Storytelling and Data Visualization

A man

Millions of stories are happening in our world—often simultaneously—each and every day. All stories worth telling have a rising action, a conflict for the central character and a resolution. It’s a familiar lesson to English majors, but this same lesson is crucial for data scientists to communicate actionable insights through data visualization. The data science industry needs professionals who can visualize information to tell a rich, compelling story. Data storytelling isn’t simply a combination of arts and science. It’s being able to use art to present the science, influencing business decisions by explaining how we got here and where we’re going.

Elements of data storytelling

Telling stories through data teaches us more about our world. The narrative can help organizations uncover something new without having to digest complex data sets. Data scientist Ben Wellington tried to demonstrate this to an audience unfamiliar with data storytelling by using large data sets in the Open NYC project.

Wellington uses data to build a narrative about the New York people are actually living in. But instead of using weighty, traditional statistics or text-heavy PowerPoints, he’s using visual storytelling with non-traditional datasets to explain the daily trials and tribulations of New York residency to the audience. His efforts resulted in the Department of Transportation making changes based on his recommendations.

Storytelling with data allowed that audience to see a coherent picture of how New York’s Open Graph can be mapped and made relevant

How to begin a data storytelling strategy

First, we should understand a critical point about data storytelling. When building a narrative, we should understand that data visualizations are meant to show a moment, but narrative explores a period of time. Stories must have a beginning, a middle, and an end. Data science makes each part of that data story rich, and our role as an analytics professional is to use a flow that makes it valuable and interesting to our audience.

Data analysis professionals need to apply solid data visualization techniques first. Here are a few we can start with:

  1. Know your medium. Are we building a narrative for an in-person presentation where we have limited time for the audience to digest findings? Or will we be creating content marketing e-books where longer, more complex stories can be told with more detailed data analysis? However you present your data visualization, it must be tailored to the amount of time your audience will allot for the narrative.

  2. Use colors that fit the narrative.. In order to show confidence and trustworthiness, brands use a deep blue color. In order to show profits, the human brain expects green or black. In order to show losses, our minds expect to see red. Ensure that your story is supplemented with consistent design technique across your data visualizations.

  3. Choose your conflict, rising action, climax and resolution.. Before you begin to build your content, sit with your data and imagine how the stories will unfold. Did you perform enough data analysis to communicate a credible problem or insight? Have other data scientists told a similar story that will help you build the foundation for your own? When you start here, you’re able to sequentially uncover the rising action, your main insight and the problem’s resolution. These literary devices are vital for storytelling with words and just as vital for storytelling with data.

  4. Use clean data.. Data storytelling must persuade an audience. Your credibility is important, and data sources that are meticulously vetted for accuracy and bias are powerful narrative elements. The resulting data visualization needs to do more than deliver insights. Each data visualization in our data story needs to help communicate that the path forward is one your audience, whether peers or stakeholders, can confidently take.

How companies use data storytelling

Companies are beginning to use data storytelling in exciting ways. They recognize the importance of sourcing data analysis and finding the right talent for their data science teams.

Spotify understands this well. Spotify’s contextually-relevant advertising narratives focus on the time of year to present data insights as a creative story to their consumer target. In their annual end-of-year ad campaign, insights from the Spotify data science team humorously highlight users’ listening habits. Besides showing the size and scale of the product, their data scientists used data to place the customer at the center.

Media companies are using data storytelling to help their clients as well. Intersection, the company behind LinkNYC, a prominent digital advertising kiosk found all over New York City, uses their data feeds from weather providers and available New York data to get commuters’ days started with lightly-branded content. The Weather Channel, The New York and Brooklyn Public Libraries and Instagram are all featuring contextually-relevant messages and visuals using to start the day for NYC commuters.

Consumer brands have numerous examples, but business-to-business (B2B) corporations and consultancies are crafting narratives for their content marketing in whitepapers and presentations. McKinsey’s quarterly publication builds a narrative for their consulting product using client-centric language. Their visualizations take advantage of the recommended techniques: using color and purposeful visual design, and bold, easy-to-digest graphs that draw immediate inferences from large data sets. Good storytelling with data can often be a pleasant surprise for B2B companies since their professional target is conditioned to expect traditional, text-heavy whitepapers.

Tips to help guide data storytelling

Data storytelling is a complex process and, when it comes to choosing and parsing your data, it is easy to lose your audience, the most important element. Data storytelling veterans know how to keep focus by adhering to these principles:

  1. Know your audience. People get excited about a message when the information is relevant to them. For your data to grab the audience, your findings have to be timely, emotionally significant and in a style your audience can connect with. A presentation to an audience of parents at an elementary school will use very different language and visualizations than one for a room of investors.

  2. Make it people-focused. The oft-repeated business advice “start with why” couldn’t be more relevant when data storytelling. Introducing the importance of your narrative before you introduce the story itself can make a difference in engaging the audience. If you’re building content marketing initiatives such as a white paper or branded video, it may be more effective to withhold data visualizations at first. Introduce, instead, the importance of the story and the impact your audience can have by using it to take action in their own lives. Once invested, they are ready to digest the next step in the journey, whether it’s a written narrative from your data analytics or a rich visualization.

  3. Understand the strengths and weaknesses of data visualizations. Will a bar graph or a histogram more accurately tell your story? Will a pie chart be the best way to communicate the impact of a large data set? Tableau notes, “Pie charts are powerful for adding detail to other visualizations, but aren’t as effective on their own.” Keeping the most effective use of charts in mind will be a critical step in laying out your data analytics work in the most relevant format possible.

Why the MS in Data Analytics and Visualization Online at Yeshiva University

The Katz School of Science and Health at Yeshiva University (YU) offers the Data Analytics and Visualization online program to give students a well-rounded background. Not only training to build and manage complex data sets with Python, SQL/NoSQL, Tableau, AWS, and AutoML, but also how to humanize this data into a rich story that makes a larger impact.

A mindful approach to data storytelling galvanizes the effort of individuals and corporations to use analytics to make an impact on the world. At all levels, an audience is made of human beings who live in an era of noise and distraction. A sharply-crafted narrative from data scientists that weaves each visualization into a larger story will be key to all businesses that rely on data analytics. Learn more about applying to our flexible online Master’s in Data Analytics today.

Share this article