Lessons from NYT on Data Dump Research
I often look to excellent journalism as a model for how researchers should turn data into stories: Do investigative work, interview the right people, analyze the details, then lay out a meaningful and contextualized narrative about the who, what, when, where and why. In fact, when I talk about our research and analysis approach that turns data into stories, think journalistic stories, not fairy tales or novels with characters, plotlines, tension, and resolution.
That’s why I’m so disappointed to see this trend in New York Times’ journalism: The Data Dump. Instead of writing stories, they edit and publish brief interview transcripts. One section I recently read contained nine stories—two of which were interview transcripts. These are not regular interview columns, like The Corner Office or others. They are lazy, utterly uninteresting data dumps that reflect an urgent need to get into print with no time and staff to do it right.
The problem is, who wants to read them? I’ve tried, but these “stories” are as uninteresting as reading your data tabs. They are as uninteresting as reading your focus group transcripts. They may have all the makings for a good story, but nobody bothered to find it, extract it, and tell it. This is the very same problem that plagues so much marketing research.
Here’s how to do it better: Write and revise (and revise again) until you can communicate the story in the data without reference to any of the data at all. No tables, no charts, no quotes, no interview transcripts. Then weave the data back into the story to prove that the story is true. In short, don’t write about the data; write about what the data prove.
It’s possible that The New York Times is just experimenting with alternative formats; if so, I desperately hope this one disappears before they do themselves in by adopting the failed strategies of, say, market research. In the meantime, every time I come across one of these articles, I appreciate the reminder of how crucially important it is for research professionals to tell a story, and to use data only to support that story. Doing anything less may well put your research on the same downward trajectory that I worry our journalistic brethren are on.