So, how are most managers evaluated? By the amount of "managing" they do, as measured by the amount of process they impose on their team. Evaluating a manager by the amount of managing they do is exactly the same thing as evaluating a programmer by the amount of code they write. And it produces results like you describe, where the manager works to produce evidence of their management in the form of processes and decisions from the top down, rather than facilitating the work actually being done.
In a simplistic world, the answer would be to change the incentives and the behaviour would change itself. But as they say, "correlation does not equal causation." The incentives have to change, but so do the people. Results-oriented managers don't work in those kind of environments to begin with, and after a year or two in such a place they will already have left. You need to change the incentives and the culture and the people all together.
Middle managers are rewarded for making budget and meeting release dates. Seems good, right?
So they agitate for (wait for it) maximum budget and minimum feature set. So that success is assured and their metric is optimized.
Unfortunately those things are exactly contrary to company goals. Why does this happen?
{opinion} Middle managers are too remote from either customers (financial goals) or top management (company goals). They're in the middle, right? With layers between them and either end.
And when you try to optimize any process with too many degrees of freedom, you have too many variables and get to choose which ones to look at. So middle managers look at their own career and ignore the rest.
My suggestion: line up all middle managers in the parking lot (important), stand at the end of the line, and put one bullet through all of them (optimizes cost in bullets).
If my company Ever has middle managers, its time to call it quits.
I think it's not the presence of middle managers; but how they are evaluated and incentivized. I don't envy their jobs; the upper managers have entire teams with clear tasks; the engineers on the other side have stuff to do; the middle managers are, well, caught in the middle. So they try to make themselves relevant by injecting themselves into various processes; by blocking things to make sure that everyone knows that they are present; by taking credit wherever possible.
So eventually it's the fault of the upper management, if they can't come up with the right incentive scheme to keep things moving smoothly.
I hate middle-managers too with a passion; but having seen them operate, I can't blame them for doing what they do. They're just playing the game by the rules. Blame the one who made up the rules.
Peter Drucker has claimed the lack of Middle Managers to be responsible for the early declines of Ford and Edison Electric Company. Their presence itself isn't a failure, it's their role and management.
It is funny that you mention architects because, my job title where I work is "Architect".
I totally agree with you in that it is the people that is the most important of all things. Tools are just tools and processes are just processes, it is the people when given right tools and shown right processes that make all the difference.
You're right. Ideally, people should be rewarded for the actual results they get, not for the amount of work they appear to have put in it.
A guy who would on the surface do very little yet gets excellent results, should be appreciated more than somebody who appears to do a lot of work yet has no end results to brag about.
So, how are most managers evaluated? By the amount of "managing" they do, as measured by the amount of process they impose on their team. Evaluating a manager by the amount of managing they do is exactly the same thing as evaluating a programmer by the amount of code they write. And it produces results like you describe, where the manager works to produce evidence of their management in the form of processes and decisions from the top down, rather than facilitating the work actually being done.
In a simplistic world, the answer would be to change the incentives and the behaviour would change itself. But as they say, "correlation does not equal causation." The incentives have to change, but so do the people. Results-oriented managers don't work in those kind of environments to begin with, and after a year or two in such a place they will already have left. You need to change the incentives and the culture and the people all together.