Mini-Microsoft has a blog post on middle management at Middle Managers, Bureaucracy, and No Birds at Microsoft where he offers pointers to some counter arguments to his regular bashing of bureaucracy and middle management at Microsoft. The various linked posts and some of the comments do a good job of presenting an alternate perspective to the various complaints people make against bureaucracy and middle management.
Derek Denny-Brown, a friend of mine who just left Microsoft, has a blog post entitled The curse of Middle management where he writes
I had a long discussion with a friend of mine about Longhorn aka Windows Vista. He had just caught up with news and some of the recent interviews with Jim Allchin. He knew I had some involvement with the OS divisions, and was just generally curious for my perspective 2 weeks out of the company.
In my view, a lot of the problems at Microsoft stem from bad middle management. Microsoft has built up a whole ecology of managers, who are at least as concerned with their career as they are with making good decisions. I've interacted more than I like to admit. The effect is that upper management doesn't hear a clear story about what is really going on. I think the phrase I used was that they 'massage the message'. Combine that with long release cycles and lack of accountability falls out as an inevitability.
One of the reasons I left is because I just don't see any way out of that mess. I am humbled by MiniMicrosoft and his determination to be part of the solution.
I tend to agree with Derek about lack of accountability being a problem here. One thing I've noticed about bad middle managers is that (i) it is almost impossible to get them out of their positions and (ii) all it takes is one or two to seriously derail a project. I've personally been surprised at how many bad middle managers just keep on ticking at Microsoft. It seems it is a lot easier to see individual contributors or even VPs pushed out for incompetence than middle management (Dev Managers, General Managers, Product Unit Managers, Group Program Managers, etc).
It is also surprising how much damage, a well-placed yet broken middle management cog can be in the smooth functioning of the corprorate machine. I've lost count of the amount of times I've asked some member of a team that creates a much reviled project why they product sucked so much and the response would be "our test/dev/general manager didn't see fixing those problems as a high priority". As Derek states, it is even worse when you get the ones that present a false impression to upper management about how things are going.
Michael Brundage also covered some of this in his essay on Working at Microsoft which is excerpted below
Company Leadership
Bill Gates and Steve Ballmer get most of the press, but it's an open secret that all of the division heads (and their staff, and their staff) are top-notch. I'm (happily) oblivious to how that circle operates, so I can only judge them on their results.
Given that Microsoft's been convicted of monopolistic practices, it may shock you when I say that Microsoft's upper management strikes me as very ethical. They talk about ethical behavior all the time, and as far as I've seen, lead by example. Maybe I'm being naive, but I find Microsoft's upper management to be very trustworthy. They're also thinking very far ahead, and doing a good job getting the information they need to make solid decisions.
Microsoft's leaders are also very generous, and frequently encourage the rest of us to make charitable donations (both money and time) a priority. Giving is a large part of Microsoft's corporate culture.
It's refreshing to work at a company where you can trust that the upper echelon is smart, hardworking, and making right decisions. I don't have to worry that my general manager or vice-president will drive our division (or company) into the ground through incompetence or greed. Microsoft's no Enron or WorldCom.
Managers
In contrast, most of the middle management should be tossed.
Did I mention I've had six or seven managers in five years? I've only changed jobs twice the others were "churn" caused by reorganizations or managers otherwise being reassigned. In fact, in the month between when I was hired and when I started, the person who was going to be my manager (we'd already had several phone/email conversations) changed! It's seven if you count that, six if you don't.
None of these managers were as good as my best manager at NASA. Of the six-seven managers I've had, I'd relish working for (or with) only two of them again. Two were so awful that if they were hired into my current organization (even on another team), I'd quit on the spot. The other two-three were "nngh" -- no significant impact on my life one way or another. I'd love to think this is some kind of fluke, that I've just been unlucky, but many other Microsoft employees have shared similar experiences with me.
I think part of the problem is that Microsoft doesn't generally hire software developers for their people- or leadership-skills, but all dev leads were developers first. Part of the problem is also that (unlike some companies that promote incompetence) good leads are usually promoted into higher positions quickly, so the companies best managers rise to the top. Consequently, the lower ranks are filled with managers who either have no interest in advancing up the management chain (which is fine) or else are below-average in their management skills (which is not).
But it's more complex than this. At Microsoft, many managers still contribute as individuals (e.g., writing code) and are then judged on that performance (which is mostly objective) as much or more than they're judged on their leadership performance (which is mostly subjective). Because individual developers have so much freedom and responsibility, it's easy and typical to give individuals all the credit or blame for their performance, without regard to the manager's impact. Conversely, managers' performance often does not translate into tangible effects for their teams (other than the joy or misery of working for them). For example, I can still get a great review score even if my manager is terrible. I think these factors contribute to management skills being undervalued.
Microsoft also suffers from a phenomenon that I've seen at other companies. I describe this as the "personality cult," wherein one mid-level manager accumulates a handful of loyal "fans" and moves with them from project to project. Typically the manager gets hired into a new group, and (once established) starts bringing in the rest of his/her fanclub. Once one of these "cults" is entrenched, everyone else can either give up from frustration and transfer to another team, or else wait for the cult to eventually leave (and hope the team survives and isn't immediately invaded by another cult). I've seen as many as three cults operating simultaneously side-by-side within a single product group. Rarely, a sizeable revolt happens and the team kicks the cult out. Sometimes, the cult disintegrates (usually taking the team with it). Usually, the cult just moves on to the Next Big Thing, losing or gaining a few members at each transfer.
I think these "cults" are a direct result of Microsoft's review system, in which a mid-level manager has significant control over all the review scores within a 100+ person group (so it's in your best interest to get on his/her good side), and conversely needs only a fraction of that group's total support to succeed as a manager (so it's in his/her best interest to cultivate a loyal fanclub to provide that support). The cult gives the manager the appearance of broad support, and makes the few people who speak out against him/her look like sour grapes unrepresentative of a larger majority. After a string of successes, the manager is nearly invincible.
Fortunately, these managers are unlikely to move further up the ranks, due to the inherent deficiences in their characters (which are usually visible to upper management and enough to prevent their advancement, but not so severe as to warrant firing them).
These "personality cults" always negatively impact the group eventually (while they're there and/or when they leave), but counterintuitively sometimes these personality cults have a large positive initial effect. Many successful Microsoft products have come into existence only through the actions of such personality cults. Some of these products even survived after the personality cult left for the Next Big Thing.
I totally agree with Michael's analysis. Like Derek, I'm unsure as to how one would go about reversing this trend. However I definitely think the way we assess accountability of folks in [middle and executive] management needs an overhaul.