Tools for Thought

Failure is usually approached with one of two ways. It’s either not an option, or it’s permitted from time to time as the price of learning.

In the first approach, no one really believes that a person can avoid making mistakes, but the idea is that adopting infallibility as a working mental set raises the bar, and at least reduces errors that might otherwise be made with lower standards.

The second approach reflects the humanist ideal of trying your best. Examples abound of famous people making embarrassing mistakes, only to go on to great achievements. We all have setbacks, so there’s no point in dwelling on them. Mistakes should be treated as learning experiences.

I think the latter approach is more realistic, but not radical enough.

Shifting from a mindset of failure to failure rate

A more reliable path to reaching a goal is to assume failure, or rather, to assume a failure rate. If it’s obvious that mistakes are inevitable, they need to be factored into the planning process so that the number of efforts and the length of time necessary to achieve the goal can be reassessed accordingly.

Consider a man who sees a beautiful woman and considers asking her for a date. He has second thoughts, wondering what he has to offer that could possibly interest a woman so wonderful. Another man treats approaching women as a numbers game. He assumes that most women he asks will turn him down so his natural reaction is to approach more woman to account for longer odds. Which man is more likely to have a date by the end of the day?

Not long ago I hit a slump in my writing. I keep a long list of article ideas in reserve, so if I ever run into a situation where I can’t think of anything new to write, I just refer to the list. Only this time there was a problem: everything on the list was either an idea I had outgrown, lost interest in, or require access to resources that would have taken more time to summon than was practical.

After falling into despair for a couple of hours, I stopped and ran through alternative definitions of the problem (an APC). Among them was one that resonated: I didn’t have enough ideas. Which is to say that even though I had “a lot” of ideas, I fell into the trap of conflating a lot with enough.

Then I remembered one of my favorite quotes from George Bernard Shaw: “When I was a young man I observed that nine out of ten things I did were failures. I didn’t want to be a failure, so I did ten times the work.”

So I spent the next 10 minutes drafting 17 article ideas to add to the 32 “reserve” one that had been sitting on a list for weeks. Now when I’m stuck for an idea, I sit and do a round of zero-base brainstorming, writing at least 10 new article ideas without referring to a previous list.

Failure to define an acceptable failure rate in advance leads to crippling perfectionism. The failure rate will vary from person to person, and from project to project, based on available skills, resources and the return on effort. For me, a 90% failure rate for coining article ideas is totally acceptable, but a 90% failure rate for articles would be totally unacceptable. If I’m reasonably satisfied with 80% of my blog posts, I consider the remaining 20% a learning tax. By assuming a margin of bad posts in advance, I don’t surprise myself when I’m not happy with all of my output, so disappointment doesn’t turn into proud self-flagellation.

For Edison, a 99.99% failure rate was acceptable for discovering a practical filament for the light bulb, since the return on effort would be boundless. Once the exploration was reduced to a single variable, running through substitutions was a simple a numbers game.

Deterministic and stochastic control

When someone turns on a light by flipping a switch, she has deterministic control over its activation. The odds are theoretically 100% that the light will come on. When she drives to work in the morning, she has stochastic control over whether or not she’ll arrive on time. “Stochastic” refers to the measure of randomness in the system. Traffic and weather conditions are more complex variables than the integrity of a filament. Nevertheless, the driver still has a reasonable degree of influence of whether or not she arrives on time by simply leaving earlier.

Luck is always a factor in poker. But the difference between a novice and a championship poker player is that the latter has studied the odds of failure for each scenario, and developed strategies, responses and guidelines for dealing with them. Building a base of sufficient information allows a person to act without complete information.

Many goals are comprised of complex variables, so we often have to compensate by doing more homework to find the variables with the most leverage. When planning a complex project (anything from starting a blog to finishing grad school), consider the worst case scenario. What factors would contribute to the project’s failure? What would put it behind schedule? List these factors, then for each of them determine how they might be eliminated or mitigated.

Technorati Tags: Thinking Operations, Personal Development