Hacker News Reader

Given that lousy managers would cause more people to quit, which we can reasonably suppose, wouldn't a system like this favor the lousiest managers? After a while, you'd expect the organization to end up with the worst possible managers, and a steady employee churn rate?

As a selection process it sounds pretty much the worst one imaginary. However, these organizations seems to not only survive but prosper. There must be more to it than this.

My theory is that they’re super successful in one area, so it probably doesn’t matter at all what the rest of the company is doing. Just one major successful project can float thousands of failed teams, due to the vast sums of money involved. I’m also positive the “golden goose” teams aren’t subject to the same kind of scrutiny and have more autonomy so they’re not subject to these kinds of policies

See my other comment in this thread. Regretted attrition rates will indeed penalize managers, and weed out the lousiest managers. The article is only describing unregretted attrition rates.

That’s compatible with the “everyone cries at their desk here” reports from a few years ago

> manager's performance review based on whether or not they fired enough people to match a company-wide expected attrition rate is completely bonkers

Amazon is a company which gained a lot from their policy of exploiting the employee. Amazon loves how they can hire mostly subpar candidates and still make profit by driving them with fear. The same is reflected in their mandatory URA policies. Its not a big deal that some great people get fired too. They don't expect such people to stick around for long in the first place. Policies to manage developers and warehouse workers have come from the same person, so you can't expect much difference in philosophy of how to treat employees.

On Blind's Amazon private lounge, there's a popular saying, don't make noise until Amazon starts shooting it's own employees. Anything less is either old or unsurprising.

> Managers should possibly be penalized if they fire (or lose) too many people (it shows that either ... or that their people hate them and leave)

There is a nuance here, which is that Amazon is only targeting unregretted attrition rates. Which indicates the number of people leaving Amazon, who have been previously marked as poor performers. The system also doesn't allow simple gaming, such as marking someone as a poor performer after they tell you they are leaving. Having a high regretted attrition rate is indeed something managers will be penalized for.

Not defending the system in any way. Just wanted to clarify the above

> system doesn't allow marking someone as a poor performer after they tell you they are leaving

Haven't seen this being done firsthand. But I've heard from trusted sources, that it's a common practice to mark people leaving on their own as URA.

Tracking is fine, the issue is targeting.

It's pretty well established that whatever you measure will affect people's decisions even without explicit targeting.

It can be important to refrain from continually tracking that which you don't want to manage.

More generally, it is very hard to turn a useful, indirect metric into a useful target, because of perverse, unforeseen incentives.

It is the correlation/causation problem, often times the metric is not causal to your success.

Presumably they've identified that under performance is spread uniformly amongst managers with a large enough number of indirect reports. Presumably the target URA is arrived at via other independent measurements. Presumably they control for negative feedback loops like hiring people just to preserve your team by stratifying the URA target, or subjecting new hires to other measures too. Etc.

Its cut-throat no doubt, but if its also well rewarded, why shouldn't there be a place for this kind of practice?

> why shouldn't there be a place for this kind of practice

Here's an article that explains why - https://www.inc.com/jason-aten/amazons-controversial-hire-to...

Where does it explain it? There's a lot of words in this op-ed with almost zero actual substance. It's a pure play on emotions, without even the effort to talk to past or present employees potentially affected. Just more lazy "journalism".

Taking the article at face value (a risky move when journalists get warmed up against business) - Amazon is effectively lying - they are claiming that they have work for someone to do and that they think the person can do it.

In reality - the hiring manager doesn't have work for the person and doesn't care about their qualifications. So Amazon is misrepresenting the situation to their hires. People make big life decisions based on their employment - doing that is very unethical and disruptive.

If you want the conversation to be about legality then that is one thing. It seems like a practice that should be legal to me. But it is objectionable.

The OP article, the referenced Insider article, and most of the comments here are all pure speculation... based on a single memo. The Insider quotes are laughable, themselves speculative "it could happen" comments rather than "it did happen" examples. "Some say that leads to stack ranking" isn't actual valid criticism or journalism, it's a comment on a possibility.

It fuels the anti-Amazon rhetoric and so everyone just runs with it, assuming every worst case scenario is true and extrapolating every rumor to the extreme to validate their disdain for the company. It's exhausting.

Strangely the vaunted leadership principles of Amazon dont include one for ethics

That’s a lot of presumptions. It may be as simple as noticing that teams with a higher attrition rate are more productive on average and faithfully treating that observation as a causal link.

It may be successful in the short run, but that doesn’t make it a good long term strategy. It may be a good long term strategy but that doesn’t make it ethical.

To the thought police: notice how I engaged the parent in a thoughtful argument instead of moronically hitting the downvote button.

I tried to make my assumptions clear. I suppose I'm saying that Amazon's massive success should give us reason to think about its HR processes in a constructive light. I feel that many posts in this thread take the view that Amazon management is stupid and/or short-sighted. That stack ranking is somehow flawed at conception and will doubtless fail. It builds a strawman out of one of the worlds biggest employers which is a bit silly.

I guess this is a political issue for many. If you view this discussion through the prism of fundamental rights, fairness and so on then its bound to cause a down-voting orgy. If ever HN was a place one could discuss alternatives with political implications, it no longer seems to be.

>> I suppose I'm saying that Amazon's massive success should give us reason to think about its HR processes in a constructive light.

Business success does not justify employee exploitation.

Amazon can afford to make mistakes because it’s cash cow has already been bred, so I wouldn’t presume that their macro success is evidence that their micro strategies are also successful. It’s more likely that the micro failures can become invisible.

But playing the devil’s advocate, one could make the case that having a quick fire culture keeps employees on their toes and maybe keeps wages suppressed. Maybe that’s worth the cost of hiring poor performers?