GUESSING IN POPULAR PROBLEM-SOLVING METHODS

Many businesses teach their people structured methods to help them solve problems. Structure can be very helpful in certain stages of the problem-solving method, adding rigor to defining the problem and finding a pattern of failure. These are important steps beyond simple guessing or brainstorming, and they are critical to quickly solving problems of fairly easy or moderate difficulty. Many direct the problem-solver to spend significant effort studying the problem in situ, which is a significant step in the right direction away from solution-guessing at a table, in a conference room, or behind a computer. Understanding the pattern of failure allows a problem-solver to quickly eliminate some of the root causes by testing them against the pattern of failure. This can shorten the list of guesses and accelerate progress on some moderately difficult problems.

Where most of these structured methods break down is that they ultimately resort to guessing to determine what root causes may be. While they can help you solve some moderate problems, you still depend on the hope that your guessed cause is on the list you developed. Hard problems are immune to them.

For example, consider a classic problem-solving methodology such as the PackCorp Scientific Approach, which was popular in the 1960s and was one of the first to introduce rigorous problem definitions. Its method has the following nine steps:

1. Pick a problem

2. Get knowledge

3. Organize knowledge

4. Refine knowledge

5. Digest

6. Produce ideas

7. Rework ideas

8. Put ideas to work

9. Repeat the process

Steps 2 through 5 are dedicated to studying a pattern of failure, which was a breakthrough in problem-solving. But step 6, “produce ideas,” depends on insight, inspiration, and brainstorming to determine potential root causes.

When you look at most popular problem-solving approaches, you’ll find that they devolve into structured-guessing at some point. Many have steps such as, “develop possible root causes” or “deduce probable causes.” Whenever we develop some list of possible root causes, we’re guessing, even if it’s structured guessing. Some of these guessing steps are disguised as “forming hypotheses” or other seemingly scientific approaches. Many of these methods are designed to focus on simple problems quickly, where one needs to just organize guesses—Five Whys is great for this. For hard problems, though, the likelihood that you’ll include the true root cause in the list of “possible root causes” that you guess is tiny.

For sufficiently complex systems, it’s inconceivable that one or a group of human minds could comprehend it in order to effectively guess the right root cause. The Fault Tree Analysis for Boeing’s 747, which lists known potential causes of catastrophic in-flight failure, has thousands of elements. In some in-flight failures, like TWA Flight 800, the root cause is not on the prebuilt FTA—there are just too many possibilities.

The structure that comes with some of these methods can accelerate problem-solving for easy and moderate problems by pointing them in the right direction. To solve truly hard problems, you’ll need to use a method that doesn’t involve guessing in any step. There are methods that avoid guessing, but they are rare. You should find one you like. The one I’m most familiar with can be found in Chapter 10, “How to Choose Your Method,” along with some guidance on how to pick the method that’s right for you.