[friendly_deer]

I note that there two pervading ethea on HN, both exemplified in the parent comment:

1) Poor management (broadly defined) creates misaligned incentives, busy work, etc. that "gets in the way" of good talent executing well.

2) Management, who "doesn't really do anything," is overpaid relative to the IC's who "Do the work."

These two ideas, though, are in contradiction. Management deserves higher pay exactly because poor management can destroy value so quickly and easily. The problem is identifying, within the "company system," good management from bad management.

[opportune]

There are a lot of other factors at play:

A lot of less experienced employees don’t understand what managers do. There is also a certain personality I’d call “the geeky engineer” who thinks the same way.

Some things that are actually signs of an effective manager are hated by employees. Some individual employees for example need micromanagement or frequent manager involvement to make sure they don’t go off track, get stuck without telling anybody, or blow a time budget on something unimportant (eg a task is scoped at 2 weeks and they spend the entire first week playing with something speculative that may help).

A lot of managers are both highly paid and truly bad. I think a lot of businesses don’t do as good of a job vetting EM hires as they do SWEs the EM will manage. And I don’t mean that the EM needs to be Staff-level necessarily.

Even a good manager is incentivized to do things that are globally suboptimal (empire building, overextending) for career growth. This is also true of ICs though.

Personally, I think the biggest problem is the EM hiring bar though. Where I work has high standards and is very selective when recruiting ICs, but it seems like all that goes out the window for external EMs. I do think EMs deserve high pay and to be incentivized to do good work, but that requires a hiring process that better distinguishes between good and bad EMs.

[chw9e]

If your employee is blowing their time budget they are most likely either:

a) a bad employee who you should not try to micro manage to make effective but should just replace with a better employee

or

b) unaware of the context around the task and does not have aligned motivations with others to get this done quickly

If b - then a good manager should do a better job of figuring out how to develop the intrinsic motivation of the employee and find ways to align the incentives and motivations of management and employees better.

Trying to micro-manage to avoid employees doing things you don't want is putting a band-aid on the problem. In fact it's even worse because you're creating a bad work environment that will attract more sub-par employees who accept micro-management and will scare away the high-performing employees you so desperately desire. Eventually you have a bunch of micro-managing managers who themselves are being micro-managed and your company bloats to several multiples in size of what is actually required and efficiency is spiraling out of control.

[opportune]

There is not necessarily a dichotomy between good and bad employees, and I’m certainly not saying every engineer should be micromanaged, but IME certain engineers require it to be able to perform well.

I think it should be applied on a case by case basis when necessary, and I’ve seen effective managers do this in the past. Anyway, the type of employee who requires this treatment often is precisely the one that least enjoys it, which is my original point.

[phkahler]

How about C - someone who sees a lot more at hand than a specific problem being worked and goes on tangents fixing issues management isn't aware of or doesn't understand. It's career limiting to operate this way, but is long-term better for the company.

[esafak]

These people deserve a better company, and should seek one. Or start their own.

[kitsunesoba]

> A lot of managers are both highly paid and truly bad. I think a lot of businesses don’t do as good of a job vetting EM hires as they do SWEs the EM will manage.

From my vantage point as an engineer it also seems like what qualifies as "good management" is hard to pin down because not only is it difficult to quantify in the first place, but it also varies between teams and organizations… a management style that worked in one team and company might fly like a rock elsewhere. If that's true, it'd make vetting managers challenging even if the business in question is trying to vet them as well as they do engineers.

[OkayPhysicist]

The problem, IMO, is the muddying of the managerial role, such that they now do 2 separate roles, one of which is overpaid, the other not. The computer all but killed the secretary, and the remaining (computer-assisted) secretarial duties got split between the ICs and management. Because you can only dump so much secretarial work onto your ICs before they stop being able to produce anything, the managerial level got a bulk of the secretarial work, which in turn prompted more hiring at the managerial level. On top of that, it represented a flip-flop in power dichotomy, because the managers still have their managerial duties/powers: the people doing a lot of the secretarial work now have the power to fire the people doing the product work. Combined with people's self-interest in protecting and building their own positions, this leads to secretarial work bloat, and thus even more managerial bloat, and an increase in secretarial workload on the ICs.

My pitch for a the ideal corporate structure: bring back the secretaries. Re-introduce the division between your secretarial work and your managerial work, and shrink the number of managers by a significant fraction. Ideally, the result is A) fewer managers means more budget for hiring better mangers B) less secretarial work among your ICs and managers boosting both productivity and job satisfaction C) flatter management hierarchy means less information loss between ICs and the top decision makers and D) the people doing secretarial work no longer have the power to shunt that work off onto the ICs.

[esafak]

I'm curious about this. What kind of secretarial work do you have in mind, for a typical tech company? What fraction of time do you estimate it takes? Meeting scheduling is automated today. What else?

[peteradio]

Document management and user access control are two I can see.

[inetknght]

> 1) Poor management (broadly defined) creates misaligned incentives, busy work, etc. that "gets in the way" of good talent executing well.

> 2) Management, who "doesn't really do anything," is overpaid relative to the IC's who "Do the work."

> These two ideas, though, are in contradiction

I don't see them in contradiction. They're also neither mutually inclusive nor mutually exclusive. They are separate observations and both may be valid or invalid for a given set of management.

I happen to think that there's a lot of good "management" but not a lot of good executive teams. Sure there's definitely a lot of bad management too, but IMO there's a far bigger pie taken from poor (or outright employee-hostile) C- or S-level executives.

[wootland]

I said that there is no "talent scarcity" and anyone claiming that there is can count themselves among the rank of bad management. If management can't attract and retain talent, it's by definition bad management.

[mistrial9]

no, you are missing the usary aspect of market forces, where getting more talent giving fewer privileges for $PAY is the stated objective. People who practice skilled work are blind to business motives to abuse skilled work specifically, because "why would you do that." The commoditization of replaceable "skilled" work roles, with increasing oversight and metrics, is absolutely a factor in high tech.

[Temporary_31337]

I think the only solution is more managers to manage the existing managers