Once you’ve gathered sufficient data, you’ll need to conduct a root cause analysis (RCA).

The purpose of a RCA is to identify and articulate what you think are the root cause(s) of the QLR issue you identified in the data. You may have several root causes, though you need not address all of them in order to address the issue you've identified. Some root causes are "high impact" root causes. Alternatively, there may be some clusters of "smaller" root causes that if addressed together may yield better results than tackling a "large" root cause that demands a lot of resources.

Determining which root causes to address through a CIP project requires some discernment and judgment. You should frequently turn back to the data to check assumptions and make sure you're addressing the right things. There are several ways to conduct a root cause analysis. The CBCC uses tools like fishbone diagrams and cause maps to conduct root cause analyses, but you are free to choose another tool if you find it more useful.

The key is to conduct a RCA that is empirically rigorous and rooted in data, not “hunches,” ”gut feelings” or anecdotes.