top of page

Practical tips for data visualization and storytelling.

Join over 800 people who receive tips every month

Thanks for submitting!


How to Become a Better Data Storyteller - Part 2

Demystifying the Narrative Arc

You’ve heard this before: data storytelling is hard. It requires special skills. You either have it or you don't. This belief is particularly common if you think of yourself as being a data person, not necessarily a storyteller. But how hard is data storytelling, really?

Data storytelling

Personally, I believe that any skill can be learned. No matter how easy or hard that skill is. Generally speaking, I always like to break down the process of learning a new skill into smaller, more achievable steps:

  1. look for inspiration

  2. understand the "ingredients"

  3. apply the lessons


1. Look for Inspiration

If you're looking inspiration, the sources are endless. Remember how your mom or dad used to read you bedtime stories when you were young? That might have been your first exposure to storytelling. Or, do you remember how a few years later, you learned how to read and you started flipping the pages of books such as The Adventures of Nanny Piggins or Battle Bunny? We all grew up with stories.

Whenever we want to relax or be entertained, a story is our go-to choice. So whether you think of yourself as someone who is good with words or as someone who is good with numbers, stories are your second nature.

Kids books

In this blog post, I'm going to explore a specific genre of storytelling - documentaries. I love watching documentaries. Like data, documentaries help me broaden my horizons. There is so much that one can learn from documentaries in terms of storytelling.

Let's explore what "ingredients" can be transferred from one type of storytelling (documentaries) to another (data storytelling).


Learn more about our virtual data storytelling and visualization training for your team!


2. Understand the "Ingredients"

The "When"

As I was diving deep into the narrative structure of documentaries, I came across a wonderful interview with Doug Block, veteran documentary filmmaker.

In the interview, he mentioned something that resonated with me and that is applicable to data storytelling:

"You hope for a narrative arc before filming and try your very best to get it while filming. It can be elusive and go in surprising directions, of course, and editing can certainly help bring it out. But you need to be thinking about it all the time from the earliest stage.”

Just like in the case of documentaries, we need to be thinking about the narrative arc before we start creating a chart or before we start drafting a data presentation. In fact, I'd go as far as saying that you should consider the narrative arc before looking at the data. Based on what you know about your audience and about your organization's needs, you should have an idea of what the narrative arc should look like. So sketch it out and then refine it along the way.

Now, let's see how the typical documentary narrative arc looks like.

The "What"

While each documentary has its own unique style, the typical narrative documentary narrative arc is comprised of three acts:

  • Act 1: Intro to the topic, hero, and conflict

  • Act 2: Obstacles and seat-backs

  • Act 3: Resolution

Act 1 introduces the characters and the challenges that will later be covered throughout the documentary. The goal is to use Act 1 to capture the audience's attention, to ensure that the audience is intrigued and wants to find out how the story unfolds.

Act 2 includes many scenes through which the tension builds and the conflict is detailed. In this middle act, the filmmaker is trying to keep the audience engaged and provides solutions to the challenges mentioned in Act 1.

Act 3 is the resolution. In this final act, the story comes full circle. The filmmaker reiterates the original conflict and the solution. Potentially, the filmmaker also paints a picture of what's next to come.

Example: Vox - The Race for a Vaccine

To "test" the feasibility of this narrative arc, I watched and carefully evaluated the the documentary The Race for a Vaccine, produced by Vox in partnership with Netflix.

Act 1 introduces the topic, hero, and conflict. The topic in this documentary is the race for a COVID-19 vaccine. The hero - the public and private organizations that are working tirelessly to create, test, and produce a vaccine. The conflict - timing is critical and we need heard immunity. Act 1 leaves the audience intrigued and eagerly waiting to learn more: "How fast can we develop a COVID-19 vaccine?"

In Act 2, obstacles and seat-backs are introduced one by one, as the story unfolds.

The first obstacle to making a vaccine available fast is the clinical trial stage. Typically, clinical trials take 4 years and vaccines need to be tested on 5,000 people. As a workaround for the COVID-19 vaccine, scientist are planning to test the vaccine on the same number of people, but shorten the trial timeline. Shortening the timeline comes with its own challenges and raises ethical concerns. The documentary provides examples of past vaccines that were rushed and that had disastrous effects.

The second obstacle is resources - public funding is needed to develop and test a new vaccine. More often than one imagines, vaccine trials are cancelled because of lack of funding.

Once a new COVID-19 vaccine is tested, the next step is production. Remember, we need heard immunity in order to eradicate the virus. This means that 60% of the world population (or close to 5 billion people!) need to be vaccinated. This raises the issue of scale. Currently, we don't have enough factories to produce that many vaccines. Factories are currently being built.

Final obstacle - politics. As we've learned from history, politics and medicine are intertwined. Politics will likely influence what countries first get access to a COVID-19 vaccine.

Act 3 reiterates the idea of the race for a vaccine mentioned in Act 1. The meaning of "race" is extended in this final act. The race is at this point described not as a competition, but rather as a process of all of us coming together to end this pandemic. In a record time. And a clock starts ticking...

3. Apply the Lessons

As I was writing this article, I came across a remarkable data storytelling which follows the 3-Act narrative structure described above: a map of the COVID-19 deaths created by The Washington Post in partnership with Lupa and Google News Initiative.

Act 1 introduces the topic and the conflict. Topic - COVID-19. The conflict - people have a hard time seeing the scale of the lives lost because they didn't all happen around them. Before moving to Act 2, a question piques everyone's interest: What if all COVID‑19 deaths in the United States had happened in your neighborhood? I was intrigued!

Act 2 starts at your location. It then introduces the scenarios one by one. In doing so, the story advances from the scenario in which one of your neighbors died, to slowly zooming out and presenting a timeline of the growing deaths around you. In the end, in this hypothetical scenario, everyone within a radius of 1.3 miles around me dies. It was a terrifying map to see.

COVID-19 deaths in my neighborhood

Act 3 leaves the simulated scenario behind, and shows us the actual map of COVID-19 deaths in the US. Like in the Vox documentary, Act 3 comes full circle and reiterates the idea that all the 202,329 US deaths could have happened close to home, in your zip code. Then, similar to the Vox documentary, the meaning of seeing the deaths around your zip code on the map is extended. The data story ends with this:

"Since these people were not around you, it may be hard for you to see the scale of these losses in your own daily life. But each one was a tragedy for those in their circle."


People and stories have a lot in common. We love stories - they entertain us, inform us, give us new lenses to see the world. Whether you're a quant or a fan of words, storytelling is a skill that you can learn.

As a next step, take the documentary narrative arc and apply it to your next presentation. Start with Act 1 - introduce the topic, the hero, and the conflict. Make sure your audience is intrigued and curious to learn more. Then, move into Act 2 - introduce the obstacles and the elements of the story one by one. Let the story unfold itself. Finally, Act 3 - present a resolution. Reiterate the original conflict. Expand on the initial meaning of the topic and conflict.

Practice, practice, practice. Eventually, you’ll master the craft of data storytelling.

1 comment

1 Comment

Scott Arnold
Scott Arnold
Apr 10

The word is "set-back" not "seat-back." The latter is the thing that has to come to a "full and upright position" prior to the plane landing. ; )

bottom of page