I have noticed over the years that most folks that are getting started using the Six Sigma methodology have as many problems with some of the early decision-making issues as they do handling some of the more quantitative tools. Specifically, which problems should they focus on? If one listens to the voice of the customer, then this might not be a big problem. If you have a problem, the customer will usually let you know. However, if one is trying to be proactive, and look for opportunities to reduce cost and improve performance, one needs to consider their Critical-To-Quality (CTQ) metrics. I have found that this is the problem for so many people. In large commodity-like industries, the output metrics are usually very well defined. In smaller, less defined, industries, it can be difficult to identify.
One must first understand that there are just a few areas in which we compete. They are Cost, Quality, Delivery, and Customer Service. All of these dimensions lead to Customer Satisfaction. Each of these four dimensions is driven by something. For example, in quality, it could be driven by functionality, dimensional characteristics, reliability, and appearance. The CTQ should then be defined by metrics that that are easily measurable. For example, if the driver is functionality, one has to demonstrate, usually through sampling that the part does what it is supposed to do. The metric could either be measured such as something like pressure or temperature, or it could be measured in something like proportion nonconforming.
The point is, until we learn how to measure our output in a simple and easy to understand way, we will always have something that is prone to be out-of-spec. and/or out-of-control, periodically.