Slashdot Mirror
Microsoft Kills Stack Ranking
Nerval's Lobster writes "Microsoft once demanded that its managers place their subordinates on a scale from 'top' to 'poor,' a practice that fueled some epic backstabbing within divisions. Last year, a Microsoft contractor with knowledge of the company's internal review processes told Slashdot that Microsoft was actively working to fix that system; just this week, the company announced that stack ranking was well and truly dead (and that's certainly one way to fix it). 'Lisa Brummel, head of human resources for the company, sent an e-mail to employees notifying them of the change today, according to my contacts,' ZDNet's Mary Jo Foley wrote. According to the memo, there are 'no more ratings,' 'no more curves,' and 'Managers and leaders will have flexibility to allocate rewards in the manner that best reflects the performance of their teams and individuals, as long as they stay within their compensation budget.' They're trying to encourage more teamwork and collaboration throughout the company. As we discussed on Saturday, Yahoo is adopting this method just as Microsoft is abandoning it."
Maybe Microsoft will be able to re-invent a better version of itself.
Get free satoshi (Bitcoin) and Dogecoins
And yet both companies will have the same outcome - continuing their long decline into irrelevant mediocrity. Maybe both companies should consider looking a little further up the management chain to discover what truly ails them.
Yahoo seems to be on a roll with this, as they adopted a Win8 pane interface on Flickr right about the time Microsoft was forced to concede people without tablets, smartphones, and touchscreens on their computers (and some who do have those things) dislike it greatly.
Laughter is the Spackle of the Soul.
A bunch of years ago a company I worked for was doing something similar.
They essentially demanded it be placed on a bell curve. So, in our group of 5 people, all of whom were good solid people who worked well together and got stuff built, management was insisting there be 1 awesome, 1 pretty good, 1 good, 1 needs work, and 1 terrible -- and that had nothing whatsoever to do with the individual strengths of the team, just some idiots vision of how these things should be managed. My manager didn't feel that anybody belonged below the top 1 or 2 rankings.
If you decide in advance that your ranking has to take on an artificial distribution, you end up with a really pointless management system which really just serves to give people with no knowledge of what really happens a nice easy to read (and often incorrect) metric.
It really does make for a pointless "management by inapplicable metrics" kind of culture. And so often it's all about making managements job easy and something they can point to the formulas -- and seems to offer zero insights into what is actually happening. The more companies blindly use metrics, the less they actually grasp what their organization is actually doing.
Lost at C:>. Found at C.
OK we know from MS history they treated customers poorly. We know they treated mom & pop shops poorly. They treated the companies that make apps for them poorly. Now we find out they even treated the employees poorly. Honest question, did they ever treat anyone right? I mean besides the management figures making 7 figures. Wait, that might not even be true. Wow, sure am glad I never worked there.
I'm old, not dead. Well that's my 2 cents worth, your mileage may vary. I say what I think, not what you want to hear.
You are a shitty manager if you need to resort to "stack ranking" or whatever.
1. Your recruitment policies suck - that's a given for all of IT/Development/Software - industry - if you can't find qualified people, it's YOUR fault. If you do find "qualified" people and you still fail - look in the mirror.
2. If they get hired and fail, then WTF is the problem? Unrealistic deadlines? Changing scope? Death marches?
3. Every problem is management's fault. Period. End of story.
Don't get me started on the idiocy of Silicon Valley: Kids, don't work there. They are milking the reputation of true innovators like the Dave Packard (Business guy) and Bill Hewlett.(engineer) - today, they are a bunch of marketing phony assholes and cunts - looking at your camel toe Ms Mayer .
Silicone valley is for posers. Pass the word.
Yahoo is adopting it because it's a great way to get rid of dead weight, as long as it's used BRIEFLY. It's really not meant to be used in the long-term (as MS and several other have tried to). In the short-term, Yahoo will lose some dead weight. In the long-term, they'll get paranoia, indecisiveness, etc. (in short, a company culture of fear).
The cow says "Moo." The dog says "Woof." The Timothy says "Thanks, valued customer. We appreciate your input."
My ex-employer KPN (Netherlands) also does (did?) this; each manager was telling it just had to be in any group of at least 8 employees, and would then show a Gauss curve to prove it ... I'm happy I'm not there anymore.
Obligatory
"Unheard of means only it's undreamed of yet,
Impossible means not yet done." ~~ Julia Ecklar
I think you glossed over my point. It's not these silly management initiatives which determine the outcome of a business's success but the core intelligence and culture of the business itself, particularly in its executives and management. Poorly-run companies are always latching on and off the the latest management fads because they lack core direction and competence.
...Yesterday's Solutions Tomorrow.
"I don't have to outrun the bear; I only have to outrun you."
Welcome to the Panopticon. Used to be a prison, now it's your home.
Lockheed Martin also employed stacked rankings. The local manager had no clue who people were. How can you even rank your employees when you have no idea who they are?!?!? I was called by another coworker's name multiple times. I finally called my manager out on it in front of everyone at a picnic. He didn't confuse me with the other individual after that... There was so much turnover we basically lost a contract due it and having to retrain new people ALL THE TIME. I don't blame those of us who left. Many people busted their asses and did an excellent job, only to be rated average or below because the manager had a certain number of slots to allocate certain rankings. AND THAT'S IF HE KNEW WHO THE FSCK YOU WERE!!!!!!
I worked at a large corporation whose name started with an A and ends with an E.
They too had a ranking system, the lowest got sent to a certain team in our general group where they were needled to death over their stats and either quit or accumulated enough "black marks" to get canned.
When the team rotation came around, the lower ranking people got suicidal, dread is the word of the day, when your name appeared on that "special team" list it was like getting sent to a death camp.
That person is now tainted and must be shunned.
I saw good techs go down for not having enough "personality" (flashbacks of *37 pieces of flair* from Office Space) and it was a dismal atmosphere.
I left that sh*t hole, never got my turn on the death team.
Frankly every large corporation I have worked for is the same in that they have all the makings of a cult... I mean if they wanted to go that way.
"If any question why we died, Tell them because our fathers lied."
While they don't have to be ranked so strictly you're still fighting over table scraps against your own team.
If everyone surpasses expectations and achieves a good result then everyone deserves to be compensated fairly.
Yeah, if you know you want to lay off 20% of a large workforce, it makes sense to take some metrics-- including some subjective evaluation-- and develop a ranking of employees from "extremely valuable" to "a drain on company resources", and then cut the bottom 20%. Do that as a one-time thing, or even do a couple rounds in relatively short succession. That could work.
But if you make it part of the company culture, you're going to end up with a company of paranoid back-stabbers.
And yet both companies will have the same outcome - continuing their long decline into irrelevant mediocrity.
Somehow despite geek opinions, Microsoft's revenue keeps going up. Yahoo is starting to look up as well, though how much of that is Alibaba is hard to say.
All industries eventually mature. Being on top of a mature industry is a good place to be, as long as you occasionally shake things up enough to stay on top.
Maybe both companies should consider looking a little further up the management chain to discover what truly ails them.
You mean like getting a new CEO, which Yahoo did and MS is doing?
Socialism: a lie told by totalitarians and believed by fools.
Stack ranking works great if you use it to get rid of the bottom 1% every year. Surely in a department with 100 people there is at least one hire who didn't turn out great.
The problem is when it is applied at a 10% threshold. It is not hard after a few hiring/firing rounds to end up with teams of over 10 people all of which are very good, yet stack ranking still demands that you fire the "bottom" perfectly OK person-decile.
stupid HR fucks don't understand simple statistics.
Of course these are the same morons that want to play keyword bingo with your resume, think everybody in a 200,000 person company needs ethics training to make up for the moral deficiencies of the executives in the boardroom and want 5 years of experience with some technology that's only been around for 2.
the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff
But if you make it part of the company culture, you're going to end up with a company of paranoid back-stabbers.
And what is the problem with that?" asks Larry Ellison.
I work for a government-owned, contractor-operated lab where we are ranked 1..N, and it's not all that destructive. Why? Because there have been no significant raises in years! So, staff and management wastes months on performance reviews, then the results are put on file and never looked at. The end.
Someone wrote that grading on a curve works in academia but not in industry. Why should it work for grading exams when it doesn't for ranking the workers? Especially the academics that are using it should know better.
Grading on a curve (or the MS stack ranking, which is the same) is one of the most unfair and vile ranking/grading systems invented. Why? Because your actual skills don't matter. What matters is how many better (or worse) colleagues you have. If you have are in a large team (or class) of good performers, you are screwed, even if you are good - someone will be given the short end of the stick only because there are only so many "good marks" available. An extreme example are students "hacking" their exams by handing in blank sheets. Even if they all (or sufficiently many) do that, with curve grading they are guaranteed some 75% chance that they will pass - by doing nothing, because only the low 15-20% fails. Shouldn't we be marking their skills and knowledge instead?
This system also demotivates the good learners/workers - what is the point of trying to work hard, when you will not get that good mark only because there is only a limited amount given out and simply too many comparably good candidates. Essentially the system forces (undeserved) bad marks on people even though they performed equally well as the best ones. This sort of thing does wonders for morale.
Finally, the second fallacy why this is fundamentally broken is the assumption that the skill distribution in a work team or class is normal (follows a bell curve). There is absolutely no guarantee of that, because, heck, you aren't hiring the idiots, are you? I am sure that the company is hiring only "rock star" developers. Same with the students - they have to pass stringent exams and fulfill admission criteria that the majority of the population isn't able. So you have a sample here that isn't representative of the entire population (where the bell curve would be valid) and all bets are off, because the system was built on an invalid assumption. The most extreme example of this is the constant distribution - the case when all students turn in blank sheet of paper (identical "skill" level) for their exam and still pass. You would have to pick the students or hire employees randomly out of the entire population if you wanted to have a normal distribution of skill. Not very practical, though.
To conclude, if you are responsible for examining students or for evaluating employees, for the grace of God, stop using relative ranking schemes like this. Comparing people to each other is certainly easier than to evaluate their "absolute" skill, but it isn't fair, doesn't represent what you think it does and it creates a toxic environment for everyone.
If you have to fire someone it can only mean one of two things. Either you didn't train the person well enough or you hired the wrong person. If you abstract what is going on enough all firings ultimately fall into one of those two categories, both of which are a ultimately the responsibility of management. This is why stack ranking is a bad idea. If you didn't train the person well enough then improve your training program. If the person was the wrong person for the job (insufficient work ethic, incompetent, unethical etc) then improve your recruiting program. Stack ranking treats the symptom instead of the disease. It takes emphasis away from focusing on hiring the right people and training them well.
No company will get every hire right (some people just aren't what they seem to be) but creating a culture where everyone is playing a game of "devil take the hindmost" will get people to worry less about getting the right person because if they are wrong they won't last. Hiring someone only to break them off later means someone made a very expensive mistake.
While I was at Microsoft, at one point my manager instructed me to stop having ideas that were outside my assigned area, because it was making another team member look bad, and this would impact the stack ranking of both the team member and my manager. So I saw up close how stack ranking sucks. Still, if I was still at MSFT today, I would be very concerned that the new system drives even more of the compensation process into closed-door management sessions, along with the horse trading and cronyism that invites.
A) It fosters competition, which should also foster a better product (I don't actually agree this is the case, though)
Where I work, if someone else is good at their job, that's good for me, because it makes my job easier. With stack ranking, if someone else is good at their job, I'd have to try hard to make them look bad without them noticing, so that I look relatively better. Where I work, I'd help somone getting their job done so we get a better product. With stack ranking, as long as it looks like their fault if the product is shit, I'm Ok.
You spelled 'pain' wrong.
I had a prof in college who explained the whole 'grading on a curve' thing.
He started his explanation by saying that a good exam would have:
* several of no-brainer questions to check if everyone had at least done the minimum and give them a warm-up for the real questions
* several of real questions, to make sure people actually did study, and to produce some differentiation amongst the general population
* one or two incredibly difficult questions. Questions so tough that most people are NOT expected to get them. Questions so tough that if you do get them then you the prof should come talk to you about majoring in the area (the prof taught Freshman/Sophomore level math).
According to him the point of grading on a curve was to be able to put that third category of questions on the exam without destroying the grade of everyone else in the room.
I, personally don't grade on a curve but I thought it was interesting that it can serve a purpose in some situations.
> Yahoo is adopting this method as MSFT ditches it
Well, that's certainly one way to keep ex-Microsofties from applying. :-) "What did you hate most about your last employer? Uh, yeah, we got that now."
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
That's definitely better than the "every department cuts a fixed percentage of their workforce" nonsense that a lot of companies do. I've always thought that the best way to go would be to do a ranking and cut the bottom X% and give the top Y% cash bonuses in the very same swoop. Layoffs have a way of shaking people up, and your best people can easily find another job if they decide they don't like to live with uncertainty.
Give them a pat on the back and a bonus offer the same day you're laying people off so they know that they're not at risk, and make the bonus vest over a few months to keep them around long enough to see that things are OK. You're saving recurring salary and HR costs on X% of your workforce, so the cash is there for a one-time payout.
An interesting anagram of "BANACH TARSKI" is "BANACH TARSKI BANACH TARSKI"
The place where I work recently introduced OKRs and Stack Ranking, bragging about how teh awesome it was at great companies like Zynga, so it MUST be teh great idea at a stupid place like ours.
This was when Zynga was deep in death throes and shedding value like a hairy dog in July sheds hair. For HR, pride. We're like some internet company the executive assistant has heard of! For people who know things about struggling companies, completely laughable.
We are teh bullshit INC. Let's be like Zynga! Oh yeah!
We're about to see what the first quarter of OKRs will bring, where, as they say, the trickle down cascade goals (which nobody has bothered to discuss with me at all) are not actually supposed to be reachable. "Because reaching them means you didn't set the bar high enough." Not reaching goals ALSO means you no longer qualify for pass/fail bonuses or promotions so the meager cash kick (typically one third of a regular paycheck; that's right a fraction of, not a multiple of) we get is effectively eliminated. Nobody is going to meet goal any more. But they promise OKR scores are "not to be used" for eval purposes.
Then what the fuck ARE they for? Shits and giggles? They expect us to believe this bullshit. "Your metrics show... oh you didn't meet any of your goals! Tsk Tsk. You are now on automatic probation!"
I expect, no, I WANT to be first against the wall when the stack ranking cuts come. Cash me out. Give me my unusable vacation time and some severance and free me from this madhouse. And they damn well won't DO it! They know what I want and won't do it.
Damn them.
Sig for hire.
... I used to regularly score 'above average', or in MSFT stack rankings, a 3.5 or 4.0 (the latter was hard to achieve if you weren't the golden-boy - required to balance the team score). This meant I would get a performance based bonus, which was great.
I made the mistake of pushing for a promotion. I felt that because I was consistently out-performing my role, that I should be promoted. Eventually they promoted me and a few other guys. We got a 'Senior' title. Now comes the problem.
The promotion only came with a 2% pay rise. The following annual performance review, it was now deemed that I was not exceeding my role (due to the new title), so I only scored a 3.0. This score means 'you met all your objectives'. Unfortunately, at the time, the policy was bonuses were only awarded to those exceeding their job description. I got no bonus. That year, or the following year. It probably left me on average $5k/year out of pocket.
Moral to the story? Don't be an employee :-)
"Never let the truth get in the way of a good story..."
My understanding is that at Microsoft, the bonuses are largest for rank 1, slightly smaller for rank 2, etc... and the bottom of the list get kicked out.
It doesn't matter if your ten person department has people with IQs from 180 to 189, the person with 180 is going to lose their job unless they game the system to rank ahead of a colleague. (Might be fun to watch a group of super geniuses outwit each other, though.)
I don't know that I'm skilled enough to make it through Microsoft's hiring process, but the stack ranking system is one of the things that prevented me from even applying.
At Microsoft the ranking system determines who is fired, and who gets the largest bonuses. That's why it was so divisive - if you're near the bottom, stabbing a colleague in the back could be the difference between keeping your job and getting fired. If you're not in the bottom, stabbing a colleague in the back, or at least failing to help your colleagues, could make a $10,000 difference in your bonus.
Measuring people is fine. Giving the measurements an impact on employment and pay destroys collaboration, and as a secondary effect it attaches a larger incentive to working fast (so you can show your manager a big list of accomplishments) instead of attaching an incentive to doing high quality work (which might lead to a shorter list of accomplishments, but fewer security holes and other errors that need to be fixed later).