Little Bets, Big Breakthroughs


I woke up a few weeks ago to an email from Amazon.com. The company was announcing the launch of Amazon Vehicles, a new "car research destination and automotive community." I have no clue if Amazon Vehicles is better than its competitors, but I had to shake my head in amazement: they just never stop experimenting.

Really, that shouldn't come as a surprise to anyone. In his 1997 letter to Amazon shareholders, CEO Jeff Bezos laid out one of the company's guiding principles: "We will make bold rather than timid investment decisions where we see a sufficient probability of gaining market leadership advantages. Some of these investments will pay off, others will not, and we will have learned another valuable lesson in either case."

Surprise! This isn't a post about companies, really. This is a post about individuals. Bezos and Amazon are featured prominently in a Peter Sims book called "Little Bets." But the book's lessons appear to apply as much to regular folks as they do to corporations. I recently read a terrific book by Dilbert cartoonist Scott Adams called "How to Fail at Almost Everything and Still Win Big." The hint is in the book's title: Adams is a serial experimenter in business and his personal life, and credits his success to continuous trial and error. Similarly, self-help writers like James Altucher and Tim Ferriss espouse a philosophy of experimentation. Ferriss even goes so far as to call himself the "human guinea pig." I think we've become accustomed to the idea that successful people demonstrate more grit, i.e. try harder, than unsuccessful ones. But it seems like they also try more often.

A few caveats are in order if you, like me, are interested in experimenting more. First, some amount of rigour is necessary. You can't do a "true experiment", as defined by research purists, on yourself since you can't create a control group. Instead, we have to approach these quasi-experiments with intellectual honesty, and be open to the possibility that we're fooling ourselves sometimes.

Second, good experiments have to have low costs if you want to do a lot of them. But "unknown unknowns" exist. For example, a nutrition experiment could appear to have benefits in the short term but be detrimental over a longer period. By definition, there is risk inherent in experiments, and some kind of rigour is again required to determine when an experiment is proving too costly.

Finally, the hardest thing about making little bets is the probability of failure. Again, Bezos articulates this brilliantly:

Creating what Gerd Gigerenzer calls an "error culture" is hard. Like most people, I'm terrified of looking silly. But developing a personal error culture and spirit of experimentation is a little bet I'm increasingly willing to make - one failure at a time.