The success of 99% failure

When I ran the research laboratory for our company, I used to say (somewhat facetiously) that 99% of everything we did ended failure. And yet, I considered our efforts a great success. How can this be?

Failure is a word that has very negative connotations in our vocabulary. But if we look at what happens when we fail, we can see a different side to the story. First, all failure is not the same. Failure differs in magnitude, consequence, and information gained. If you bet a million dollars on a stock and the company goes bankrupt the next day – that is a failure of great magnitude (for most of us). However, if you only invested a dollar, you can probably live with the loss.  Either way, you learned something about the stock and the market. As inventors and researchers, we are always probing into the unknown. How will we know something if we don’t try it? Well, again, there is the question of magnitude and consequence. Jumping off a high cliff to test the theory of survivability after impact would constitute a high magnitude of risk and a great consequence of failure. You would certainly learn the answer, but at what cost? However, if you gauge the magnitude of the risk, the consequence of the failure, and what you might learn from the trial; failure can be transformed from a negative to a positive. Sometimes finding out what doesn’t work is as important as finding out what does. Probing to confirm direction is something that inventors and researchers do all the time. This not only puts theory to the test, but also provides new information to reformulate the theory. The key is to take controlled risks – risks that won’t “break the bank” or anything else of great value. The goal is to exceed what you lose in time, money, material, etc. with what you gain in knowledge.

As I said at the beginning of this post, we were extremely successful and yet experienced a “99% failure rate.” What made us successful was the ability to gauge risk as we probed into the unknown, and to learn valuable things that we could then apply to our next efforts. The failures were small but informative; the successes built on those trials were substantial.