It's worse than that. A lot of these management tools don't even measure outputs, only inputs, and then only partially and with doubtful predictive value.
From the viewpoint of outcomes, everything beyond outputs is simply too obscured to make sense of.
I don't think that's always true but you definitely have a valid point in a lot of cases. For inputs I was thinking about stuff like measuring time spent on specific activities without reference to whether the employee had been set up to make good use of that time or whether those activities were themselves worthwhile. That would contrast with outputs like measuring a software developer's productivity by the classic lines of code metric or some modern version like number of commits per day. A developer working 80 hours/week and producing 1000 lines in 20 commits in that time will look great by those metrics but I'd rather hire the developer who would write only the 100 useful lines that would survive into production in 20 hours and 5 commits because they thought before they wrote.
Nearly every performance tool I’ve seen only measures outputs. Now, outputs are critical to outcomes - but they’re not the goal themselves.