There is an extensive interview with Donald Wheeler in Quality Digest. He wrote Understanding Variation and many other excellent books to help people understand what they are seeing - and not seeing - in their data.
I usually enjoy what he has to say, and look back to Understanding Variation often when I am playing in data. The interview went in several directions, and I particularly enjoyed the thread about how tools can often confuse the situation:
The ability to do a computation does not guarantee that the result will be meaningful.
Just because you have the software and computational capability doesn't mean you need to use every bell and whistle. The goal is to get something that makes sense and is explainable to everyone. If you get a "perfect" answer but know one can understand it, you are in the same boat as having no answer. This comes up in the business world often - one size does not fit all.
[Photo: "minim++ - Tool's Life" by José Luis de Vicente]