Fail fast, fail cheap

One of the Freakonomics podcasts was called The Upside of Quitting; you can also listen to it. It is about the balance between the sunk cost (I put so much into this already, it is hard to abandon) and the opportunity cost (the cost of not doing something else). Their point is simple: the stigma of quitting skews our decisions; we hold on to our failures for too long. And there is also strength in persistence, no doubt. The problem is when we hold onto something just because we already poured too much money and effort into it, NOT because it still has a promise. The Vietnam War is one tragic example of the failure to fail fast.

I thinking about it while reviewing a list of projects I have started and abandoned, trying to discern good failures from bad ones. Here is an example of a better one: last semester, I thought we would do a service to many people if we accept paper copies of some assessment forms here, and have our work studies punch the numbers into our electronic databases. Some objected that this is going to mean too much work for us, but I figured faculty and cooperating teachers’ time is in the end much more valuable than the work-studies’ time. We were also trying to save on the cost of training many people and answering their technical support requests. The trouble, however, came from an unexpected direction. Because there is always a few days’ lag, it was very difficult to monitor compliance. The data could be in one of two electronic databases, or in a pile of unprocessed papers, or in transit somewhere. We just could not tell who completed and who did not, especially in the few critical days at the end of the semester. And yes, work studies and GA’s stopped working because of their own finals. So, OK, we failed, and are going back to all electronic submission. But we failed fast, relatively cheaply. On the balance sheet are negatives (a few people think we’re morons, because we change the process every semester; we lost some data; some instructors never completed the forms), and positives (we learned something from the experience, we have moved on to the next, and hopefully better system).

Bad failures happen when the sunk cost keeps increasing, because people invested their effort, their egos, and their identities into something that does not work. It still fails all the same, just too late, and too expensively. I wish we could be a little more tolerant to failure. The world is changing; it is a much more dynamic and fast-paced place than even 20 years ago. No need to be flaky and try a hundred different things, without following up on any of them. However, we should always have a few more initiatives going than we expect to complete. And most importantly, we need to be attentive and forgiving to each other’s and collective learning processes. Otherwise, people will develop an aversion to risk, and just do the same thing over and over again. And that would be a very boring place to work in.