I heard in NPR today that Netflix fires people that are average and only keeps those that are exceptional. Does any one have any experience working in such a company? When I heard about netflix, it seemed like life would be very stressful if you always have to worry about getting fired. But on the plus side, you get to learn a lot and work with exceptional people, which is a great way to learn. ANy experiences or opinions?
In general, performance reviews are great in theory, but fall down in application. Theoretically, a review is a time to talk about the great work you did, what you learned, and what your goals are. Your manage can respond will honest feedback and a merit increase.
The reality is reviews are a part of a process, and your rank in them depends in part on the quota set by HR. Most of the time, HR wants employees to fit within a bell curve for rankings. In some extreme examples, people's reviews were based on a predetermined track and not on actual performance.
Netflix's policy is just HR marketing. If all you do is keep your 4 and 5, then what makes those people 4 and 5 will start to become 2 and 3. It's a moving goalpost, and sounds like an effort to increase productivity without increasing compensation.
If you fire all the 1's, 2's, and 3's, you have to hire people to fill in those positions. And in one year, the whole group gets re-evaluated, including the new hires. I'd think it would take quite a few cycles before the people who were 4's and 5's became 2's and 3's, while maintaining their same levels of productivity.
I think GE was well known for doing a similar practice under Jack Welsch and they still had plenty of long-term employees. I don't think the system is particularly desirable though.
Personally, I think the idea of turning "performance" into a number is just fundamentally flawed. Giving the same salary to an entire team is a better idea
I work in a law firm that will fire great attorneys. It's part of the big law business model. Up or out and very few get up'd into partnership. So eventually you get fired (after 8-12 years depending on firm.)
So I know eventually my ass will get fired. It's just a matter of when.
I don't really understand it. But it's how the industry works.
Edit: 8-12 is for unambiguously great attorneys. A lot of decent attorneys get shown the door after 4-6 years. People who can't hack it (usually because they won't put up with be worked like a slave) will only make it 2-4 years. A lot of people quit.
At 8-12 years do you get paid more than a new hire? Seems a "good" way to keep costs down, having the bulk of work done by the new guys - and a few partners to oversee everything.
The downside is the average experience of your employee is going to be a lot lower, but that must not be the top goal of the law firms that employee this.
From what I've seen of UK-based law firms, this is typical.
A young law-hopeful will face insane competition to get the relevant work experience and eventually get on a vacation scheme (2% success rate) [1]. Here they will do 2-4 weeks work with the firm they want to apply for.
Next, they need to apply again to receive a full training contract at that firm. Competition for these is even tougher (~0.5-1%).
Now, as part of the graduate intake (training contract) this person will be one ~100. Here are where well educated, nervous, fresh-faced young staff become the firm’s best workers - it's basically a two-year job interview. At the end of the two years, around 70 will be hired by the firm as new lawyers. From there another 20 will parachute[2] out per year. Of my brother's 35-person intake at a magic circle firm only two remain seven years on; many are in completely different fields. Around one person per intake will make it to the top - by this point they are the 1% of the 1% of the 1%. The entire business is structured to concentrate wealth in the hands of as few people as possible, with ludicrous hourly rates that junior staff never see, rigid seniority levels and the illusion of prestige.
Source: Brother at mid-sized magic circle firm + my own 3 months' subjective experience at two top-tier law firms before I decided to jump ship (I'm biased!).
[2] A popular term in the field referring to the ease in which a lawyer working at a prestigious firm can move in to something less prestigious and a whole-lot more rewarding somewhere else. I met people applying for top firms with the sole intention of doing this.
To your point below, I suspect the 8-10 year person is creating more value with less supervision than a new hire. Being able to manage a project/client solo, even a small one, makes your work incremental revenue to the firm. More so if you're involved in bringing in the business....
I have new hires; they take a LOT more supervision... and our group's output is limited to the speed at which our senior folks can originate projects...
Most lawyer work is bog standard paperwork, and clients pay by the hour. Extra years of experience is not useful. You need a few experienced people, but mostly for their sales and strategy skills.
Some types of legal work are bog standard paperwork, but the kind of stuff the big firms tends not to be. It takes about 5-6 years for a transactional person to be able to run a small deal by themselves. In litigation it's even steeper. At big firms, most associates (even 8 years in) probably couldn't run a complex case. And the older partners just have better judgement and wisdom that only comes with doing something many times.
But with stuff like writing wills, drafting simple contracts, entity formation, patent applications, etc. tend to have a shorter learning curve.
A number of my boss's reports had landed in the achieves category for all of 2014, but nonetheless received a communication from HR this past April that they were in the bottom 5% of the company for 2014 and at risk for termination if their performance did not improve. This surprised my boss and the reports.
Obviously if you're in the bottom 5% you're well below average. The question is how you can consistently be ranked in the middle category yet nonetheless be in the bottom 5%. That would seem to make the quarterly ratings a rather pointless exercise for communicating to employees what their ongoing performance is.
>Obviously if you're in the bottom 5% you're well below average.
This is not true.
If the bottom 50% are all very close in absolute performance, you get exactly the situation described, which is exactly the steady state of the process you desribe. The process is working too well.
Sounds like Marissa Mayer took Google's stack-ranking machinery and replicated, awful detail by awful detail. What a queynte.
At Google, the bottom categories are rarely used and "Meets Expectations" is a huge window (3.0-3.4) from the 3rd to 65th percentile. Since you only get your bucket, your manager could be saying that he's giving you a perfectly respectable 3.4, while actually raping you in the ass with a 3.0.
Often, you have managers who give the 3.0 and 3.1 to keep reports immobile (they can't transfer) while conveying that they are getting average or average-plus scores. But whenever there's a witch-hunt-I-mean-dishonest-layoff-I-mean-"Low Performer Initiative" those people given 3.0s get hit with PIPs, which is not what even their managers wanted.
We've banned this account. Calling Marissa Mayer a cunt is an egregious violation of the civility that HN requires. (Using an archaic spelling doesn't change that, it just makes it weird.) No one gets to do this here.
I told you just three days ago that if you wouldn't or couldn't stop doing this, I would ban your account [1]. This comes after more warnings, reminders, cajolings, and please-dont's than anyone has received in the history of HN by a long shot. It's painfully evident that you won't or can't stop, and it's time we applied the rules equally.
I worked in a company like that. Everyone was ranked against their peers, and if you were below ~40% more than two years in a row, you were "counseled out."
I didn't find it stressful. Partly because I was never in danger, and partly because I knew what I was getting into. But I think it's a bad practice, mainly because it just doesn't work. Empirically what I've observed is a bunch of people in the wrong role, not the wrong company.
It's expensive to fire people: companies could save a lot by making lateral moves easier, and doing better internal matchmaking. As in matching people <--> responsibilities, not like... dating :).
What is the normal tech company attrition? I would think 20-30% seems normalish, but I never had full HR docs of a fortune 500 company or anything to look at. Generally a bunch of people quit after 1 or 2 years - if not you stayed for 10-20. If most of the fired people were the going to quit anyways people, no big change. (But i assume at least some of the low performers are the 20 year vets as they are burned out)
The short answer is that we didn't find enough; we were perpetually understaffed. However: most of that was just because of industry-wide talent shortage in information security.
Generally, you fill in the void with people who recently left competitors (who have the same attrition rates).
Well, they miss a lot of higher quality experienced people (many with families) who just don't apply at Netflix because of this high turnover and risky experience.
I have a theory that the talent level at most companies has almost nothing to do with ridiculous HR policies, but rather two things:
1) Project/Field
2) Name brand recognition
The game industry has a lot of overqualified people on very poor salaries in poor conditions because games are sexy. Google has both sexy projects and name brand recognition, which allows them to recruit top level talent despite the fact that their interview process is absurd and frankly, pretty insulting to qualified professionals (can you imagine doctors or lawyers going through similar processes?). Reason being, having google on your resume opens a lot of doors, and there's a chance you get to work on something really cool.
So I don't know anyone that works at Netflix, but I would believe they have mostly above average engineers not because of this bizarre practice of firing "average" people, but because they have enough recognition to find above average people in the first place.
"average" with respect to the company, or the population of programmers as a whole? If normally distributed and sufficiently large, you'd be getting rid of half your company under the former. The latter seems like a tough thing to measure to begin with. I can't imagine either really being the case.
Though it does sound somewhat similar to the concept of "stack ranking", which a few companies (such as MS) are notable for having used as part of their annual review process in the past.
Stack ranking has always sounded to me like an absolutely poisonous thing to implement in an otherwise healthy office. But if the organization knows that it needs to implement a reduction in force regardless, then I guess it might make sense if management does not have a feel for who their best engineers are.
This hints at a pretty interesting analysis of whether this benefits companies.
If you fire anyone below "market average", you can reasonably expect to sustain an above-average workforce as long as you compensate well enough to retain decent employees.
If you fire anyone below "corporate average", the results are more interesting. You'll bring up the average quality of current employees, but have to backfill the missing employees.
If your hiring is just a normal distribution on average skill, the result is that you quickly reach diminishing returns. You fire the vast majority of your new hires every year, while slowly turning over a few of the old ones in favor of new superstars. If your hiring is better than average, though, you can reach far high equilibrium points. In effect, it makes sense to fire all the employees up to your hiring midpoint, trusting that you can efficiently raise your average quality by doing so. After that, you'll have to cycle through too many employees per position to see speedy improvement.
Of course, all of this is crippled by training delays and the problems you'll face when "we fired half the employees last year" gets around. Interesting theoretical model, though.
The advantage of stack ranking is that it's a formal process, and one in which managers' behavior is more measurable and reviews are definitively scheduled. It also aligns employee churn with performance to some degree, moving weaker hires out and encouraging better hires to stay, again via a formal process.
What it reduces is the tendency of less formal processes to give raises to those who ask and to stiff those who don't and to throw up the "your salary is confidential". All those things it avoids are known to create a high potential for bad company culture and moral.
Considering the quality of software and innovation produced by MS while stack ranking was in place, and the drop in software quality coming out of Google after its own version of stack ranking was introduced, it seems unlikely to be effective.
As I understand it individuals were stack ranked within teams, so even if a team outperformed the company or industry average, individuals were still tagged as underperformers.
And generally it became a matter of politics, popularity, and self-promotion, not objective competence - which is hard to measure anyway.
To me, it seems like a fast road to madness.
As for Netflix - I can't imagine any company needs to be staffed entirely by ninja rockstar code demi-gods. Many development projects are mundane and by the numbers, and basic competence is fine. If you want to be disruptive, hire a core of creative innovators who can code. They probably won't be ninjas, but for product development, talented customer-oriented innovators are a really good thing in any business that sells stuff to real customers.
Stack ranking is a solution to a larger problem, which is the issue of growth in overall "size of the tribe" and political networks that shade / obscure / fabricate performance views.
By the time you're stack ranking, you're deep into large company territory... dealing with the above issues.
Medical equivalent: People taking cancer drugs have a higher than average likelihood of dying. However, that's probably a function of having cancer vs. the performance of the drug.
I think you make a good point about how to define average. I am not sure what the npr people meant when they said average. I was assuming industry average. However, the questions then is - if the managers are used to seeing only high performance engineers, how do they gauge whether the engineer is really average performer. It seems like managers subjective views and their biases would matter a lot.
If you eliminate the "average" segment of your staff, then the "exceptional" team members become the new average, and you terminate the next average group, you eventually wind up with one - maybe two.
In practice, many firms eliminate the "average" section of staff and replace with fresh grads who aren't particularly likely to be better than "average" but are cheap.
Rinse and repeat often enough (and preferably with an incentive scheme that justifies average starting pay and ridiculous working hours with the promise of fantastic rewards if you manage not to get fired for the first three years) and bizarrely, you get a reputation for high standards.
I work in academia, where things move slowly. When lag is expected, turnover is expensive. Also, it being a university, my employer is much more concerned with educating us and broadening our skill sets, rather than discarding employees like a half-eaten slice of pizza that didn't quite have as much pepperoni as you wanted.
Have you ever met someone who was a hobbyist or amateur engineer at best, but always has grand ideas about how things should work? But has no idea of the type of investment or execution it would take to pull off. That is what I think of whenever I hear management strategies like these. It is like the management version of the compost fueled cars guy:
Deciding that you only want the best people is easy. Is there really a business out there who decides they want terrible staff? Determining who those people are is extremely difficult. Even if you are successful in identifying the best people, can you keep them? Whenever I find exceptional people, it takes more than 'not firing' them to keep them.
More realistically. Policies like these aren't designed with any idea of identifying or retaining talent. Instead they are usually a means to affect work culture. Like grading on a curve. It makes everyone work harder.
I would be worried that it make performance management of managers difficult. Which is to say a manager who fires people as a way of preventing themselves from being fired. The trick being "Yeah, last quarter did kind of suck but we fired that loser Bob, this quarter will be great." In fact there are a couple of reasons why the quarter was poor, it could have been poor Bob but it could also be the manager isn't cutting it. Unless the group has the option of voting the manager out of the organization, the whole "save the best, fire the rest" algorithm has problems. Or more accurately it has a blind spot with respect to management.
The interesting question for NetFlix will be how durable it will be as a structure. My impression is that it selects for the wrong attributes in management. But time will tell.
Because there is a structure, I suspect stack ranking makes it easier to identify empire building. Highly ranked individuals should exhibit behaviors that correlate with their high ranking such as moving to more challenging work within or without the team and company. High ranked individuals should be the sort of people that the next couple of layers of management are aware of because they create value and solve hard problems. High ranked individuals should be the ones other managers are trying to poach sideways. High ranked individuals should be the ones that the manager relies on to explain technical detail to upper management layers. They should natural have a particular kind of visibility on email strings.
Any culture where a manager can explain away a bad quarter by blaming their underlings is already screwed.
I agree with that last bit, but there is anecdotal evidence from employees that there is some validity to keeping a ready supply of people you can jettison out of your group in order to maintain altitude.
If I put myself in the shoes of a good manager, what do I do under stack ranking?
If my team is strong bottom to top, I try to find soft landings for my lower ranked employees because...well making people successful is what good managers do and concern for people as individuals beyond instrumental value is also a positive trait and it just seems fair and right in a karmic sense of the world.
On the the other hand, if the bottom people aren't really strong, a good manager gets cover for going against their positive traits.
As with most things, it works great with the design ideal person in the key positions. All organizational structures can work fabulously well when all you have are good managers and employees[1]. The question then is how does the organizational structure/guidelines deal with bad actors? All management structures include some sort of performance management, both for individual contributors and managers. Generally there is an incentive plan, a retention plan, and an attrition plan. You provide incentives to encourage people to perform beyond their own expectations, you provide a retention plan to keep the people who are delivering day to day, and you provide an attrition plan for helping people who aren't working out into a position somewhere else where they have a chance of being more successful. This is just management 101, the trick though is forced attrition and grain level boundaries (outlines of various attrition pools).
Forced attrition plans set up the dynamic that you find yourself in a pool of peers and you know that no matter what, every year one of you is going to get let go. You don't want it to be you, you know that you can be meeting all of the objectives of the organization and still be "the guy (or gal)" because no matter what, someone is going to get picked. My assertion is that bad managers, ones who embrace behaviors that good managers do not, can find ways to protect themselves at the expense of the good managers. It isn't good for the company of course, but the dynamic is already launched into motion. Given a very small epsilon that bad (in this case unprincipled) managers are probabilistically slightly less likely to lose their jobs, the system will slowly over time select for effective, but morally ambiguous managers. This is a very common evolutionary experiment folks run in simulation all the time. The smaller the epsilon the longer it takes, but it always happens.
[1] Sometimes that leads people astray into thinking it is their organizational structure, not their people, which accounts for their success.
I think it's wrong. It creates an arrogant, elitist culture within the company. Also I don't think anyone in the world knows what an 'exceptional engineer' is - If there even is such a thing.
If you put too much pressure on engineers to produce lines of code, they will optimize for that at the expense of quality/maintainability of the project.
I have worked for a number of startups which advocated the practice of 'cutting corners'. I think it doesn't work in practice because technical debt tends to creep up on you much sooner than you think. Unless the whole company's future is hinging around meeting a specific deadline for a specific feature, then there is no excuse for cutting corners.
I think the personality of the engineer is much more important than their skills. I have worked at companies that fired people after the first month - This is ridiculous.
You wouldn't fire a business executive after 1 month on the job because of poor growth metrics - The downtrend is probably a flow-on effect from past periods of mismanagement.
This is exactly the same in engineering. You can't measure the consequences of technical decisions until many months of even years after they are implemented.
While you might think you're firing 'bad engineers' - You are in fact just firing random engineers based on random events like market timing, which team they are part of, direction of the wind, phase of the moon, etc... Just chaos.
The question is also very much, how performance is measured. I saw performance measuring that was on the level of "solved tickets" -- with some special cases: E.g. when you did all the research and somebody else closed the ticket (with minimal effort), it was his.
There where plenty of fraudulent abuse of this system and in the aftermath, I guess it would have been better to flip a coin!
In such a system, the performance measuring as such, is undermining the corporate culture.
"Fire you for average performance" is a bit of a loaded term and implies a certain unfairness or vindictiveness.
I'd rephrase that to be– do you work in a team with uncommonly high standards? I'd imagine this to be the case at any cutting-edge company. Consider say, the special forces in any military. If you don't meet the grade, or you don't carry your fair share of the weight, you don't get to stay. That's just tha name of the game. I assume that this must be the case at all the exciting companies– Tesla, Google, Apple, Facebook, etc etc.
It's not stressful if the expectations are set in advance, and the precise conditions for firing are clearly stated, and you get the feedback and assistance you need to stay afloat.
I work with a small team that prides itself on being results-oriented, and I love it. As far as I can tell, everybody we hired is hired because they bring their A game.
I'd much prefer this to being a part of a team that "carries" people who don't perform.
Of course, we've all also heard horror stories by now of ridiculous young founders who come up with extreme ideas about productivity, 80-100 hour work weeks and whatnot. Sustainability is important.
My company will fire you for average performance. They do it quite pointedly. One day you're at your desk, and an HR rep with an armed security guard will come and get you. They'll escort you to an office, explain that you're being terminated, and remind you that you're an at-will employee. They'll give you some forms about insurance, and then the guard will walk you to the door. They do not let you clean out your desk. They not let you say goodbye to anyone. Sadly, this is the same procedure that my company uses if you give notice. They do not except two weeks notice. The same day you give notice, or perhaps a day or two later, you will go through this ritual.
To your point, I work for a sales organization. It doesn't matter if your performance reviews are outstanding. If you go more than a couple of weeks without results, the director of sales operations begin sending you nasty emails. Within a month after that, your escorted out the door.
This might be just the sales world, but at some point people deserve to be treated like human beings. Not like replaceable cattle.
The keyword you're looking for here is "topgrading". Basically it's management social darwinism that breaks down a team into "A Players" "B Players" and so on with the end goal being to weed out anything below "A-" by applying extra pressure on them until they either turn into "A players", or quit. Firing is a last resort. Netflix in particular is well known for encouraging people to voluntarily quit if they can't hack it.
The question is of course if all the "A Players" live and breath the company, voluntarily coming into the office on weekends, doing conference calls at 4AM etc - and you are the kind of person who values their family for instance, and wants a little work life separation, what score do you get? What happens if when offered a year long paternity leave, you actually take them up on it? How hostile will the environment be at the company when you return? Or what happens if you have a medical problem that impacts your job performance? And so on...
But, to answer your question... I once got sucked into a management overhaul at a company that used Netflix-style topgrading as justification for a wave of layoffs. The whole self-deportation aspect of it doesn't work very well - people dragging their feet in a job are often reluctantly there because they need things like medical insurance, have bills to pay and families to feed. Being able to just up and voluntarily quit a job because you're "not a good fit" is a luxury. So everyone who was a "B Player" (aka in a role that wasn't well defined) got put through hell for a month then fired, and then the remaining "A Players" lost morale and took up the voluntary quitting aspect. So basically it backfired into brain-drain and a chronic "A Player" retention problem.
Interesting post, I hadn't seen that before. Here is another factor to worry about: If you hire too many mediocre programmers, the good ones will leave. (Trust me, they can get jobs easily.) You'll be stuck with the mediocre ones.
I think the NetFlix strategy is partly a response to this.
People pretend there is a 10x difference between programmers. I assure you that is way too low. The difference is actually Infinite, because contributions can be negative (I've seen a bad programmer set back a project by 6 months). Even worse, a programmer can create a toxic atmosphere, causing good programmers to leave.
I like the NetFlix strategy. Some people are just a hell of a lot more productive than other people. (Example: http://bellard.org/ )
I assure you that is way too low. The difference is actually Infinite
Exactly. There's also the case when someone is highly productive but only for themselves. They create all their own tools, their own environments, their own platforms, they follow their own homemade software development practices, use their own testing tools, and none of it integrates with the rest of the world. But, because one person made everything, that single person is super productive (as long as they never step outside their own platform). They don't need documentation because they made everything from the ground up and their systems reflect their personal brain state.
Then you have to ask, even though one person is 10x, does it really help if nobody else can integrate with their work? Then every request, feature, bug fix is a bottle neck of 1 person no matter how much outside help you give them.
"Whenever I started feeling like I was falling short of what the company required from me, I viewed it as an opportunity to learn and grow, instead of blaming the culture for my anxiety"
- I would argue that the way this guy thinks is a personality trait. I like the way he thinks. Does this mean, if you are going through some tough time in your personal life and can't put in as much hours, places like netflix is not for you?
I worked at one of the worlds largest investment companies. Everyone in each group was reviewed and then fit to a bell curve. All compensation, title, and responsibility issues take the bell curve into consideration. This had the effect of making it possible to get fired for being "above average". If you have a group of 5, and they are the best employees in the company (all above average), well, a couple of you will still get, at the least, substandard raises and bonuses or, at the most (generally in times of cutbacks), fired. This is not uncommon for similar companies.
> But, wouldn't it be that per definition that the most people are average?
That depends which average. The usual "average" (arithmetic mean) isn't defined in a way that any actual individuals in the sampled group have to be at the average. (Some other "averages", like the mode, must be represented, but even with the mode it only is the most common single value, its quite possible, and normal, to have more not at the mode than at the mode.)
While empirically lots of things are distributed so that the area near the arithmetic mean on the most common scale is most heavily populated, there are cases where that isn't the case on any scale, e.g., bimodal distributions where the values cluster at two extremes and the average is somewhere between them.
> You can only be exceptional compared to something that is not exceptional
Yes, but a firm could in theory (if it could measure performance of its own staff and in the marketplace accurately enough) fire people who are at or below the market average, and only retain those of above market average performance, and everyone there (that survived those performance reviews) would be (by their most recent measure) above average -- exceptional, in that way -- for the market. (And this is true for any definition of "average" you want to use.)
This seems to be what Netflix is suggesting that they do.
My only experience with rated evaluations at a larger company were pretty ridiculous; I was rated a 4.5/5 by my immediate managers, but the higher-ups decided that everyone should be nerfed to between 2-3. Otherwise, they'd have to pay people. I'd bet netflix does a little better of a job but I'm sure it's mired in politics and other bullshit
Sales for a publically traded company. If you don't show growth above average, you will be fired. Unrealistic quotas and complete ignorance of market conditions (quota is the same in a boom or bust) leads to high performance standards. It's a constant high pressure environment and most people quit before they are fired to avoid the shame of a firing.
No. I work for a very large organisation. There have been cases of staff who did virtually no work at all and it was still a couple of years before they were fired.
The reality is reviews are a part of a process, and your rank in them depends in part on the quota set by HR. Most of the time, HR wants employees to fit within a bell curve for rankings. In some extreme examples, people's reviews were based on a predetermined track and not on actual performance.
Netflix's policy is just HR marketing. If all you do is keep your 4 and 5, then what makes those people 4 and 5 will start to become 2 and 3. It's a moving goalpost, and sounds like an effort to increase productivity without increasing compensation.
I may be jaded.