A bedroom example in another discussion showed a tricky edge case: someone standing near the edge of a bed zone made the light switch on and off, even though the room was occupied. Before letting a zone turn a lamp off, I would log zone transitions alongside whole-room presence during normal use, look for brief bounces near furniture or doorways, and test whether a short delay or a whole-room-vacant condition avoids false offs.
How do you validate zone placement and timeouts before enabling a lighting rule? What observations tell you the rule is reliable enough for the people using the room?