Isn’t that most places? It kind of makes sense, because actual productivity isn’t necessarily the easiest to measure (although I think it comes across just in attitude.. how much they care about he project.)
Not really: this is just the streetlight bias in action. A high-precision estimate of a useless metric isn't better than a lower-precision estimate of a relevant metric.
I've only worked at places well-run enough (at least in that regard) that people are competent enough at their jobs to measure overall productivity: it doesn't generalize to an automatic rubric like "# of commits" or "lines of code", but it is doable.