The rebellion is taking different forms. Some, such as Accenture and Microsoft, have ditched “forced rankings” that use a bell curve to distribute employee performance, among other things. Others, such as Adobe and General Electric, have either dumped the rating scale that labels employees’ performance with, say, a “3” or “meets expectations,” or are testing the idea with groups of employees. Big banks like Goldman Sachs and Morgan Stanley have made changes to the way they rate and review workers in recent weeks.
But as the uprising gains steam, a big question remains: How good are the systems replacing them? As companies revamp reviews, what's working and what's not?
New survey data shared with The Washington Post by the advisory firm CEB reveals that one of the big changes employees have cheered may not remain so popular over time. It finds that companies that eliminate ratings altogether — replacing the “grades” workers get, whether in the form of a numerical scale or descriptive terms like “excellent” or “meets expectations” — could be in for some complaints.
CEB surveyed more than 9,000 managers and employees across 18 countries and found that those who worked for organizations that had scrapped ratings from the review process actually scored the performance conversations they had with their managers 14 percent lower. Employees who’d gotten a top score under the old ratings system missed them most, with satisfaction scores dropping even further. And among the group that had no ratings, the number of employees who believe their organization differentiates pay by performance dropped eight percent, the survey found.
"For organizations that have abandoned some sort of categorical rating type feedback, what we actually find is that experience is pretty negative," said Brian Kropp, CEB's H.R. practice leader. Without a rating to focus on in the conversation, Kropp said, managers may feel it's harder for them to deliver a clear message. The survey also showed that when companies drop ratings, managers spend less time on performance management.
Finally, employees appear to have a harder time seeing the link between pay raises and performance. When there wasn't as clear of a link between a score they got on their performance review and the size of the merit increase they received, employees' "perceptions of pay differentiation fell," Kropp said. In focus groups, "they tended to say 'my manager’s just going to give more money to the person he likes.' "
While these new results may spark caution in companies thinking of jumping on the trend, it also seems too soon to say eliminating ratings is a bad idea. After all, years of research have pointed out the challenges of grading employee performance -- how ratings can lead to manager bias, demotivate workers, and, especially if forced onto a curve, prompt the kind of internal competition that isn't helpful in a team-driven workplace.
In addition, some companies that have dropped the ratings appear encouraged by the results. Microsoft's director of global performance programs, for instance, told the Wall Street Journal in October that “the lack of rating, we have heard back from our people, mitigates the threat, distraction and internal competition." Accenture, GE and Adobe were unable to make executives available to comment for this article, but in an interview with The Washington Post earlier this year, GE's vice president of executive development and learning, Jack Ryan, said its changes, which are still in pilot mode and haven't been finalized, have led to "much richer discussions, where you're really focused on outcomes and not focused on labels."
Meanwhile, another new paper from McKinsey & Co. argues that when it comes to most average employees, differentiating pay based on individual performance may not be worth the effort. Because the top 10 to 20 percent of employees make particularly outsize contributions in today's companies, taking time to make micro-distinctions in pay for average performers simply may not have a big payoff.
"For those who meet expectations but are not exceptional, attempts to determine who is a shade better or worse yield meaningless information for managers and do little to improve performance," McKinsey's consultants wrote.
In an interview last month, McKinsey principal Bryan Hancock said the firm's "people analytics" research crunched data and found that for average performers, "ratings weren't very predictive of future performance" -- only for very top performers or very bottom ones. "It looked like a scattershot when you start to look at the data."
As a result, their paper argues that except for employees who make significant contributions -- "I'd be willing to disproportionately pay Steph Curry to shoot three-pointers," Hancock says, referring to the Golden State Warriors star basketball player -- it may not be worth the hassle to distinguish between individual performance for average performers, especially when it may amount to small sums each year.
"What we're saying is for the middle, making small changes in somebody's pay has just as much of a chance of ticking someone off as it does motivating someone in the future," Hancock said.
So what does seem to work? Kropp says their research shows that revamped reviews that get more input from employees' peers, focus on the future rather than the past and give more frequent feedback -- at least quarterly, if not monthly -- seem to help the most. "The speed of work now just dramatically outpaces the speed of HR processes," he says. "The idea of an annual performance review or a six-month performance review just doesn’t make sense any more."
Kropp does say their survey revealed places where the negative effects of ditching ratings didn't show up: When the quality of the managers in place were especially high. "If you’ve got really great managers, you can really be successful in any system," Kropp said. In such places, he says, there's a "small benefit from not having scores."
Kropp also agrees that the dissatisfaction some managers and employees are reporting in the rating-less reviews may be because the changes are new. Most companies who've dropped ratings altogether have done so in the past couple of years, so managers could get better at them over time, and employees may grow more accustomed to not having them.
The problem: Corporations tend to be impatient. Kropp says he's already hearing from companies that are thinking about going back. "Publicly traded companies aren’t known for having a very long time horizon," he says. "It’s unclear if companies are willing to wait long enough to try to get there."