Not only is it graded on a curve but to prevent ratings inflation there are quotas for each ratings bucket. Many deny this, even when presented with concrete proof that such quotas exist (but were called "performance targets"). This itself would be fine except that negative ratings have a target percentage of ~10%. This includes regrettable losses and ratings below Meets All Expectations.
Such targets are enforced at a reasonably high level, typically at director or higher over an org of 100-150+ employees.
Why is that bad? Because it creates a hugely toxic environment. Why? Because such directors and VPs will demand their managers find sufficient "low" performers to meet their targets.
The other aspect to this is that all of these ratings are essentially bullshit anyway. Why? Because it's a popularity contest. If your manager and director likes you, you'll pretty much always get a good rating. If they don't, you won't. The exact same set of facts can be used to argue ratings pretty much +/- 1 bucket, possibly more.
Stack ranking and related schemes are so obviously going to be gamed, will definitely measure something other than the nominal thing, and will gum up the works. It's one thing to have a Hunger-Games-style hiring process. Bad as that is, there's an expectation that once you've passed those trials—once you're hired—that you can then devote your time to building things or leading a team that builds things.
But when it extends into quarterly or annual reviews, where there must be a loser, then of course it will serve as a distraction from accomplishing actual goals and doing useful work. Now you have to spend time managing your image, making sure you don't piss off the wrong peer, and choosing assignments that are trendy.
Companies try to avoid all of that by adding even more complexity to the review process. More layers to prevent favoritism, more concrete metrics to sidestep subjectivity, a longer list of metrics to mitigate any one of them being over-weighted.
This just means more time and attention is stolen from productive work, and directed instead toward image management and politics.
The goal of corporate employees usually is not maximizing profit. If it were, things like private equity, LBOs, and shareholder lawsuits would not be necessary. Rather, the goal of organization-man is maintenance of the hierarchy. Why? Because people who aren't interested in this goal either quit and take jobs outside of the corporate world (like consultant, therapist, small business owner, etc.) or they circulate freely throughout many different companies and rely on up-leveling at the interview stage to get raises and promotions.
With this goal, all the stack ranking, annual reviews, politics and image maintenance makes sense. For there to be a hierarchy, there must be people at the bottom of the hierarchy. There must also be a pathway to the top of the hierarchy, lest people tire of the game and quit. Therefore, people seeking to preserve the hierarchy will create processes to ensure some people are defined as needing improvement, and others are defined as being worthy of promotion and upward mobility.
The folks who are purely interested in maximizing profit usually enter finance and become hedge fund managers, where they have a small staff (largely dispensable, other than themselves) but outsize influence on markets, and can keep 20% of their profits (100% if trading their own capital) for themselves.
At the end of the day even upper management doesn't matter. This is about Owners and Workers and everything in between is just fluff. The $300k - $400k these engineers are making is a rounding error. Turning middle management into cannibals so that one day they might reach slightly higher management is exactly what is supposed to happen, because it means they are willing to kill for the owners.
So many people lose sight of the bigger picture, workers are always cogs in the machine. Luxurious, comfortable cogs, sure. But its crabs in a bucket.
Intelligence isn't a spectrum, it's an infinite-dimensional space. Someone can be smart on one axis and a dunce on infinity-minus-one other axes. Every genius ever born was almost certainly mediocre at nearly everything outside their expertise.
And even today such people exist. Encountering them is often quite an experience. Seeing them being better than you at almost everything is on one hand really nice, on the other hand it makes you wonder what you did wrong. Such folks are usually quite rare though, 1 in a thousand types of people. But I'd imagine Google should be full of them. Or maybe it isn't, idk, given GP.
Without any examples of his work it’s hard to believe that they are of any consequence. I don’t think anyone can just write a piece of music, but you don’t present a strong case that anything he wrote is even remotely worthwhile.
Hell, Bruce Willis has a band that’s released records. I just don’t see any reason to think they’re worth listening to over anyone else.
Form Wikipedia:
> an individual's performance on one type of cognitive task tends to be comparable to that person's performance on other kinds of cognitive tasks.
We assume that the people implementing stack ranking have good incentives.
At one of my jobs, my manager blurted out what we have all suspected about Scrum. It is valuable to generate a neat numerical progress report for the guy above. It has no other value.
My manager doesn't give a shit about achieving anything either. He is doing the same gaming we are.
None of this is true for creative workers in actual teams with deep technical or site knowledge. Someone with a decade of experience that is a bit burnt out and underperforming may still have amazing contributions albeit less frequently or predictably. Someone that doesn't commit code often but "holds the team together" by lending a helping hand and giving sage advice to juniors can be invaluable. Fire them and the juniors will flounder. New developers not only don't bring customers with them, but likely won't be productive for months. And so on, and so forth.
Stack ranking originates in organisations where sales are in charge at the upper levels. They think that what applies to them applies to everybody else at the organisation.
This is a mistake, of course, but it's fantastically difficult to overcome "cultural" biases like this. Even otherwise very smart people have a blind spot to this kind of thing. After all, the fish is the last to know it swims in water.
Not to mention that decisions on whether or not to stay in education are often taken when fairly young, and even very clever people can arrive in their 20s with incorrect views about the world based on what their parents/teachers/etc believed - if you've spent your childhood being told that a degree is the only way to have a secure future, you have to not only be smarter than needing a degree you also have to be bold enough to think you already know better than everyone around you (which sure, some people are that confident as 12 year olds, some never get that confident even as adults).
TLDR, I get why business degrees get a bad rep, and agree that many idiots do do them; but also, a couple of the cleverest people I've ever known have done them, mid-career, because the job they were aiming for (and in both cases I'm thinking of, then went on to achieve) had hard requirements attached and they considered it worth wasting time on an MBA to progress.
I didn't realize it was this bad. This is awful.
First, it’s important to understand that managers are held to outcomes. If a manager retains low performers because “they’re friends”, over time they’ll find themselves with a low performing team and it will be visible in their results. At that stage they are now at the bottom of their manager’s stack-ranking process. Retain and promote your low-performing friends at your peril.
Of course, if you believe your entire management chain to be incompetent, perhaps those managers will never be held to account. But if you hold that view there’s no reason to believe anything is functional, let alone promotions, and I’d explore other employment opportunities if you can.
Respectfully, I disagree.
First, let me point out that when I say "popularity" I don't mean friends necessarily in the nepotism sense (in response to your later point). "Popularity" is a reductive term but still a relatively accurate one. What's really going on here is the intangible factors lead other around you to interpret the same set of facts charitably or not almost arbitrarily.
A big factor here is similarity in terms of demographic factors and background. This so often gets wrapped up "culture fit", which itself can be a euphemism for more problematic biases.
The example I like to give is this: a person may work on a project for a half that never ships. It may not even produce an MVP. Given this set of facts you will people who will say "we learned a lot and will take those learnings into future projects" all the way to "this team failed to deliver a prototype and as such we canned the project".
What's driving this are social factors.
Managers are subject to the same social forces that drive this interpretation. A given manager may get credit for dealing with reports that aren't necessarily bad enough to be exited but would otherwise be a headache for someone else in the org. Or that manager may be blamed for their team being perceived as low-performing even though they're given projects doomed to fail (eg under-resourced).
It's also worth noting that with any form of stack ranking or performance quotas, another factor that comes into play is horse-trading. That is, different managers are competing for a limited pool of ratings. You will find that a manager has decided someone should be promoted and that report will find themselves high on stack ranking as a result. There's an implicit quid pro quo here because other managers can get your support (as a manager) when one of their reports gets put up for promotion.
> Of course, if you believe your entire management chain to be incompetent
It's not competence. It's biases masquerading as objective determinations and an implicit over-reliance on social cues and factors all while managers have incentives that may come at the expense of those of their reports.
How could this be fixed? If true, what would we expect to see, versus its not being true?
“there’s no reason to believe anything is functional” should be “you don’t believe in your manager.” Good managers divorce moral edict from feelings. Good managers know they don’t get to define what functional means.
Sure you can buy ad$... but are you prepared to buy the 30% fake impressions to fake people we require?
The fact that both systems / strategies were developed by the same group of people is for sure the "why" behind these cuts.
Got a good chuckle out of this.
My experience with calibration at multiple companies is it's where you realize not only does your review depend on how happy you make your manager, but how politically savvy your manager is coupled with how much you were able to impress the people on the calibration committee.
It also gives your manager a nice out if you work hard, but they don't quite like you. All they have to do is not fight for you in calibration and you'll never be able to get much beyond a "meets expectations" rating.
It was more honest and straightforward than what silicon valley and Big Tech pretends to do. It is all ranking in some ways. Everything is. If you remove the facade, fundamentally you have a limited resource and you want to reward people that make the company successful. I don't see anything wrong with that. Doing anything else is wrong and unfair IMO.
I didn't say that and I didn't imply that. Anyways, rewarding (positive reinforcement) is mutually exclusive to punishment (negative reinforcement). Rewarding should be encouraged, otherwise, I have zero motivation to do better. There are many other companies that would value me more and I would not stay there.
Extreme meritocracy is idealistic, and I understand that cannot be achieved easily and there is always politics/networking that plays a role. But it does exist and I've been a beneficiary of it.
Take the manager's word for it and go all in on nepotism?
What's the better solution?
Note: 'in a standard way across teams'. You have jumped directly to 'can't be measured at all in any circumstances'
How to grade that into any kind of ranking, in a standard way, is seemingly not likely given how subjective the difficulty of a code base is. This also completely elides the clever and best engineers that find ways to simplify and/or avoid solving problems entirely. Further, a team that built a crap pile of bad code, but know it pretty well could seem really productive just because they have built for themselves some job security. Then compound all this with measuring the utility of what was produced.
Also, any investors out there might hear “one time restructuring charge” in the quarterly results when they talk P&L which will affect the bottom line.
Geez, I mean there are decent managers who do rate on performance and not just buddy-buddy stuff. Frankly I'd expect to be trusted with my reviews of people under me.
I saved my team a couple of times over but pissed off a manager in the process. My performance review had multiple contradictions praising me and admonishing me for the same things (being a good mentor but not spending any time onboarding new team members, etc). I left rather than deal with the gaslighting that was all but certain to ramp up.
Graded in the context of peers’ performance does not imply that they sort employees by performance.
But well, stack ranking is not only about sorting (the name is a lie), it's about placing people in fixed sized cohorts. The real question is if there is a limitation on the number of raises or a minimum on the number of people treated as "low performant".
You do get to see the average size of cohorts, and you will need to explain why your team doesn’t meet them - for example if the entire team project is going exceedingly well, or the team is really made out if a lot of extraordinarily talented people - then you can argue why you have more than the average rate of excellent rankings.
Anyway, even if it honestly for "extraordinarily talented" people, that's still stupid stack-ranking.
The same you’d rank a person’s work, you would rank a team or org’s work.
If that team is doing really well, it’s likely they will have a distribution of higher rankings than the average. If they’re doing really bad, it can be the opposite.
The point is that if you finished ranking some team and your rankings are higher than average, you need to have a reason why.
What makes a ranking into a stack ranking?
The curve determines rating and compensation, influences promotion decisions, projects people get to work on and what opportunities come their way. The seems like a sort to me given that budget and project opportunities are fixed in many places.
EDIT: It sounds like at least some people are claiming that Meta has an internal policy that X% of your engineers under you are supposed to be given "Meets Most Expectations" ratings, which would also be pretty close to stack ranking (if at least allowing for some management leniency) https://news.ycombinator.com/item?id=32930370
With large enough populations you see fairly consistent statistical information across time. Little fluctuations even out. So at org sizes of, say, 500 people you'll see pretty consistent ratings distributions between cycles. This is where you get the "we have a roughly expected ratings distribution." If you've deviated far from this then it means either that your org is training people on the evaluation process incorrectly or there is some other cause of a weird distribution. When something doesn't match, people simply investigate for an alternative solution.
What people seem to be frightened of is when somebody says to a manager of 10 people "hey you've got too many high ratings, drop some." Google and Meta are massive companies so I suspect that this happens somewhere but I've never even got the scent of this happening from my position at Google. I've had cycles where 100% of my team was in the top 20% of ratings without anybody raising an eyebrow.
On stack ranking & the penalty for being in a great team - yes, also happened, being in a great team but you still had to rank someone lower, better to be in a worse team to get higher ratings. It had obviously negative impact, but there was some combination or organizational pride that had been cultivated, plus we thought we'd all be millionaires anyway so what did it hurt. But I bet some people had to leave who were the unfortunate victim of these affects.
A team of 20 people may have no low performers. An org of 400 probably has a couple, at some point the law of large numbers does really start to apply, and you're no longer in competition with the people you're working with on the daily.
If one member of your two-pizza team is going to get fired each year, it makes sense to throw someone under the bus on the reg. If one member of your 100 person org is going to get fired each year, that strategy doesn't make sense.
I tried for you, remember me when your next job is hiring managers.
Ideally you'd do this before calibration so the rating doesn't mess it up.
They are still super atheletes, but they are judged against other NBA teammates, not average professional basketballer in the world.
I wouldn't want to work somewhere at FB scale that doesn't grade you on the curve of other engineers in your department and level, once your departments are large enough that's exactly how it should work.
What's problematic is forcefully defining X percent as being low performing, the comparison to peers is a misused kernel of benefit.
BTW "meets most" likely doesn't sound too bad to non-Metamates. Internally it's a pretty bad mark on your record. It doesn't directly relate to being fired but a ton of people quit when they receive it. I view it as sort of a "pre-PIP" state.
(* Managers/directors will say it's only a suggestion but nobody really believes that. I'm sure management themselves are frowned upon for not having enough MM under them.)
The kind of newspeak I've heard from multiple friends and former coworkers that ended up at FB/Meta at some point. Well, the ones that didn't fall into the cult of Mark.
It might be because I'm getting older but this is the kind of stuff that makes me want to puke. The co-optation of language to exude an aura of friendliness is absolutely disgusting.
MM is a warning—it means you need to improve or you'll likely be put on a PIP.
I despise this, the dissonance between a positivity-sounding "meets most" with the actual meaning being "pretty bad" is what I expect from the North Korean regime, or Russia. It's a disgusting type of newspeak/corporatespeak which, admittedly irrationally, boils my blood.
The idea is they consider themselves to have a very high bar (which I think is largely accurate). They say you “meet most expectations” which means you hit the marks on all but 1 or 2 criteria. It’s not that you’re not good at your job, it’s that that’s not good enough for them.
Of course anyone receiving a MM is having a very shitty day despite that. They could use more direct language but I’m not sure it would have any practical difference.
It doesn't matter how much friendly-sounding bullshit you slather on if that's what's happening.
relevant to the overarching topic: https://dilbert.com/strip/2021-05-24
Hmmm this doesn’t sound like the Facebook I worked at
EDIT: sorry I see you said "worked at". So I guess you won't be able to verify.
In this thread someone mentions it exactly how I remember the doc describing it: https://www.teamblind.com/post/Companies-With-Stacked-Rankin...
The curve is defined by the organization. They can carefully examine it, refine it, backtrack how it's performed... or you can have someone clueless hear that so and so company fires Y percent, so let's tweak the curve until Y percent are at the "extremities"
The multi-axis performance review criteria that requires you to show deliverables on things like your impact on team-building or culture. This means that when you might normally get a couple week low-output time between projects, you are usually looking for something to fill in the non-coding axis, like organizing a team virtual offsite event or searching for a way to improve the codebase.
Organizing team activities and code quality initiatives are a good thing, but the explcitivity in which you need to show deliverable impact on all axis is something I found unique to Meta and very exhausting.
Where it really digs is when you are 1.5 weeks into writing an in depth project plan, meeting with other engineers, coordinating cross teams, researching, reviewing code etc. you better have a good explanation why your delivery over time graph has a big flat spot.
That's 95% of what I do in my company. I guess by Meta's standards I'm not working?
I know architects that spend most of their time writing design docs. Does that not get measured as a deliverable? That’s literally their job
That definitely sounds like pushing you to be “productive” as much as your body allows without burning out.
https://www.theatlantic.com/technology/archive/2018/03/the-b...
I also suspect NIH is a big part of the problem. Bidets and washlets weren't invented in America or adopted there first, so Americans can't adopt them now since that would be admitting they weren't first for something. Americans are, as a whole, completely unable to look at what other countries are doing and learn from their example.
there are good (usually small) teams doing interesting and professionally fulfilling work that's not really common elsewhere. (things that come due to "hyperscale". a-a-and, unfortunately, in this regard all big corps are alike. so if you want to do something like that, then your best bet is a corp like FB. though of course there are smaller niche companies for everything, so just because of some niche don't rush to big co-s.)
I guess they are hiring at some BUs and firing at others?
The options seem to be:
* Cut risky or underperforming products or teams and focus company resources on the products and teams that are doing well.
* Or cut staff across the board, leaving all teams struggling to succeed with fewer resources.
The first option seems wiser. "More wood behind fewer arrows", as the saying goes.
Seems contradictory to many modern interview process opinions that low performers introduce a net negative effect.
* losing the low performer in this context means you lose a head count. If you keep the low performer during hard times, you can hope to get them backfilled when the economic environment gets better * there are often more work to do than you have resources for: if as an EM, you have to deal with PM who refuse to adapt to capacity, you can simply assign the low perfomers to the projects you don't care about. * etc.
once you reach a certain size, those issues happen in pretty much any organization in my experience.
Even at the worst companies I’ve been at I have never seen anyone rewarded for behaving this way.
The unintended side effect is that it punishes risk takers and destroys innovation culture.