One of those outcomes is paralysis due to endless analysis.
The dynamics are simple. Every test or every additional piece of data has the potential to resolve a finite percentage of the remaining uncertainty.
Suppose each "test" removes 1/3 of the uncertainty. At first blush it might appear that three tests would remove all uncertainty...but that is not how it works.
The first test removes 33% of the uncertainty but leaves 67% of the uncertainty.
The second test removes 1/3 of the 67% or 22% or the original uncertainty, leaving about 45% of the absolute uncertainty.
The third tests removes 1/3 of the 45% of the remaining uncertainty leaving 30% of the absolute uncertainty.
Removal of 70% of the uncertainty might be plenty, especially if the decision is inexpensive and less-favorable outcomes are not really all that bad.
It is easy to lose sight of the ABSOLUTE numbers when you are swimming along and making decisions by the seat-of-your-pants. Every test LOOKS like a bargain because every test removes 33% of the remaining risk.
We see what is in front of our face
Momentum keeps up focused on the problem we were dealing with last week and blinds us to other risks, other costs that are emerging.
Consider a person who is dating and cannot commit because...well, there is always SOMETHING they don't know about the other person.*
The risk is that the other person will get tired of the lack of commitment and look elsewhere. Or he/she will get to an age where they will not want (or cannot have) children. Or perhaps YOU get old-and-ugly or become ill or lose your job and lose your value in the dating market.
In product development, extra time taken up by the design phase is usually at the expense of the manufacturing tune-in phase. The absolute risk is not eliminated by the product development engineers ironing out every fussy, little detail. Rather, risk is transferred to the manufacturing side of the house, and in absolute (or global) terms the risk starts to grow exponentially.
Summary
- Make a plan
- Ensure that the "tests" comprehend a wide swath of operating conditions. In the realm of dating, that means visiting a wide range of venues and activities. It means seeing them when they are tired and/or stressed and/or when they disagree with you....or combinations of the above.
- Have a clear picture in your head regarding your tolerance for absolute risk and then pencil out how many "tests" are required to beat the uncertainty below that level. If you just need a pretty girl to take to the company Christmas party...they are easy to find. If you are quirky and unwilling to compromise about several value-like things, you might be looking for one-in-a-hundred or one-in-a-thousand.
- Don't let anxiety and tunnel-vision blind you to emerging risks. While you are looking for that one-in-a-million, you have been discarding dozens of one-in-a-hundred girls who would have slam-dunked more than 80% of your requirements and made you very, very happy. And are you really going to live long enough to honestly evaluate a million girls under a wide range of venues and activities?
- Getting plenty-good-enough quickly means you have a generous amount of time to tweak and adjust afterward. As long as the plenty-good-enough choice has flexibility and excess capacity so failed trials are not catastrophic.
*This example was chosen because it is one that most people can relate to. For those who read too much into what I write...Mrs ERJ and I are doing just fine. Thanks for asking.
I find the biggest impediment is inertia. The static kind. Once a groove is worn, it is very hard to jump tracks and get out of it, due to it's own inherent inertia. It takes twice as much energy to achieve the same delta prior to wearing in that groove.
ReplyDeleteWhen my then girlfriend now wife and I decided to build a house on property she inherited from her parents, I told her building the house would be a great test to measure how committed we were to each other. We self contracted the house to save $$$, hiring the subcontractors individually and cleaned up the daily refuse. We also applied the individual finishes to the house and bought furniture for it. All of these steps took decisions and compromises.
ReplyDeleteAfter several months of this, my girlfriend agreed - if we stuck to each other during this process, it was a good test. So far, this has proven out - 29 years in early March, two now fully grown children. a dissertation completed and the loss of three parents, and retirement of one of us. Been a full life.
Decision making is something I have struggled with for years, ERJ. Thanks for such a clear and concise description of an applicable process.
ReplyDeleteHaving been involved in more than one product development project that transferred into manufacturing, I cannot underline enough how much easily a well thought out and well defined project makes the actual implementation so much easier and greatly increases the chances of success.
My grandpa alwys said that having a plan was a great idea. It gave you some room to make a better, more refined plan when the real issue struck.
ReplyDeleteHaving an imperfect plan, is better than no plan at all.
Most people have NOTHING and are therefore victims. The rest of us are (usually) survivors.
No plan survives first contact with the enemy...or the problem. But having a plan is a leg up over having no plan most of the time.
ReplyDelete