I worked many years for Motorola; the last couple of which were spent looking at availability and reliability. (Reliability engineering seeks to decrease the frequency with which a system fails; availability engineering seeks to decrease the time to repair/replace a failed system.) Motorola then had a culture of "if you can't measure it, you can't improve it". I therefore began by looking at historical outage data for our systems. One important piece of data we had in the database was "root cause". Understanding root cause is perhaps the single most important bit of information one needs to improve any system. As I analyzed that data, going through the engineering logs of what was done to fix the system, I found that the recorded root cause was essentially unrelated to reality. Apparently the system required the person taking the call to record a root cause, but the engineer who actually did the repair never went back to update that field. ...
This is a paraphrase of the pasuk in t'hillim 84:7 -- "mei'chayil el chayil" -- which means "from strength to strength". In this case, it is my thoughts and ideas to those who are strong enough to be interested :)