Why data storytelling is a great skill to have
Beyond the Cloud - your partner for AI
Beyond the cloud - your partner for AI > AI-Driven Innovation and Intelligent Process Automation. <
Before we start: download my latest 1000+ book on Machine Learning
Data storytelling transforms a set of data into an understandable storyline for as broad of an audience as possible.
With the ubiquity of advanced data analysis tools now available, anyone can take a set of data and visualize a graph from it
Yet, according to Storytelling with Data author and guru, Cole Nussbaumer-Knaflic, modern schooling methods do not teach learners how to communicate effectively with data.
Many of us spend years of formal education learning about numbers and language, yet rarely are these combined in a way where we use our language skills to tell stories about a set of numbers.
We humans are natural storytellers, yet we are not naturally adept at telling stories about our data.
The result is a sea of ineffective, unexplained, confusing charts and graphs.
How can we do better? We can start with an understanding of how to combine numbers with narratives.
Different ways of thinking
Despite this modern educational gap, our brains have evolved to think both quantitatively (what, how much) and qualitatively (why, how).
While modern education often fails to merge quantitative data with narrative, diverse cultural approaches remind us of the value in doing so.
During my research for a project, at the intersection of computer science, Indigenous knowledge, and employment, I spent time researching how various cultures integrate both quantitative and qualitative thinking for decision-making. This blending of data with narrative illustrates effective data storytelling, offering a model for combining diverse ways of thinking.
Hard numbers work well when we are dealing with absolutes — they provide a static, abstract view on a set of data. However, this does not always transfer to the reality of daily life for someone who rarely, if ever, deals in abstractions.
An Indigenous fisherman, for example, would base their decision on how, where and what to fish for on approximations from many different inputs.
Weather, temperature, water depth, time of year, and their internal knowledge of their past results based on these variables. Adaptive thinking is essential — yesterday’s data may have no relevancy on the decisions that need to be made today.
The research of Berkes and Grant (2007) on qualitative thinking of fishermen in Grenada highlights the practicalities of adaptive, real-time qualitative thinking in decision making. In fact, many Indigenous folks are baffled by the need of Western scientists to quantify everything in excruciating detail. This Western methodology often misses the underlying relationships between the different pieces of data.
Indigenous knowledge systems demonstrate the innate human capacity to interpret complex environmental data through narrative. This approach mirrors the essence of data storytelling, where combining qualitative insights with quantitative data improves our depth and breadth of knowledge.
Modern methods have improved upon how this knowledge is gathered, disseminated, and related, yet the way in which different cultures approach knowledge dissemination needs to be considered. For many this has traditionally been qualitative in practice.
Understanding the integration of diverse ways of thinking prepares us to confront the challenge of quantitative storytelling, where traditional practices and modern needs intersect.
The challenge of quantitative storytelling
Traditionally, qualitative thinking (ie. stories) would be used and approximations applied in order to achieve results (ie. the number of fish caught under a certain set of conditions).
This knowledge would be passed down to the next generation of fishermen. This is how much of Indigenous knowledge was maintained over time — often in lieu of a written language.
This tradition underscores a natural human inclination to attach narratives to values — a skill that lacks sufficient application in contemporary education, leading to data visualizations that lack storytelling elements.
Many of us are innately wired to attach stories with values — yet many schooling systems do not combine these. Hence, many visualizations of data do not include narratives— and they should.
Edward Tufte, a pioneer in data visualization, in his classic book The Visual Display of Quantitative Information states that “words and pictures belong together”. He also notes that the separation of the two is the end result of a long-standing feud between artists and writers for “scarce space”.
Explanations that give access to the richness of the data make graphics more attractive to the viewer.
Complexity made accessible
Viewers often need the assistance that words provide. And conversely, many viewers need graphics to explain words and numbers.
Data storytelling provides a bridge between these domains. For those of us who think effectively with visualizations, a chart or graph may clarify.
For those of us who are unable to interpret a visualization, text integration in the form of captions and annotations can fill in the gaps.
Here’s a simple example.
Global trends in CO? emissions per capita over time:
Global warming is a hot-topic issue and is scientifically connected to the increases in global CO? emissions per person over time.
By visualizing these trends on a line graph, it’s easier to see the overall direction in which global emissions are moving —clearly upward (since around 1830) over the years. How can we effectively tell this story to those who are qualitative thinkers?
The use of a line graph simplifies the interpretation of data trends across multiple years, highlighting periods of significant change or stability.
The annotations for two periods in time tell a story about the effects of early industrialization and the post World War II boom. The use of a line graph highlights periods of significant change
For a qualitative thinker, this type of chart is often referred to as a “hockey stick” chart as the dramatic increase shapes the chart to look like, well, a hockey stick.
This qualitative observation removes the quantitative abstraction and detail from the raw data, providing a more relevant context to those who need it.
By converting the data into a visualization, you can:
The annotations can then be used as effective storytelling tools to accentuate the pivotal points in the data.
Tools for finding stories in your data
There are a myriad of tools available to find stories in your data. Some of the most popular:
For technical users:
For non-technical users:
Afterthought
Data storytelling is about “making complexity accessible” (Tufte).
Start with your data. What story does it tell? How can you share this story to make others understand and care?
There are many tools available to help you find the stories in your data: Tableau,
Combining words and pictures together with narrative tell a deeper story to the viewer. To provide clarity, viewers may need the help that words provide, or conversely, the visual that helps clarify the words.
Keep in mind that the point of a visualization is to clarify data.
If the visual does not clarify, then it is an utter failure.
Signing off - Marco
Other stuff you may like: