But that whole strategy is based on hiring "good people," and by what metric do you determine that? It seems that it's just pushing the measurements to another location, one that's less visible and transparent (the hiring process). Or maybe you don't think that there should be a firm set of metrics at all, but having metrics that are too loose ends up meaning the decisions tend to be arbitrary and based on the whims of whoever is currently making the decisions.
I think most managers would enjoy hiring the best people and then having greatness occur. But the reason these things are in place is because you're not always going to be hiring the best people (or someone might be great in some categories, but terrible in others you didn't know). Saying "hire good people" isn't much of a solution.
Measuring the hiring process is a different kettle of fish, to measuring on the job.
I think being overly measured in a job makes a job about maximizing the metrics at any cost. And a lot of people will cave into that pressure and take shortcuts in areas that are not measured in order to maximise their measured performance.
For example over estimating, less testing, less documentation if the metrics is about meeting estimates.
Whereas without the metrics they might just do what they think is best given their experience in any given situation.
There is never going to be zero metrics, as managers need to know where the money and time is being spent, but I prefer things on the lighter side.
I think a metric is an objective, formally and specifically defined measurement. So get people to do their research and use subjective measurements, i.e. personal judgement. It may be a bad option but it is an option, and we don't know whether it'll be bad or not until we test it.
There doesn't need to be a metric for that, it's the manager's job to use their judgement to assess that by building a relationship with the person, seeing how well they communicate their ideas (a sign someone understands things), that they finish what they said they would do etc etc.
As described here, "communication ability" and "timeliness" are not metrics -- they are just things that someone wants. They are no more a metric than "tastiness". In the absence of a well-defined measurement procedure and the intent to carry it out, one can not really be said to have a metric.
It's not exactly circular, since there are many less managers than low level workers, and progressively less people as you go higher. Presumably the people at the top were once successful managers or successful workers.
Basically, this promotes a system where metrics are given by how your peers and managers feel about you.
Perhaps while our metrics aren't perfectly capable of predicting how good a worker will turn out or judging it's progress, we shouldn't rely too much on them in place of peer judgement.
But indeed let's not fool ourselves that coworker judgement is perfect. It needlessly introduces a whole body of biases, unnecessary politics, and social engineering that can be as bad if not worse than a metric-only system, which at least are objective.
I think most managers would enjoy hiring the best people and then having greatness occur. But the reason these things are in place is because you're not always going to be hiring the best people (or someone might be great in some categories, but terrible in others you didn't know). Saying "hire good people" isn't much of a solution.