My colleague Jeff Pfeffer and I have been writing about the dysfunctional internal competition at Microsoft for a long time, going back to the chapter in The Knowing-Doing Gap (published in 2000) on "When Internal Competition Turns Friends Into Enemies." We quoted a Microsoft engineer who complained there were incentives NOT to cooperate:
"There are instances where a single individual may really be cranking and doing some excellent work, but not communicating…and working within the team toward implementation. These folks may be viewed as high rated by top management… As long as the individual is bonused highly for their innovation and gutsy risk-taking only, and not on how well the team accomplishes the goal, there can be a real disconnect and the individual never really gets the message that you should keep doing great things but share them with the team so you don’t surprise them."
And we quoted another insider who complained about the forced curve, or "stacking system:"
This caused people to resist helping one another. It wasn’t just that helping a colleague took time away from someone’s own work. The forced curve meant that “Helping your fellow worker become more productive can actually hurt your chances of getting a higher bonus.”
This downside of forced-rankings is supported by a pretty big pile of research we review in both both The Knowing-Doing Gap and in Hard Facts, and I return to a bit in Good Boss, Bad Boss. The upshot is that when people are put in a position where they are rewarded for treating their co-worker as their enemy, all sorts of dysfunctions follow. Forced rankings are probably OK when there is never reason to cooperate — think of competitors in a golf tournament — or perhaps when sales territories or (for truck drivers and such) routes can be designed so that people don't need to cooperate. And there is one trick I've seen used (at GE for example) where people are ranked, but part of the ranking is based on how much they help others succeed — but people at GE have told me that forcing the firing of the bottom 10% can still create lots of problems (in fact, my understanding is that GE has softened this policy).
As my colleagues Jeff Pfeffer loves to say, the assumption that the bottom 10% have to go every year is really suspect — it assumes a 10% defect rate! Imagine a manufacturing system where that was expected and acceptable:
Well, the Microsoft stacking system is in the news again. A story by Kurt Eichenwald in coming out in Vanity Fair that bashes Microsoft in various ways, especially the "stacking system." It is consistent with past research and reports that have been coming out of Microsoft for decades — I bet I have had a good 50 Microsoft employees complain about the stacking the system to me over the years, including one of their former heads of HR.
The story isn't out yet, but according to Computerworld and other sources, this is among the damning quotes:
Every current and former Microsoft employee I interviewed — every one — cited stack ranking as the most destructive process inside of Microsoft, something that drove out untold numbers of employees. 'If you were on a team of 10 people, you walked in the first day knowing that, no matter how good everyone was, 2 people were going to get a great review, 7 were going to get mediocre reviews, and 1 was going to get a terrible review,' says a former software developer. It leads to employees focusing on competing with each other rather than competing with other companies.
To be clear, I am not opposed to pay for performance. But when unnecessary status are created, when small quantitative differences that don't matter are used to decide who is fired, anointed as a star, or treated as mediocre, and when friends are paid to treat each others as enemies, creating the unity of effort required to run an effective organization gets mighty tough — some organizations find clever ways to get around the downsides of stacking, but some succeed despite rather than because of how they do it.
The late quality guru W. Edwards Deming despised force rankings. Let's give him the last word here. Here is another little excerpt from The Knowing-Doing Gap:
He argued that these systems require leaders to label many people as poor performers even though their work is well within the range of high quality. Deming maintained that when people get these unfair negative evaluations, it can leave them "bitter, crushed, bruised, battered, desolate, despondent, dejected, feeling inferior, some even depressed, unfit for work for weeks after receipt of the rating, unable to comprehend why they are inferior.”