I really like the five whys to get to the truth of what went wrong, but unfortunately, in practice you only take the whys so far.
For instance, where I work, if we have a large enough on-call issue, we have an RCA to figure out what went wrong. We use the five whys there to figure out what we could’ve done better. However, we usually only really go as far as the engineer’s actions that caused the issue. We usually don’t ask, why did the engineer choose to act in that manner instead of the correct one? Usually the takeaway is “We need a process to make sure engineer does right thing” and not “We need to make sure engineers don’t have too much work so that they have time and space to do the right thing.” I find it’s usually the latter is a better course of action but it’s hard to make those statements or requests, or if you do make those requests, management isn’t as open to taking action.
For instance, where I work, if we have a large enough on-call issue, we have an RCA to figure out what went wrong. We use the five whys there to figure out what we could’ve done better. However, we usually only really go as far as the engineer’s actions that caused the issue. We usually don’t ask, why did the engineer choose to act in that manner instead of the correct one? Usually the takeaway is “We need a process to make sure engineer does right thing” and not “We need to make sure engineers don’t have too much work so that they have time and space to do the right thing.” I find it’s usually the latter is a better course of action but it’s hard to make those statements or requests, or if you do make those requests, management isn’t as open to taking action.