tell your story?Well, you can’t just jump right in.

Before you can begin, you have to identify your audience and their expectations.

Cole Nussbaumer Knaflic: I would say this is the most important factor when communicating with data — when communicating in general — because the go-to we often default to when we’re communicating, as we communicate first and foremost for ourselves, is something like:

• “I want to communicate the project I just finished.”

• “… the analysis I completed.”

• “… the answer I found.”

We actually want to shift that around and think about when we’re speaking or presenting, whether it’s formally on a stage or in a meeting with colleagues, that we’re considering first and foremost the people who are on the receiving end of that. And when we do so — the more we can take their needs and desires into account — the more successful that interaction is likely to be.

Some people want to know how much it’s going to cost. And if that’s the priority of the audience I’m communicating to, I can lead with that, so then we can return to the details after that need is sated. Or for the people who want to know the theory and all the details that you looked at along the way, that’s a different path when you know that’s what your audience is going to crave.

So, What Is a Story? Where Does It Start? What Are Its Elements?

Satyen Sangani: So a story, on some level:

• Starts with knowing your audience

• Ends with some call to action, broadly defined

So, what is a story? Where does it start? What are its elements?

Cole Nussbaumer Knaflic: I’m going to describe two different structures of story, but they’re really the same structure. One is just more simplified because if you’ve never used these tactics before, then going from a standard monthly report to “All right, folks, I’m going to tell you a story!” may be crossing too big a gap right away. So, there are steps we can take along the way where we’re still sort of using “story,” but not going full-fledged.

The simplest version of a story has a plot, some twists, and an ending, and you can think about how different business scenarios or different analytical projects might map into this:

• The plot: We had a question about a piece of our business.

• The twist: We explored the data and we found some things we expected, but we found this one thing that was interesting. It was unexpected and can actually change the way we do things.

• The ending: The action we can take, based on this new information.

The cool thing about this simple story is you can apply it to a whole communication, but you can also apply it to a single graph. What does the graph tell me? What’s the new piece? What’s the twist and what’s the ending? The ending is my articulation of the takeaway. The story helps communicate the meaning of the graph.

And this is an easy practice to help you communicate with graphs. Just write a sentence about what you want your audience to see or know. If it makes sense, write that sentence directly on the page, or do things to draw attention to that part of your graph.

Even in these simple ways, you’re storytelling, which is just answering the questions, “What’s the takeaway? What do I need to know?” Making that clear is a simple version of story.

We can get a little bit more nuanced, though, and build on this. What we teach in our workshops is the narrative arc, which is just another way to characterize stories. It’s a little bit further fleshed out compared to that simple story structure that I mentioned.

Narrative arc starts out with a plot. There’s a sense of place and time, there might be characters or different players, and then as we move through time, tension is introduced and builds in the form of a rising action, reaching a point of climax, where tensions are their highest. Then there’s a falling action that acts as a buffer that leads to the ending.

We talked about message a little bit earlier, but one of the things we teach is to really reflect on — when they consider their audience and their message — is what’s at stake. What stands to break — or what are the risks — if the thing we need to happen doesn’t happen? Or the flip side, what are the benefits or the good things to come out of the action we’re trying to inspire?

Satyen Sangani: Right. What are the stakes? What are we playing for?

Cole Nussbaumer Knaflic: Exactly. Because then if we think of that arc, the tension we’ve brought to light plays back directly into what’s at stake.

It’s not what’s at stake for us as the communicator or the analyst, it’s what’s at stake for our audience. Because if we can frame our story in the context of that, and the tension we bring to light is meaningful for them in that way — because it taps into something that’s at stake for them — then the ending to our story becomes what the audience can do to resolve the tension you’ve brought to light.

It really brings the audience into our story in a way that often gets missed when we communicate in a business setting, where the approach is often very linear:

• Here’s the problem we set out to solve.

• Here are the things you did to the data you have.

• Here’s the long list of caveats about the data and our methodologies and what we went through.

It’s a very long, drawn-out, linear process — over the course of which it’s very easy to ignore our audience. For me, that’s the biggest shift. The biggest benefit that happens when, rather than communicating this linear way, we rethink things in the shape of this arc. In order to have that shape, there’s tension present.

This isn’t about making up attention. There’s always tension present.

If you have something worthy of communicating, it’s just about figuring out what that is in this scenario and how do you adequately bring it to light for your audience. It’s also deciding when you get their input versus pushing them to do something, and making all of those pieces of the puzzle fit together.

There are a lot of different nuances playing into every different scenario when we communicate. And yet, we often reach for the exact same solution when we need to communicate, rather than thinking about what a success looks like this time:

• How can I align what I create?

• How do I talk through it?

• How do I prepare myself, whether I pre-wire with my audience or get with people ahead of time?

• How do I make all of those things work together in a way that’s likely to help me achieve the outcome that I’m after?

Because when we do that, we can do a lot more.

What We Don’t Need to Communicate in Storytelling

Satyen Sangani: The Sixth Sense is a perfectly-crafted film.

Surprisingly, it didn’t come out fully formed.

Many scenes were left on the cutting-room floor because they didn’t advance the narrative.

Deciding what to leave out can be incredibly difficult.

So how do you decide what to cut?

Cole says, it comes down to your audience, your story, and the time that you have to tell it.

Cole Nussbaumer Knaflic: If we think of stories in other realms, such as books and movies, it’s not this smooth, single ascent and descent. It’s more like a jagged mountain where things go up and down all the time, pushing us forward as they do. That’s useful over the course of a 400-page book or a two-hour movie, or as we’re doing the analysis behind the scenes to get to the answer or the thing we need to happen.

What we don’t need to do is communicate every single one of those peaks from the jagged mountain to every single audience. It’s a matter of taking our specific audience in a specific instance and figuring out what combination of pieces I need to curate now for them, and how I weave them together. This means the same story looks very different for the venture capitalist (with whom you may have only 30 seconds), and for your finance partner (who’s going to want the details before they sign off on a budget). The work behind the scenes has to happen, but when it comes to how we communicate it to others, consider a curation process of picking the necessary pieces of all of that. We do need the context of the rest of it in case questions come up, but we choose what’s required to meet the needs of a given scenario.

Public Speaking: How People Will Pay More Attention to Your Data

Satyen Sangani: A director has to wear a lot of different hats.

They have to deal with camera angles, approve costumes, and, of course, work with the actors.I guarantee you there are some parts directors enjoy more than others, but all parts are necessary to make a great movie.

I know many of you Data Radicals don’t enjoy public speaking.

But like it or not, it’s part of our job. And Cole says if we don’t cultivate that skill, it’ll show up in the impact your work will have.

Cole Nussbaumer Knaflic: You will be limited in your career. There is so much potential for the person who has strong technical skills and can also speak in ways that make other people feel engaged and attentive.

People will pay more attention to your data and what you do with it if you can talk to them in ways that make them want to keep participating and be part of that discussion. It’s not easy, particularly for people in technical roles, because it’s likely they are in technical roles because they didn’t want to have to do this stuff.

I’m an introvert. My most comfortable place in the world is by myself in a room behind my computer screen, but I wouldn’t be able to drive much change if that is where I remain. For me personally, it was finding an area I was passionate about and then realizing I can take that passion and harness it and hone it in ways that will help more people.

I mentioned being strategic about how you communicate when you’re in a room with people and you’re talking to them. You get real-time data the entire time because you get to watch them and you get to see these micro facial movements, such as when brows furrow or lips get pursed. Moving around the room changes how people behave.

There are all of these really interesting ways you can use yourself as the communicator of information that I didn’t understand at all when I was starting out and making pretty graphs. This stuff that came later when I was teaching and the point where I got to know my content well enough that I was finally comfortable enough to focus on not only what I’m saying, but how I’m saying it:

• How I use my voice

• How I use my body

• How I use my hands

• When to be in the front of the room so people can see me and my materials

• When to move around and be somewhere else entirely so the focus changes

These techniques are a sort of art form that should be taught in schools because it can be learned and honed, and it shouldn’t be a trial-and-error thing everybody has to individually figure out over time.

Tips, Tricks, and Best Practices That Data Storytellers Use

Satyen Sangani: Every movie director has their favorite tips, tricks, and best practices.

M. Night Shyamalan is famous for his twist endings and long takes.

Spike Lee invented a unique dolly shot.

Martin Scorsese loves to use “Gimme Shelter” by the Rolling Stones.

Are there tips, tricks, and best practices that data storytellers use?

Cole Nussbaumer Knaflic: Absolutely! Tables are one of those things that surprises us, because simple tables are usually the first thing we reach for when we’re aggregating our data. But think of how people process information: If you imagine a table in front of you, you’ve got rows and columns and data in each…so you’re reading that table.

When you scan down columns across rows, you’re trying to compare values, which means you’re holding information in your head. Trying to compare things is a highly taxing cognitive process because when you use a table you’re using your verbal system. That’s a big benefit we get from thinking about showing data and graphs, because that is accessing our visual system, which is much faster at processing information when it’s designed well, than our verbal system, which means data in a graph will typically be understood more quickly than data in a table.

One way of showing data is a type of graph. Often, it’s worth considering different views of our data: both as part of the analytical process and so we can get a better understanding of the nuances of the data; but also, as we’re thinking about how we might subsequently communicate that information to someone else:

• What’s going to help create that “Aha!” moment of understanding?

• What do I want my audience to see?

• Am I getting them to compare things or see a trend over time?

And the answers to each of these questions can help point you to a good sort of menu of graphical options that might work. There are a common few you’ll be reaching for most of the time — bar charts, line graphs — because these are familiar. One thing to be aware of as we’re designing graphs is when we use something that’s unfamiliar to our audience, we’re introducing a hurdle. Because we have to get them either to want to spend enough time with the thing — to figure out how to read the thing — or if we’re presenting it live, we have to keep their attention long enough to make that happen.

So make sure that if you’re using something novel or less familiar to your audience, that that’s a worthwhile trade-off because some amount of time has to be spent understanding how to read the data before they can get to the data. When there’s an opportunity to use just a bar graph or a line graph and get people focused less on how to read the graph and more quickly on what the graph says or why it’s interesting, we may need to reach for something new or different because that observation isn’t obvious when using some of these more common forms.

How to Draw the Audience’s Attention With Data Storytelling

Satyen Sangani: In The Sixth Sense, M. Night Shyamalan uses the color “red” very sparingly.

This helps to draw the audience’s attention during dramatic moments.

Apparently, data storytellers can do the same thing.

Cole Nussbaumer Knaflic: Each time you show a graph, what do you want your audience to see? How do you get rid of the things that might be impeding that from happening? That’s where clutter can come in: grid lines and other visual elements that may not be necessary. We can think about pushing those to the background or getting rid of them entirely.

Where do we want our audience to look first? What’s the most important thing for them to see? How can we use contrast sparingly in those areas to make that happen fast? Typically, the lowest-hanging fruit, when it comes to focusing attention more quickly, is being sparing and selective where and how you’re using color. So we’ll often design graphs in shades of gray and then use color really sparingly and intentionally to draw an audience’s attention to where we want them to look.

And when you also use the strategy I mentioned earlier, where you connect a sentence to each graph, those two things together can work really well. Neither takes very long and both can go a long way in making your graph quickly understandable to your audience. Use color sparingly to focus attention on where you want them to look, and then use words that tell your audience why you want them to look there. You can take a graph — even if it wasn’t the ideal graphical form or if there’s some clutter present or other issues — and still make it work in that instance.

These are just a few of the basic sort of lessons that we teach, and none of it is rocket science. It’s all stuff that when you step back and think about it, makes common sense. It’s only when we pause and think about it, that we discover we’ve developed bad habits that can push us in some not-so-great directions. So, it’s good to remind people of that.

How to Build Data ACES Within Organizations

Satyen Sangani: The first draft of the screenplay for The Sixth Sense looks a lot different from what you see onscreen.

In that version, the character Cole was a crime scene photographer, and the plot revolved around a serial killer.

The movie’s famous ending was only introduced in the script’s fifth draft.

That’s a testament to the power of feedback.

You’re not going to get things right on your first draft.

So make sure you’re getting another perspective.

Cole Nussbaumer Knaflic: Feedback can take a ton of different forms, but it can be as easy as, for instance, the next time you find yourself making a graph or a presentation you’re about to deliver, tap someone across the hall, or virtually grab somebody and share it with them. Getting their perspective does a couple of things:

You get input from someone who has a fresh perspective, which can be useful because we sometimes get so deep into our areas of expertise that we lose sight of things that are not tacit knowledge to everybody else. That’s very useful in helping frame things for your eventual audience — who’s going to naturally be more distanced from your work than you are.

The person who’s giving you feedback gets benefits, as well. One benefit is they’re honing their logic, as I mentioned a moment ago, and another is when you put yourself in a vulnerable position and welcome input as a result of that. (By the way, if you ask for feedback, always listen and try to incorporate something from that feedback into your work!) By modeling the feedback process for the other person, they are now more likely to seek feedback, either from you or somebody else.

That’s the way that good feedback can start on a grassroots level: We start checking our work with each other, we have better conversations about it, and we do better work now as a result of all of these things. Imagine that at scale, and it becomes really powerful. It takes time. So it does take support to be able to do that and spend the time on these parts of the process.

Communication of the data is the part where the work either succeeds or fails. As organizations look to build their cultures around data, the more support that you can give, whether it’s resources or just the time for people to spend practicing and giving feedback and honing how they’re talking about their data — how they’re showing their data — all of that is going to pay off in positive ways.

Building a Culture of Data

Satyen Sangani: This idea of a culture of vulnerability is a super-critical aspect to building a culture of data because on some level, the idea that you could be wrong is exactly what data is all about, right? If we all came in with our biases and assumptions that we’re all right, then that’s a sort of top-down, highest-paid-person-opinion (HiPPO) kind of culture. The idea of saying, “Look, you need to be vulnerable in the work itself, but also everybody has to be vulnerable” — and getting feedback — is so critical to this notion of sort of letting the data tell you something more directly.

Cole Nussbaumer Knaflic: One way to do that — particularly if it feels scary or countercultural to how things have been done — is to start off by not making it about the day-to-day work. At your weekly team meeting, you can have each attendee in charge of finding a graph in the media and then, a week at a time or a month at a time, you talk about each of these about what works well or what doesn’t work well, or look for examples that are outside of the day-to-day work. If it feels like the stakes are too high in the workplace, this could be one way of just getting people used to critiquing other people’s work and having those sorts of discussions.

In other places, we have an active online community where people can post feedback requests. We have an exercise bank. We also run a monthly challenge where each month the topic varies. The goal is creating a safe space where people can practice and flex their skills and try new things and see what works and what doesn’t and exchange feedback.

Cole Nussbaumer Knaflic: By doing challenges like these, or making up your own and talking about them, increasingly it becomes easier to request, receive, and give feedback and then having those sorts of fruitful exchanges in the day-to-day work as well.

Satyen Sangani: There’s so much more to being a Data Radical than knowing the data.

Telling compelling stories with our data can help organizations make better decisions.It can also help us get everything we want out of our careers.

So the next time you’re getting ready to deliver an amazing presentation, keep this conversation in mind.

Maybe even go revisit The Sixth Sense or another of your favorite movies.

Think about what elements make the movie so effective — and how you can deliver that same impact.

We’ll see you next time.