I just finished Ian Ayres Super Crunchers: Why Thinking-by-Numbers Is the New Way to Be Smart. It’s very pop / light, along the lines of Freakonomics, but still worth a read for the reminder that (unlike pretty much everyone I interact with in the Web software world) surprisingly few folks are aware of the extent to which statistical analysis now powers business decision-making. In particular, I found the description of statistical applications in the government and medical spaces fascinating – I had no idea that randomized trial had become the default mode by which new human services public policy choices are being made.
The thesis of the book is a logical evolution of the behavioral insights pioneered by folks like Cialdini; it’s well-established that human decision-making relies heavily on heuristics and that these analytical shortcuts and biases often lead to bad decisions. It comes as no surprise to anyone who shops at Amazon or uses Google that computers are really amazingly good at coming up with the right answer to a broad range of questions. Ayres just documents the fact that this kind of statistical analysis of large and diverse datasets is winning out over human judgment across an ever-broader spectrum of human endeavor.
The core assertion of the book is that statistical techniques have already replaced many traditionally human functions (with broadly excellent results) and that broad swaths of high-status and/or highly-compensated human “expertise” roles (e.g., medical doctors, teachers, and many different flavors of content analysts) are likely to suffer the same fate. He’s careful not to overstep into machine-chauvinism, so much so that I was a little surprised not to find a mention of Nassim Taleb’s arguments from The Black Swan (and elsewhere), that over-reliance on models can lead to even more disastrous results than over-reliance on intuition, due to the enormous potential impact of rare but catastrophic outcomes.
If you’re a software developer or an experienced online marketer you probably wont find much new in Super Crunchers, but if you’re one of the above and are trying to sell your offering in to clients or industries that still place intuition over analysis (e.g., most traditional media / advertising businesses, A&R in music, etc.) you might want to drop this on your clients desks as a warm-up to your next meeting…