The last 10 years are often called Microsoft’s “lost decade” as the company that once dominated the software world seemed to pay more attention to internal competitions than the marketplace. Microsoft, said Vanity Fair, had “cannibalistic culture.” It laid much of the blame on the forced ranking system the company used for employee evaluations.
But now, Microsoft has abandoned that controversial “stacked ranking” system, replacing it with evaluations that forego rankings and give managers much more control over the process. Variations of stacked ranking remain in place at Google, Amazon and Yahoo, where CEO Marissa Mayer is currently under fire for her program of tough performance reviews.
In stacked rankings, employees are assigned to a performance group, with the percentage of employees placed into each group determined in advance. So, to succeed a Microsoft worker had to actively compete with everyone around them or find work in a group where they would be a top performer, regardless of their actual skill.
“This is wonderful news for all my friends at MS,” wrote a former employee in an email. “Now they can spend time thinking about the company’s real competitors rather than one another.”
More Lost Decades to Come?
If stack rankings are really the root of all Microsoft evil, will Google, Amazon and Yahoo soon enter their own lost decades? All three use some form of stack ranking, also known as “rank-and-yank” because the lowest-ranked employees are often fired for poor performance.
That’s what happened recently at Yahoo, where between 600 and 1,000 employees were sacked after receiving two or more poor rankings.
Some workers don’t miss the goners. They say Yahoo was overstaffed with mis-hires and low performers. However, there’ve been enough complaints to force one change: In fourth quarter evaluations, the number of employees receiving the lowest ranking will be zero.
How does Yahoo know this? Because Yahoo’s stacked rankings require a certain number of workers to be placed into each of five categories, in preselected percentages. Angry managers have complained they were forced to rank some workers too low, just to meet the percentage quotas. CEO Mayer has said it wasn’t supposed to work that way.
And that’s the problem with forced rankings, experts say. They make some good employees look bad, even as they weed out poor performers. As the rankings are repeated, it can become harder for employees to retain their position, allowing them to slide into a performance abyss over time.
That, in turn, incentivizes employees to look after themselves first, battle coworkers for position second, and think about company goals third.
Analyst Rob Enderle has spent his career as a professional tech industry watcher and believes forced ranking does much more harm than good. “I think this hurt U.S. industry in general,” he says. “This spread from company to company thanks to GE and [its former CEO Jack] Welch, and if you look at the performance of the firms that implemented it, in most cases, it fell off sharply as a result.
“In terms of making internal collaboration difficult, stacked ranking is the leading cause of why collaboration generally doesn’t work and why so many firms have unnecessarily hostile work environments,” he continues. “It creates a them-or-me culture where backstabbing is effectively encouraged by the compensation plan.”
Forced ranking was a “business fad,” according to Cliff Stevenson, senior human capital researcher at the Institute for Corporate Productivity (i4cp). “It seems like it’s been here forever, but it really started in the 1980s.” Used by 40 percent of companies in 2009, an i4cp survey found only 14 percent were using it in 2012.
GE’s Welch is largely credited with popularizing stack rankings, which he used to place workers on a corporate “vitality curve.” During the ‘80s, these forced rankings helped Welch to annually fire the lowest-ranked 10 percent of his executives, regardless of absolute performance.
Microsoft’s About Face
Microsoft’s change in heart came after a series of unflattering published reports described how for many years its ranking system was kept a secret and how it turned Redmond inward-looking as employees sought to protect themselves.
Even simple forced ranking, if made public, can cause trouble. In an email to Dice News, a corporate conflict resolution specialist spoke of a client who was told to choose the four worst performing team members. “Extreme competitiveness and being unwilling to work as a team were the feelings my client shared with me. And she wasn’t even in the bottom four,” the writer said.
Meanwhile, some observers say employee evaluation systems have a major effect on employee morale and, ultimately, corporate success or failure. “They play a pretty big role in the success of a company because they give you a roadmap to show you how rewards should be given out,” said Deb Keary, vice president of human resources at the Society for Human Resource Management.
“I don’t like forced ranking because you must put somebody at the bottom. If you do it once, that’s OK, there’s a person at the bottom. But if you do it again, there is a different person at the bottom. What happened? Did they go bad? At some point you’re cutting into people you need to keep,” she says. As for today’s top performers, “In five or 10 years, they’ll be out, too. Is this what you want to do? I really don’t think so.”
Not everyone agrees that stacked ranking are universally awful, just mostly.
“I think forced/stack ranking can work in some companies, those with a very aggressive culture that are able to fire 10 percent each year,” says Tom Armour, co-founder of HighReturnSelection, an HR consulting firm. “In my own experience, forced ranking is not conducive to organizations that rely on teamwork, innovation and creativity. It’s hard to force people to be more creative and innovative with a stack ranking tool.”
Armour believes tech and knowledge based companies usually thrive when their environment is collegial and innovators work in teams sharing their ideas. “Forced raking produces the opposite, putting employees into competition with each other and having them withhold information and stop sharing.”
Still, forced ranking can be a useful tool, though only when managers do it in their heads and keep results secret. Keary, for example, has a large staff and keeps a mental list of her employees’ performances. “But I wouldn’t share that with a living soul because people don’t need to be disheartened and it wouldn’t help the strength of my team at all,” she says.
Another point against forced ranking is that it can do away with the average performers that actually get most of the work done. While top-performing “stars” lead corporate growth, implementation generally falls to others. Not everyone has to be a top performer to be effective for the company.
“Everybody is not a genius,” Keary concludes. “Everybody is not a 5. This is not Lake Wobegon. All the children are not above average.”