"When I think of forced ranking, I think of a specific process where it's really across the entire organization, where everybody is against everybody else," said Chou.
Critics of Microsoft's performance management, including past employees and managers, often cited the unfairness of the practice, saying that they were forced to put some team members into the lowest-ranking bucket, even though they considered them superior to even the best in other teams.
"Think of the situation applied to Harvard's students," said Chou, coming up with an example. "Someone at Harvard has to be the worst Harvard student, but that person is probably still better than those at most public institutions. But [in forced ranking] someone has to be at the bottom."
But the belief by managers that their employees are better than those on other teams is not only widespread, but natural and understandable, said Chou.
"Managers know their own employees better than those [outside their teams]," said Chou. "In many cases, they've hired them personally. So it's natural for them to think their own [workers'] functions are the most important, and thus their own [personal] functions are the most important.
"In forced ranking, there's always a tendency to rate more of their employees above average," Chou added, referring to managers' evaluations. "If it goes unchecked, more and more will be above average."
And then what's the point of an evaluation when, as in Garrison Keillor's Lake Wobegone, "all the children are above average?"
In fact, forced ranking and/or forced distribution practices are sometimes implemented by companies to dampen that tendency, said Chou, to instill what he called more management "discipline" in evaluating employees. Some companies implement it thinking that it will be only for a while, and that once they feel like the performance evaluations are where they should be they will taper off the dog-eat-dog practice.
Chou called forced ranking "on an extreme" end of management philosophies, and said that was one reason why it was losing favor. "Some of those people [who once used it] are now saying that the benefits they got from it were outweighed by the cultural cost," said Chou.
Stack or forced ranking is not inherently bad, or as some have claimed, evil, Chou countered. Its success, like any practice, requires managers to "buy into" the practice. "Every [evaluation] method has advantages and disadvantages. But it's important in most organizations to identify those who are really moving the needle," he said, and thus reward them.
"We do see evidence that its popularity is waning, and see more companies moving away from it or thinking about moving away from it," said Chou. "In Microsoft's case, apparently the cost of doing this outweighed the benefits they thought they were getting."
Sign up for Computerworld eNewsletters.