Why Data Needs Narrative (and vice versa)
Why data needs storytelling and vice versa.
In some circles, the "storytelling" part of "data storytelling" automatically means you're trying to pull the wool over the eyes of your audience.
I see this pushback from people all over the place. And for a reasonably good reason. You hear story after story in the popular press about how a storyteller has stretched the truth to the point of breaking, all in service of a good "story". Fool me once, shame on you. Fool me dozens of times, you can't out-story-tell the storyteller. My own father seems to rankle a bit when I talk about building a narrative of a presentation I've been asked to deliver.
As the tried (and tired?) and true trope goes, "anecdotes are not evidence".
Post by @mattyglesiasView on Threads
But the extreme other side of this issue is the "The Data Speaks for Itself" camp. How many times have we seen brilliant and nuanced scientific research be ignored because one (or both) extreme side hired the better storytellers? The tobacco lobby, Swift-Kelce conspiracy theories, literally every single 'health food' trend...
The data can speak for itself when the data collects itself and cleans itself. Until then we as data scientists, analysts, engineers and architects have to speak on behalf of the data.
And to speak on behalf of the data, we have to give it a fighting chance of being listened to, understood, and remembered - we have to tell its story.
Cognitive Load & the Demand of Data
Understanding data is an effortful process. The brain needs to interpret, compare, and derive meaning from what might be a deluge of numbers. Every data point, every number, every percentage demands cognitive resources for interpretation.
(See my post on Cognitive Load Theory to understand this idea using one popular framework.)
When we're faced with raw data, especially extensive analyses without any guiding structure, it's analogous to being handed numerous jigsaw puzzle pieces, each from different puzzles. The cognitive effort required to even begin making sense of these pieces, let alone fitting them together into a coherent picture, is immense.
Without some form of categorization, summarization, or guiding narrative, we're left to grapple with these disparate pieces, each vying for our limited cognitive attention. The challenge here isn't simply about managing the volume of data, but about the very nature of how humans process information.
The Necessity of Narrative
Cognitively, humans are hardwired to understand and remember stories. This inclination traces back to our evolutionary history. Early humans, sitting around campfires, shared tales of hunting escapades, encounters with other tribes, or the mysteries of the stars. These stories weren't just for entertainment; they were critical for survival. They served to transmit knowledge, culture, and values from one generation to the next.
Narrative as Structure
From a cognitive science perspective, stories help in organizing and making sense of information in a structured manner. When information is structured, it becomes more digestible, memorable, and impactful. Consider the difference between reading a dry list of facts versus engaging with a compelling narrative that weaves those facts into a coherent and emotionally resonant tale.
Narrative as Context
Complex ideas often require a rich tapestry of background knowledge to fully understand. Narratives are instrumental in providing this context. By embedding details within a story, the audience can understand the why, how, and what of a situation, rather than just the isolated facts.
Narrative as Access
A well-crafted narrative has multiple layers and can be approached from different angles, making it accessible to a diverse audience. Whether a listener is captivated by the human element, the overarching plot, or the intricate details, there's something in the story for everyone. This inclusivity is essential when communicating complex ideas to audiences with varied backgrounds and interests.
Furthermore, rather than being passive recipients of information, narratives encourage the audience to engage actively. As listeners or readers, we predict outcomes, empathize, and draw conclusions. This active participation aids in comprehension and retention.
Narrative as Memory
At the heart of a good story is a central message or theme. While the surrounding details provide depth and richness, they all serve to support and amplify this core idea. This structure is incredibly useful when dealing with multifaceted subjects. By framing the complex idea within a narrative structure, the communicator can emphasize the most crucial aspects without getting lost in the minutiae.
Additionally, emotions are powerful drivers of memory. Studies have shown that emotionally charged events are remembered better than neutral events. Narratives, by their very nature, have an emotional component – they involve characters with desires, challenges, triumphs, and failures. When an audience is emotionally invested, they're more likely to remember - and act on - the core message. 1, 2
You (& Your C-Suite) as Characters in the Story
Yes, your presentation or document has characters. You as an analyst are a character in the story of the analysis. So are your users or customers, and so are the decision-makers you're talking to. You could even consider your data a character. What else is the Introduction-Materials-Methods-Results-Discussion-Conclusion format of a scientific article if not the same as a Joseph Campbell arc of narrative?3 As humans, we (usually) want all of these groups to come out victorious in the end.
Narrative isn't merely a luxury or an add-on; it's foundational to how we, as humans, understand and make sense of the world around us.
The Ethical Imperative of Curated Presentation
Beyond the cognitive and communicative challenges posed by raw data, there's an overarching ethical dimension to consider.
To claim that "data speaks for itself" is, in many ways, an abdication of responsibility.
This isn't about distorting data or infusing it with bias. It's about recognizing that data can be misinterpreted, misunderstood, or even manipulated. By curating its presentation, by providing the necessary context, narrative, and visualization, we're ensuring its integrity. We're ensuring that the decisions made on the back of this data are informed, accurate, and ethical.
But how does one curate data effectively without falling into the trap of over-simplification or bias?
This is where interdisciplinary knowledge proves invaluable. A grounding in cognitive science aids in the clear communication of complex points. Foundations in neural sciences guide with the understanding of how different visual stimuli are processed.4 A deep understanding of the domain ensures that the essence of the data is maintained.
Such a fusion of knowledge bridges the gap between raw data and truthful, actionable insights.
Still, there are challenges. As Davenport and Patil highlighted in their seminal 2012 Harvard Business Review article on the rise of the data scientist, the nexus of expertise required for effective data communication is rare.5, 6 Not everyone has the luxury of extensive training in both domain expertise and cognitive processing. Thus, collaboration becomes key. Teams with varied expertise can come together to ensure data is not just accurate but also ethically and effectively presented.9
Concluding Thoughts
The belief that data can speak for itself is a seductive one.
It offers a veneer of objectivity, suggesting that data is immune to the vagaries of human interpretation. Pure data, devoid of context and narrative, can be silent or even distort the truth.
But also, one shouldn't have story just for the sake of story. After all, anecdotes are not evidence.
Think of data as the raw material, and storytelling as the craft -- turning that material into something tangible, relatable, and actionable. As experts in data interpretation and communication, we are not just number-crunchers but storytellers. We breathe life into numbers, ensuring that the message they carry isn't just heard but deeply understood and felt.
In our mission to communicate data-derived insights, it's crucial to remember the cognitive science behind information processing and the power of a well-told tale. By weaving our data within the fabric of a compelling story, using potent visualization techniques, and maintaining unwavering ethical standards, we ensure that our data doesn't merely convey facts. Instead, it builds insights that can captivate, inspire, and most importantly, catalyze real-world action. When data and story work together, we can ensure that our data doesn't just speak – it sinks in, it resonates, and most importantly, drives action.
Footnotes & Citations
1. Damasio, A. R. (1994). Descartes’ error: emotion, reason, and the human brain. New York: Putnam.
2. Greenhalgh, T., & Hurwitz, B. (1999). Why study narrative? BMJ, 318(7175), 48-50
3. Olson, R. (2015). Houston, we have a narrative: why science needs story. Chicago, The University of Chicago Press
4. Paivio, A. (1986). Mental Representations: A Dual Coding Approach. *Oxford University Press*
5. Davenport, T. H., & Patil, D. J. (2012). Data Scientist: The sexiest job of the 21st century. *Harvard Business Review*, 90(5), 70-76
6. Davenport, T. H., & Patil, D. J. (2022). Is Data Scientist Still the Sexiest Job of the 21st Century? *Harvard Business Review*, https://hbr.org/2022/07/is-data-scientist-still-the-sexiest-job-of-the-21st-centur
