Quantified Storytelling

Storytelling is often seen as an art; something subjective that taps into our emotions. However, as data analytics become more popular, it is becoming clear that even the art of storytelling can be quantified. According to a recent article from the New York Times, Vinny Bruzzese and a team of analysts offer the service of evaluating new movie scripts and comparing them to past scripts that have either become box office hits or have flopped. NarrativeScience from Northwestern University uses algorithms to extract key facts and insights from your data and presents those insights in an easy to follow narrative. At the 2013 MIT Sloan Sports Analytics Conference, ESPN put together a panel entitled “ESPN’s Use of Analytics in Storytelling” to discuss how anchors and writers are using data to tell better stories.

When we started researching quantified storytelling, we first observed that there are many ways to break down a story. In grade school, we learn how to do a plot diagram. We analyze a story looking for the exposition, rising action, climax, falling action, and resolution. Nancy Duarte, an expert in presentation design, outlines the pattern all great talks have of connecting the past and present to an ideal future. In a video interview, Ira Glass of NPR defines a story as nothing more than a series of events. Harvard Business Review portrays how sales linguistics data demonstrates that successful sales stories have an opening stage, main stage, and a closing stage (beginning, middle, and end). The following TED talk from Sebastian Wernicke displays a statistician’s formula for creating a TED talk. He takes the art of storytelling and simplifies it into an easy to follow template.

Although stories can be adapted for different purposes, the basic structure of every story remains the same. Even the shortest story ever told, from Ernest Hemmingway, “For sale. Baby shoes. Never worn”, has a beginning, middle, and end. It implies a series of events that are up to the interpretation of the reader. It was told with a purpose (supposedly to win a bet), it appeals to the emotions of the audience, and it follows a concrete structure, all in six words.

Nevertheless, as one critic in the New York Times states – “It’s [Quantified Storytelling is] the enemy of creativity, nothing more than an attempt to mimic that which has worked before. It can only result in an increasingly bland homogenization, a pell-mell rush for the middle of the road” – quantified storytelling is not without its skeptics. Storytelling has a 10,000 year head start. Only time will tell if data science can catch up, and we look forward to leading the charge and sharing our quantified storytelling analytics soon.