Our goal was to deliver clustering examples with results that are repeatable with good performance. This is a tough balancing act, and it's important to keep. Quality and repeatability are must haves for trusting the results while good performance encourages experimentation with different filters, levels of detail, input variables, and so on, which is the path to more insights.

In the next chapter we will see how to an unsupervised learning technique requires a different approach than the ones you have seen previously.