Stories and Data

Last week I spent 3 days in storyland. I was in Chicago at a meeting of the Innovation Learning Network, focused on the use of stories in business and innovation. I listened to lectures on the psychology of stories; played games with a Hollywood script writer, a radio show editor, and an acting troupe; and spent a morning redesigning airport security procedures.

It got me thinking about the interaction between stories and data.

Ideally, stories and data have a symbiotic relationship. Stories add richness, context, and meaning to data while data adds richness, context, and proof points to stories. Used together well, they are a powerful combination for telling the truth about the past, present, and future. But if we get sloppy, they can undermine each other.

I think there are two ways that data supports stories:

1. Data helps us figure out which story to tell
2. Data helps us tell the story

These are two very different uses. Confusing one for the other can have tragic consequences.

Data helps figure out which story to tell
First, data helps us analyze reality to determine what's actually happening in the organic stories of real life. By collecting and analyzing data, we begin to piece together a narrative that represents what's going on--who is doing what? how often? how many? how consistently? etc. That's what data analysis is all about--good data helps me understand reality.

For example, if I want to tell the story of how someone uses my web site, I can look at data about user demographics, site traffic, user reactions, and user behaviors. This data enables me to construct a story that accurately represents something real. It may also enable me to create a vision for the future that is grounded in today's reality, or to describe a future that is demonstrably different from today's reality.

With a foundation of good data analysis, I can then create narrative stories that describe reality to my audience in ways that are accessible, compelling, and memorable.

Data helps us tell the story
Once I know which story to tell, I can use data to improve the impact of the story I'm telling. Data enhances the narrative, adding proof points to make it believable, adding context to make it understandable, and adding details to make it compelling.

The pitfall
Problems arise when I mix up the two uses of data. If I'm telling a story, and I grab some data to make my story more compelling, but I have not yet used data to make sure I'm telling the right story, then I'm in trouble. For example, let's look at data and stories about global climate change.

If I want to tell a story about the relationship between human activity and the climate, I need to start by examining the data in order to figure out what's real. If I'm paying any attention at all to science, I will conclude that the story to tell is one in which humans are driving the earth to the brink of climate catastrophe. I can then use data, both qualitative and quantitative, in the telling of that story. E.g., I can cite average temperatures rising, thinning ice sheets, oceanic acidification, and displacement of species. All of these will support my story. These data make my story more accessible, compelling, and memorable.

However, if I skip the first step (using data to determine what is real), then I may tell a story in which global warming is a crazy liberal plot, and I can still use data to support my story--it snowed in April this year; global temperatures historically rise and fall; etc.

I think the critical step is for a storyteller to be self-aware of which of these two roles data is playing. Am I using data to figure out which story represents reality, or am I using data to support the story I'm telling? Both are important, and both are powerful. But if I do one without the other, I either end up with a less compelling story, or I'm distorting the truth.