The Data Within
A few weeks ago, I found myself coughing and shivering under a blanket as the latest fall virus settled in for an extended stay. Unable to do much else, I turned on the TV and started looking for a movie to watch. But everything the streaming services suggested just felt … off. I recognized myself and my interests reflected back at me in the suggestions: a documentary about the financial system; a movie about food; a show about travel in East Asia. But none of it clicked.
Then, suddenly, I knew what I wanted to watch: a movie I’d seen with my wife and son 15 years ago. Undemanding, funny, brimming with cozy associations. It was the cinematic equivalent of chicken soup. Almost any other day, I wouldn’t have been interested. But just then, it was exactly what I needed.
The data the streaming services have on my viewing habits is more than impressive. It is clean and complete. It captures almost everything I have watched over the last decade, with the exception of a couple of hours of viewing on flights or in hotel rooms. Normally, the algorithm serves up a menu of options that includes something that will satisfy me. And that’s the thing about algorithms: They are tuned to normality. They make predictions based on statistical likelihoods, past behavior, and expectations about the continuation of trends.
I don’t get ill very often. Faced with an outlier like that, the algorithm was useless. And that was an important reminder. I spend a lot of my time enabling organizations to make data-driven decisions. But sometimes, the past can’t steer you. Sometimes, probability is just a roll of the dice. Sometimes, you have to go with your gut.
Making decisions based on what feels right or on what your gut is telling you can be a........





















Toi Staff
Gideon Levy
Tarik Cyril Amar
Sabine Sterk
Stefano Lusa
Mort Laitner
Ellen Ginsberg Simon
Gilles Touboul
Mark Travers Ph.d