Microsoft's EAT-your-OWN-YOUNG management system AXED Microsoft may soon be a much nicer place to work, thanks to the company's announcement that it's doing away with its infamous "stack ranking" employee performance reviews. In a company-wide email obtained by the Wall Street Journal, Lisa Brummel, executive VP of Redmond's human resources department, explained that managers …
It would be interesting to know what drove this change. It is hard to believe the leadership team that has done so much for the company over the last decade has had an epipheny. Similarily for labour, there must be many unemployed in the US/India who have lost all shame and who would be prepared to put Microsoft on their resume. Dig deaper please el Reg, what couldn't they do or who couldn't they get under the stack rank system? Who in senior management happened across a Google tech talk or is recovering from head trauma?
" It is hard to believe the leadership team that has done so much for the company over the last decade has had an epipheny."
I'm more interested in why this was announced by the VP of HR and not Ballmer himself. If you really believed this would lead to a turnaround you'd want your CE yelling it to the world ... (or wall st, which for public companies is the same thing).
Really?
The decision is coming with enough second guessing having been announced by the VP of HR. If this had been announced by Ballmer himself, Alcoa couldn't make enough aluminum foil for all the hats the posters on IT Tech sites would need for their hats. So even if the idea had originated with him, somebody else had to announce it.
She was tasked with developing the new system. Being in charge of Human Resources, that's a fairly sensible move. She probably hated the old system as it increased their turnover so her staff were constantly counseling staff & searching for replacements. Balmer, who brought the idea to microsoft & is leaving, would be the least best choice since he has little capital with his employees
Cheers !
Quite right too. I left a previous employer partly (among other things) because of the performance management and employee rating bullshit, because it was frequently gamed or otherwise massaged and absolutely soaked up valuable time which could be spent doing something useful.
Most people just did X for the sake of their performance rating and bonus, not because X was worthwhile or needed doing. As it became a mundane and dreaded box ticking exercise, it ended up making people not actually care about their jobs which was the exact opposite of what HireRelease anticipated but never mind what do I know lol. Even HR themselves hated the overheads involved in commissioning and processing it all so why bother? Same goes for employee surveys which were all gamed (HR again either didn't know or didn't care), we were pretty much all told what to say by management and nobody dared be honest in case the survey wasn't anonymous after all. There were frequent witch hunts resulting from this.
The ratings were arbitrary too. You could cure cancer and be told your performance was "Satisfactory" while your colleague could do something completely and utterly pointless, unrelated to his role and get "Excellent" with a payrise.
You could cure cancer and be told your performance was "Satisfactory" while your colleague could do something completely and utterly pointless, unrelated to his role and get "Excellent" with a payrise.
Of course - if you cure someone he will not need further medicine, so the one with the "pointless" results generates more profits.
So who is the better performer from a company view?
My old employer did this too. Instead of the appraisal (which everybody hates anyway) being over fairly quickly we had people gaming the system, and actively trying to sabotage rivals. Not exactly goal congruent. Then after we’d had our appraisals there was a moderation process to see if we’d been judged fairly. This was rubbish and simply down office politics, so if you were “in” with the manager who shouted and bullied you’d get a good rating because he said you were a “good chap” (or woman) and he would protect his cronies. Others doing a better job but with more moderate managers would suffer. This process took weeks, and removed senior management from the business for much of that time. All that to fit a distribution curve that some HR wonk said was how it should be.
No wonder people left.
I did see a fraud at a company that did “rank and yank” whereby the bottom 5% of rated employees would be removed each year. As it was so bloody stupid and morale damaging several enterprising managers put ghost employees into their team, paid them a salary and then sacked them at the end of the year. The salaries were collected by real employees and divided up amongst the team at year-end. The team was happy with this game, senior management not so much.
Office 2000 came out around yr 2000, including a major upgrade of database functionality to work with SQL Server 7. 3 Months later SQL Server 2000 came out, and broke that major upgrade. They must have been working on that stuff for two years, and it 2013 now: Microsoft has had distructive disfunctional competitive silo's for at least 15 years.
Don't some bits of BT still run the idiotic and counterproductive system which MS have just said is a Bad Thing? (E.g. the bits where the CWU still have a bit of influence don't do this madness, but it still goes on where "professional" employees are "represented" by some toothless management-lapdog organisation).
Read technical new sites? Managers up to band 10 - quite possibly; DE's and Fellows - you would hope so; letter banded non-technical executives? I think they have a training course for them to stop that sort of thing.
I almost sent a link to this to my manager, but honestly, what's the point. He agrees with me on the forced bell curve front but he just has to play the system (and to his credit does it as humanely as possible).
I've never had a bad appraisal marking (luck as much as anything I'm sure), but I still despise the system.
...my favorite article about MSFT, that told the brutal truth about the utterly incompetent Ballmerian troupe, shined the light on this so badly mismanaged, wasted decade-long period, finally got the right treatment - after many links in comments by me it's got mentioned in an article! :)
Good job Microsoft. Great smokescreen. What! Do people really think stack ranking will go away? What do they think will happen, some socialist system where all are paid equally? The core philosophy will remain at Microsoft as with most every modern company, Pay For Performance. And, you have to pay your top performers competitively or guess what happens. And then you have to pay your next to the top performers competitively...... If your pay reflects your performance, there is ranking, no matter what you call it.
They made it clear that there are no changes to the overall compensation budget and that managers have to manage salaries within their budgets. How does anyone think this will happen without some kind of ranking going on, explicit or implicit.
I admire Microsoft for their marketing...of their HR.
"What do they think will happen, some socialist system where all are paid equally?"
Perhaps they'll just be paid based on their seniority within the company, moderated by what the market usually pays for someone of their specialisation. Or is that too socialist for you?
"They made it clear that there are no changes to the overall compensation budget and that managers have to manage salaries within their budgets. How does anyone think this will happen without some kind of ranking going on, explicit or implicit."
That's true, there will be a sort of ranking that goes on in the managers head. He/she will reward his/her team based on their contribution to the team's goals. If everyone does well, they'll share it equally. Of course this means that teams working under different managers will have different rewards. All they have to do is have a system in place whereby action can be taken if they think they're being treated unfairly.
Is that the only alternative?
Is it no longer possible for managers to decide how many of their team are top performers - does it have to be an HR set figure or everyone?
Ranking goes on, and will continue (and even exists in socialist approaches), however the flaws with stack ranking are based around the set limits of the bell curve. Removing the forced bell curve is the improvement - or didnt you read that bit?
Good news anyway. A healthy Microsoft might keep IT development on its toes. Maybe after Windows 8 is buried like Vista. Agree with other commentards over ranking being gamed. No-one trusted the system, rankings or their managers managers. In one interview I was told I was rated well above requirements, but that meant I was rated as only meets requirements as I was a senior and meant to be at top of assessment heap. Very discouraging when you see a reasonable bonus vanishing over mindless continuous improvement waffle. In a high productivity team, it was all the more galling to see a compulsary bell curve fit to staff who were very competent. This, despite team leader and manager who did their best to support their staff. Seems the looting classes had never heard of asymmetric distributions, which would be the logical outcome of a continuous improvement program.
Putting a score on employees is bullshit. If an employee can improve then help them improve. If they do something exceptional give them a bonus. If they're hopeless fire them. But Good Christ, don't 'score' another Human being. That's just royally fucked up.
Outside of the (im)moral behavior, a score will bite you in the ass sooner or later. Somebody will get hold of a spread of scores and sue the shit out of you. Which, incidentally, is what I expect happened here.
------------------------------------------------x
--------------------------------------------X. X
------------------------------------------X.
----------------------------------------X
-------------------------------------x
---------------------------------X. X
---------------You are here--X. X
Let me guess - first companies require that employees be ranked on a set curve requiring some number of people to fail, and then they all sit around wondering why 20% of their employees suck no matter what they do?
Tell you what: When I'm going into the third nail-biting month teetering between wealth and insolvency, there's nothing like a reminder of why starting my own business was the right decision...
You have your job and you have your goals. Not always the same thing. The most important thing is to be able to say the goal was done at the end of the quarter. Notice I didn't say done right.
A lot of people have gotten poor reviews because they tried to do a good job on a project, say write and procedure, but they couldn't complete it in time. While the people who wrote something completely useless could say they were done and got the raise.
AC for obvious reasons.
Hands up who was surprised?
--- Microsoft will focus more on teamwork and collaboration, and future performance reviews will take into
--- account how employees work with others, in addition to evaluating their individual success.
--- This is a fundamentally new approach to performance and development
No it's not, it's called common sense, and is basically a succinct summary of what is expected of management. Seems that the rapidly eroding status of the company is deflating some self-reinforcing bubbles of dogmatic nonsense that have inflated during their tenure at the top of the tree.
I'm surprised! In my experience everyone else is blamed when a management theory is proved a failure. There's a rotten 'ya canny break the laws of management' attitude out there and facts have never stood in its way before. I'd be inclined to think this is just management speak for 'Oh yes you will all get rewarded for the work you do....if there's anything left after management share options are exercised'.
Once a company has gone this rotten it will be almost impossible to undo the damage as the people in charge are the people who got there via the old route - they're didn’t reject their promotions then, they're not going to let them slip now.
.... "there is more joy in heaven at one sinner that repenteth than in nine and ninety just men".
On a somewhat more serious note it has been my experience that this vile practice much beloved of The Managerati has poisoned work place relations for over a generation. A thoroughly inhuman system that brought out the worst in people and never achieved what it said on the tin - not even from the point of view of the managerial daleks that implemented it.
one might instead consider the compelling motive for the change:
""This is a fundamentally new approach to performance and development designed to promote new levels of teamwork and agility for breakthrough business impact," Brummel wrote."
Translation: "The bleedin' boat is sinkin' faster than the pumps can dump it overboard. All hands stand down from slammin' each other wid yer oars, and all of ye grab these bailin' buckets and turn to! It don't matter who's gonna get the credit; if this damn scow sinks, we're all gonna drown!"
"Yeah?" one scallywag shouts. "Then git yerself over to Ballmer, there, and belay his damned drill!"
I've heard of companies which aim to ditch the bottom 5% of employees. The problem comes if the bottom 5% of your employees are better than the best you can attract to new positions!
This also assumes that there is no benefit to being familiar with the company practices and culture. Mind you, at companies like these, I can see why that might be the case.
The managers always complained about the system; it was done by middle level managers (not line, two levels up in the hierarchy) and they were forced to rank everyone, "even if everyone performed exactly the same"; ho ho, aren't managers jolly chappies.
The MS fanout was 3-4, the hierarchy depth was about 7 (well, I was seven levels below BillG, I counted once), for those who can't do math there were a little over 20,000 employees at the time. Using 4 for the fanout there were 16,000 grunts, 4,000 line managers (I was one) and about 1000 middle managers.
From a hierarchy point of view the system worked, and probably still works, because the upper management can only handle a certain amount of information. Whether I or anyone who reported to me was particularly brilliant is particularly uninteresting; what matters is what my manager's reports as a group do. Or, more realistically, my manager's manager's manager, who is stack ranking him (or in a couple of cases only, her).
The downside to the system is that individual excellence or, for that matter, individual stupidity, has no route to the top or the bottom - you live or die along with the rest of the people ("team") you happen to end up with.
I believe Microsoft didn't care - it had a somewhat more enlightened attitude and realized that every one of us can do remarkably good, or remarkably bad, work given the right environment.
Alas somewhere along the time someone fell for someone else's marketing bullpuppy and maybe now believes in the model pursued by a certain other company which I once heard referred to as "prima donna programmer." (That reference comes from around 1994 from someone who never worked for Microsoft or the other company in question.)
John Bowler
Intel were doing it in 1986/7 when I was there.
Got out just as the appraisal process finished, but a colleague told me she was unhappy that I'd been graded "marginally meets" (this was "get that again next year and you're out") and would have supported me if I'd appealed it.
Personally, I thought I'd done bloody well as they'd walked away from the market I'd been recruited for before I actually started (Xenix systems across Europe) and I got stuffed with 386 PC motherboards and systems. For which we had zero share-of-mind from the chip division as our volumes were orders of magnitude less than the "proper" manufacturers.
The high spot was the badge-engineering box-shifter who thought he could get away with calling them "Imtel" systems (the boxen were white-label).
With a passion the bell curve. It's never made sense to me, surely a good tech company wants to hire over the normal amount of "extraordinary" individuals overachieving. Then you get told that because you were exceptional the previous years you obviously can't be this year (as no one can ever keep being excpetional!) so your goals are now even higher so you are just satisfactory or good this year... Others get praised for handing in a completely pointless management report which was the managers pet thing this year (or covered his ass). Morale? What's that?
The Human Remains staff who thought the stack ranking system was a brilliant idea are some of the most stupid people around. The bell curve is a *result* of looking at a large population, it it not something you work to. They are doing it the wrong way round and it doesn't work for small teams. The bell curve will be a result of looking at the whole company, not teams.
Fitting people to a bell curve doesn't make a gereat deal of sense. Yes, if you put a random selection of people in a type of job, say sales executive, and manage somehow to have no external variable there may well be a bell curve of performance. But in a real company, over time, the good ones get promoted or go to other companies for higher pay. The bad ones get booted. But the ones in the middle tend to stay there as they've found their level. So most companies have a much tighter distribution than a bell curve.
No doubt some companies will use the bell curve shaping approach intending to get rid of poor performers. Unfortunately all too often it makes staff all too aware of how the company measures poor performance. And the measurement of individual performance is seldom aligned to what is good for the company.
What alternative methods are out there ?
Not condoning or slating the bell curve but if it's down purely to managers discretion, whats stopping the cronism there ? Even if theres no curve us drones will always have an issue with why Jones in DB infra got more than me. He's a lazy basterd, that Jones....
D
"that the HR drones who introduced it in the first place were summarily fired and tasered off the premises by a group of overweight and sweaty rent-a-cops"
Now then, don't blame it all on HR, they mostly just have to do what senior managment has been told by the most recent team of consultants hired striaght out of doing PPE at Oxford tell them is the next gee whizz idea. HR drones don't make this sort of decision, the finance director does - it is easier to work out your remuneration budget if there is a nice Gaussian distribution of achievement to link to reward.
The problem is that they do not understand statistics, that it is unliely that a company would ever have a normal distribution of achievement and capability reported every quarter not only overall, but for each colleaction of >10 individuals, particularly if you are in any way selective in your hiring and firing processes.
Folmi, I think, reckoned that you shold reward people according to their value to the company, if the man who stoked the boiler made the greatest contribution he should get the most pay. But in feudal organisations (like any large company) there is the belief that each manager must be paid more than his or her staff because managers are MORE IMPORTANT than the people who do the actual work the company provides to its cusotmers.
Things like this are kind of two part, but the primary component is a lack of strong management. The last couple generations of managers and execs are simply afraid to take risks based on observation, they want metrics. Humans just can't be measured like that and it would have been obvious to them had they ever had a life outside college and their first cube.
For example, you can't put a score on the less than stellar developer who is also the peacemaker between the leading egos in the group. You've got to have the guys who excel at their tasks, but you've also got to have the guys that keeps the wheels greased. You've got to have the guy who can put up with extra departmental stressors and make them go away without pissing anyone off. They might not be as good at their assigned role as some of their peers, but they are just as valuable.
Without those people in the group squabbles and problems get escalated beyond the point where a reasonable response can be expected. For example, I have zero time or interest in your problems with Bob in the Corner. If there's nobody to mollify Bob and the other person I'll just fire one of them and hire somebody else. Nobody is irreplaceable and I would rather have reduced output than be stroking egos all day.
Behind every exceptional employee or group are the people who nobody knows their names, but are actually making everything work. Too many modern managers lack the ability to identify those people and to take a chance on them. That's just dumb.
At last some sense in MS, although you do have to wonder why the people who propagated the system are still there. Stack Ranking can work but only as an emergency measure over a short term by reducing the personnel crud that can accumulate in legacy industries.
But for 10 years!! Even GE realised there mistake and dropped the system years ago.
However I still have a problem with any individual rating system.
Ratings should be about the team performance not individuals. If a team is doing well and meeting or exceeding their goals then the make up and composition of the team should be of no concern. Reward the team and they will self form into a successful entity, individuals hampering success will be excised as a matter of course.
The only downside is that HR and senior management then feel they are not doing there jobs
WRONG: GE still uses/abuses the system!
Work like hell and get an excellent rating one year - your goals are set so that becomes the expected, and if you work like a normal human being the next year you are so far below the expected that you are out of the door. (Regardless of whether you are still contributing more than your peers!) And don't worry that they can't get a new employee who is anywhere near as good, you become the overhead reduction. (The aero-space sites where I still have friends are all staffed at about 75% of what they need to do the job properly.)
Don't get me wrong, the bell curve rating might be suitable for production line productivity which is all Jack could understand, but not in a technical development scenario. The 'cult of Jack' is apparently still strong with one high ranking HR droid last year openly saying that if 20+ year "experienced designers were in the bottom 10%, there were plenty of people working in the local supermarket who could replace them." And these are the guys who enforce the system!
One of the 10% of under-achievers a few years ago.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
