Story telling in data science

Advertisement
Font Size

Taxwise Or Otherwise

What is data storytelling?

It is a structured approach of communicating insights from data through stories and visuals to an audience.

To some, data storytelling may seem like an oxymoron. People often view data analysis as a technical function and storytelling as a creative task. Data and stories are perceived to be opposites in the cognitive process. This thinking is fueled by the left-brain/right-brain myth — one that has since been debunked by scientists from the University of Utah after they scanned more than 1,000 brains in 2013. There was no evidence that people dominantly use their left or right brain. All of the regions in the brain enable humans to engage in both creative and analytical thinking at the same time.

With the unprecedented increase in the amount of data that businesses create — whether from their social media, purchases, sales transactions, or customer profiles — the potential value from data also increases. However, we can only realize the value of data when the insights we derive move people into action. The question is: How exactly can we do that?

To move people into action, data and stories must work together.

Telling stories is more effective because it is more memorable and persuasive than reporting statistics. Chip Heath, a professor at Stanford University, underwent an exercise to test his students’ ability to retain information when stories are used versus statistics. After the students heard different speeches, 63% remembered the stories and only 5% remembered any individual statistic.




In a separate study, researchers tested the tendency of individuals to donate more to named victims (presented as a story) than to statistical victims. For example, ‘Baby Jessica’ fell into a well in Texas in 1989 and people sent over $700,000 for her rescue effort. Statistics (e.g., thousands of children will almost surely die due to undernourishment) seldom arouse strong reactions.

However, those that dive deep into data, such as data scientists, may not be naturally skilled in storytelling. Data scientists are skilled in managing data, building models, and running algorithms, but may not have the skills to effectively communicate insights from data.

How do we bridge this gap and fully realize the value of data? We must develop a culture of data storytelling in our organizations. It is our responsibility to give justice to the insights from data through effective storytelling.

The three components of data storytelling

There are three components to data storytelling: data, visuals, and narrative.

Visuals transform the data into illustrations that help uncover insights that are hidden in the facts and figures. They give the picture or the vision that helps people absorb what the data represent. However, they do not provide context on why the data should matter to the audience.

The narrative tells the reason behind the statistics and why the audience should care. The storyline also enhances the audience’s understanding, and is our best defense against confusion.

The intersection between visuals and the narrative is also a critical factor in driving people to change — emotion. While the story aids in understanding and encourages empathy, the visuals enhance the sensory experience that captures the audience’s attention. This combination of story and visuals makes the audience more emotionally receptive to the message. However, data are still essential to supplement this emotion with hard evidence.

Data make up the soul of the data story. The visuals and the narrative work together to bring the data to life. Stories build that emotional connection between the audience and the data, while the data itself establishes trust.

When people are presented a vision with clear reasons that stir their emotions, we move them into action.

Antonio Damasio, a neuroscientist, studied people who had brain injuries, specifically those whose prefrontal cortex — the part of the brain that is responsible for emotions – is damaged. They performed normally in all of their human functions, but they had lost the ability to feel emotions. He found that their ability to make decisions was seriously impaired. Subjects could logically describe what they should be doing, but they found it challenging to make decisions about where to live, what to eat, and other day-to-day choices.

This shows that humans make decisions based on emotions and not on logic. Without emotion, it is difficult for humans to make decisions. These decisions range from the amount of money to donate to charity to selecting which projects to pursue. Emotions are shortcuts built into our brains to generate feelings that guide us in making decisions and direct us in taking actions. This, coupled with a clear vision and a moving reason to pursue an initiative, helps sustain the momentum of the action.

Without data storytelling, more and more of the value of data could remain unlocked. We need to break down the notion that data analysis and storytelling are two irreconcilable skills. By communicating the data through storytelling, only then can we harness insights, inspire change, and move people into action.

The views or opinions expressed in this article are solely those of the author and do not necessarily represent those of Isla Lipana & Co. The content is for general information purposes only, and should not be used as a substitute for specific advice

 

Weddy Anne R. Diamada is a senior associate and the data analytics champion at the Deals and Corporate Finance Group of Isla Lipana & Co., the Philippine member firm of the PwC network.

63 (2) 845-2728

weddy.anne.diamada@pwc.com

Advertisement