I was one of those who was laid-off, or fired, or whatever you want to call it. I've been really confused over the last week because the reasons for my termination were pretty vague. But wow, after reading that, so many things make sense now.<p>One of the areas this type of a bell-curve system falls apart is for highly specialized and well-qualified teams, such as the one I was working on. Who knows, maybe I was a poor employee, but after all the raises, accolades, and the promotion I received (by my firing manager), I'd like to think that wasn't the case and it was just a manager put in a tough spot.<p>Regardless, it was a nice run and the job market is great. I'm taking a bit of time off, but if you are a company in SF and looking for someone with lots of JavaScript, Node.js, and frontend experience at startups and (of course) big companies, links are in my profile.
If you're a manager in a stack-ranking organization like this, in this market, why not take a stand? Refuse to classify team members who are not actually "occasionally misses" as such. If it's straight-up stack ranking, give everyone a tie for 3rd place or something.<p>What's the worst thing that happens? You get let go? What a great way to get let go from a dysfunctional company.
These types of forced ranking systems are very tough. Microsoft has had a lot of complaints about there's. GE, on the other hand, has had success with theirs. I've seen them work successfully at large firms where they were ingrained, and the companies had other cultural norms to increase teamwork. I've seen it flop at a few software firms.<p>What they're good for:<p>- Forcing managers to make hard decisions and give real feedback. (You can't just tell everyone "You're doing fine.")<p>- Highlighting who at the bottom should be let go, and who a level below should replace them. (In most firms, the top 10% at a given level outperforms the bottom 10% at the level above them)<p>- Providing internal guidance for larger layoffs.<p>- Encouraging constant trimming to avoid those larger layoffs.<p>The problems of these systems are numerous though:<p>- It can discourage teams of superstars from working together on large projects. "What if only 1 or 2 out of us can be ranked high?"<p>- It can encourage selfish behavior and infighting if there aren't other counterbalancing forces.<p>- There is frequently disagreement for anyone not in the chosen group.<p>- Much of the differentiation is based on managers ability to fight for their employees, rather than employees actual ability.<p>It will be interesting to see how this plays out. In Marissa's defense, she wasn't hired to continue "Business as usual".
I worked at Yahoo, and quite frankly, Yahoo needs to go through a few years of culling the bottom 10% like this. After several iterations of this, it could be a great company again.<p>What Mayer is saying is that she is willing to sacrifice those employees that are at the lower boundaries of acceptable, in order to make more room for the top 85%<p>And quite frankly, the managers that are complaining about having to put people at the lower boundaries are highly likely to be underperforming as well. When I was there, it was mostly filled with bloated and lazy employees.
> In 1981, Ford's sales were falling. Between 1979 and 1982, Ford had incurred $3 billion in losses. Ford's newly appointed Division Quality Manager, John A. Manoogian, was charged with recruiting Deming to help jump-start a quality movement at Ford.[19] Deming questioned the company's culture and the way its managers operated. To Ford's surprise, Deming talked not about quality but about management. He told Ford that <i>management actions were responsible for 85% of all problems in developing better cars</i>. In 1986, Ford came out with a profitable line of cars, the Taurus-Sable line. In a letter to Autoweek Magazine, Donald Petersen, then Ford chairman, said, "We are moving toward building a quality culture at Ford and the many changes that have been taking place here have their roots directly in Deming's teachings."[20] By 1986, Ford had become the most profitable American auto company. For the first time since the 1920s, its earnings had exceeded those of arch rival General Motors (GM). Ford had come to lead the American automobile industry in improvements. Ford's following years' earnings confirmed that its success was not a fluke, for its earnings continued to exceed GM and Chrysler's.<p>source: <a href="http://en.wikipedia.org/wiki/W._Edwards_Deming" rel="nofollow">http://en.wikipedia.org/wiki/W._Edwards_Deming</a>
Not that it's quite the same, but I can't help thinking of decimation in the Roman sense.<p><a href="http://en.wikipedia.org/wiki/Decimation_(Roman_army)" rel="nofollow">http://en.wikipedia.org/wiki/Decimation_(Roman_army)</a>
I work for a company with stack ranking. and was forced to stack-rank my team.<p>The success of this system depends on 3 things
a ) qualifications of the managers doing stack ranking
b ) perception management of the employees being ranked.
c ) quality (impact and sophistication) of the projects people are doing<p>( a ) is hard to come by in middle-management ranks (because very few talents individuals stay in those ranks -- they either move up, or move out).<p><pre><code> ( b ) perception management (essentially ability to blame others, to create inflated perception of your contribution) -- is a horrible thing to do for a technology company who benefits from having introverts who just love to program.</code></pre>
Those will often fall the victim to the 'smoke and mirror specialists'.<p>( c ) can simply be traced to poor up stream decisions/conditions.<p>---- -<p>But how to actually deal with accumulated incompetency (as it happens especially in companies that must grow very fast) ?<p>--- -<p>The answers I heard here would work but in combination<p>a) stack-rank projects<p>b) within project evaluate performance of project's stakeholders/initiators in addition to the project team<p>c) stack-rank all employees by using peer-and-outsider model that takes up 40% of the overall rank
(so that if an employee has excellent enteprise impact, but an incompetent manager, as noted earlier, -- he/she will not be totally dependent on that manager's score)...
And that also builds up better team work.
This is the "sequester" approach to layoffs.<p>Instead of deciding which projects are not paying off and cutting them, cut employees across all projects, regardless of how well they perform. Similar to the budgetary sequester backed into by the US government.
Whoa what? “Mayer, in fact, tried to address the issue yesterday morning in a Q&A, which she began by quoting her favorite children’s book about the value of experience.”<p>How do you start addressing concerns of working professionals by quoting a children's book?
I worked at IBM Research for 6 years and they have absolutely the best way to achieve the best possible productivity from <i>all</i> employees:<p>The secret?<p>"Trust Your People".<p>I was a contractor, working offsite, from another country.
I used to come to IBM Research facility once a month and we had quick discussion: "Should we add this? Ok. Let this guy (me) do that". After that I disappear for a month or two, charge IBM hourly and then present a report of what was done.<p>Result: we had the best enterprise security product on the market that actually generated plenty of revenues for the company.<p>Hats off to IBM and to SRW, JM, JK, BA, and of course DC!
Granted we were smaller team - 20 people in total or so.
Lots of new people were coming in and doing their best to deliver their best.<p>There was something spiritual about this, but it worked like a charm.
Can't believe that employees would actually complain about their work by posting "on an internal message board for anonymous feedback".<p>If it's anonymous, I have a bridge to sell you!
This is a pretty popular thing at many large companies. Everyone somehow wants to hire the cream of the crop, yet there are only so many cream of the crop candidates.<p>My own experience is that after a couple of rounds of these layoffs, most of the fat is cut from a lot of teams. After that you start knocking off productive people.<p>I'd be more impressed by someone who invigorated their workforce and then selectively cut people after they figured out who wasn't working out. Choosing a random manager assigned number that wasn't specifically requested to be a ritual offering of headcount isn't very smart. It'll bite you in the ass eventually when productive people actually end up leaving because of the work environment.
If you read "Straight From the Gut" by Jack Welch, you would know that this is standard practice that is supposed to weed out the bottom 10% every year.<p>The "GE Way" was to keep pruning from the bottom, and yes, it would mean that sometimes decent performers would get cut. In the book, there's an anecdote about how at some point, groups would take on crappy workers, just so that they could fire them for the yearly purge.
In theory, a system like this could work much better if those "bell" rankings were only forced on teams which aren't performing well (as a team). This would both encourage teamwork and lower the risk to let very good people go only because they're in an exceptional team.
In the documentary 'Enron: The Smartest Guys in the Room', it mentioned that the company had a "rank-and-yank" system, whereby annually everyone would rank everyone else. Those ranked in the top 10% would be promoted, and those in the bottom 10% would be let go. One guy commented "I don't know a successful company that fires 10% of its employees every year".
Isn't this the system Microsoft is using that has created such huge problems for the company? There was that Vanity Fair article recently that talked about what a mess the product teams are at Microsoft. It blamed the stack rankings in part for creating a hostile environment where nothing could get done. Nobody collaborates because they want to come out on top of those rankings, and some even go so far as to sabotage their co-workers to stay atop the rankings.<p>I seem to recall the article suggesting that the only place stack ranking ever worked was GE.<p>Sounds to me like a recipe for disaster.
Here's the deal. As an employee with boots on the ground, you're usually the first to see this sort of stuff coming a hundred miles away. Why anybody would put up with this BS is completely beyond me. If you can get a job at Yahoo, I'm pretty sure you shouldn't have any problem getting another well paying job.<p>I was in a pretty large corporation and they started doing this and a ton of their top developers started leaving in droves. I'm talking 10 a week, and not entry level guys. We're talking senior BA's, developers and engineers when it was at its worst. They had a turn over rate in the 90% range. After 6 months of not being able to do anything because they were in a constant training mode, a bunch of directors and middle managers got axed and they took a different approach. It takes a lot to move the needle, but when it does, it will make a difference.
So stacked ranking - why do companies not understand that the granularity level can't be too low (i.e., teams of 5 or so)?<p>Or is it just the case that stacked ranking works in theory but doesn't scale to real organizations (kind of like communism)?
Kara Swisher: She's clearly in the monkey quadrant.<p>On the topic of the chart: why would you ever work for a place why 85% are structually constrained to underperform or be average? If 100% of the empolyees start out the 15% "Target Hires", in a single generatino they have 6/7 failure rate of performing to their potential? Makes no sense if you're actually any good.
C'mon. Right or wrong, everyone knows the "no working from home" policy was a way for Y! to do a mass layoff without having it look like one to the street. This is merely an extension of the same policy. Mayer can afford to make a lot of mistakes. It's not her dime after all. So she will make them.
This suggests there's a major problem in the mid-level management that's communicating the message. Telling a front-line manager to put someone in a bucket to meet the quota results in bad feelings, poor decisions that lead to loss of talent and bad publicity.<p>The alternative doesn't have to be abolishing stack ranking. The manager's manager should instead ask for justification and help the manager raise standards if they are too low. If not, he should repeat the same process with his manager. The +-3% variance indicates that the executive team expected this to happen. Since the feedback indicates that it's not happening in some (many) cases, hopefully this has triggered top-down conversations on the right way to implement.
When it comes to employee ratings, absolute vs. relative is a red herring. Both lead to adverse outcomes. With the more common absolute ratings, managers have an incentive to rate everyone highly, so people who work for honest managers get screwed. With relative ratings, the weakest members of the strongest teams get screwed. Rating employees just doesn't work unless the departments are also rated. If you know the relative strength of each department then you can adjust the ratings of its members accordingly. There's still a serious question of what you do with that information, but at least then it won't be totally distorted by the differences between teams and managers.
It would be interesting to look at companies with stack ranking and see if any objective performance measure is a significantly better predictor of a layoff than: 1) A random number generator; 2) the personality of your boss.
It seems to me there are two problems with this system. One is that you are grading on the curve so a "less good" but still good employee is getting fired. The other is that the only person doing the evaluating is the employee's manager.<p>I would propose a system where there is an absolute, not relative, scoring method. Pick 10 criteria and score them with 1 - 10 points. Set a minimum score for continued employment, say 70 points.<p>I would also crowd-source the reviews. Not just the manager, but peers and people that report to the employee would all give a review. Then take the average review score.
What difference does it make? If your committed to reducing an organisation by X%, then a certain number of people are going to be fired. And it makes sense they are the lowest performing percentile.<p>So maybe some people were rated as "misses" when they weren't, but they were still the lowest ranked employee in their unit. Would this argument be mute if they were ranked "least effective" instead?
Wouldn't this encourage cross-department collaboration? My logic goes like this: Collaborating makes one succeed. But, you don't want your team to succeed, because then you might be laid off. But, people on other teams aren't in competition with you. So, if you collaborate with them, you can really kick the ass of that jerk next to you that you used to be friends with.<p>I'm joking...mostly.
I had put down this comment in reply to one of the thread's below...however its better that i put it as a separate comment too..for a larger audience...and in appreciation of true leaders..<p>management is one shit where the board's decision or the management's decision is responsible for a company's failure..what these guys were doing when they were hiring or buying companies...is employees responsible for the failures...<p>In any company's failure, the first one to blame is the management because they failed and that's why a lay off has to be done..I had always felt that layoff is an easy way for a CEO or the management to wash off their hands rather than trying to solve the problem at hand..<p>A leader is someone who leads his team and makes other perform...<p>In management, I have found most people just licking the bosses above them...and i hate these bunch...in such a management structure, one actually creates a bunch of followers who are unfortunately called managers..<p>There are exceptions...and they are very few...and these are actual leaders who have the power to turn things around..
For now, all the yahoo services I use stink as much as they always have. It will take years for these sweeping cultural changes to pay off, if the do. Definitely a heroic effort on Mayer's part, considering the shelf life of Yahoo CEOs is so low.
This is silly in my opinion. If she wanted to prune the comoany to make room for more (acqui-)hires do it in stages. First let the managers feel like they can more easily fire the most underperforming people. Donmt force them to do so.
I mean at this point Yahoo is an anachronism, they might as well go all the way with it. Perhaps this will accelerate the inevitable and we can all just be done with it.
Am I the only one wondering why Yahoo! is still around? I struggle to even comprehend what the hell they even do today after jumping out of the search engine race.
<i>> I feel so uncomfortable because in order to meet the bell curve, I have to tell the employee that they missed when I truly don’t believe it to be the case. [...] More often than I’d like I’m told we are executing a certain way ‘because Marissa said so.’</i><p>If you have an underproductive department with overprotective management, layoffs (and negative performance reviews) won't happen as frequently as they should.<p>Maybe members of management just want to protect their respective fiefdoms. Maybe they have their empathy makes it difficult to lay people off. Maybe they genuinely believe their teams are doing well. <i>Maybe their teams actually are doing well</i>, and exceptionally few layoffs (if any) need to happen.<p>Regardless, "because Marissa said so" is a perfectly valid answer. There will be collateral damage, but it's likely that she's dealing with some extremely protective, entrenched management who won't negatively review anyone on their teams unless they're forced to.