Skip to main content

Performance management is broken. How do we fix it?

Ultimately, senior management decides. But all of us have a stake in the system and so all of us should understand the options.

The first step, as discussed last week, is to review your current approach to gauge fairness. What part of the system is or can be perceived as unfair? Even if fair, who on your staff may be implementing it in ways that could be deemed unfair? Probably more than you have wanted to admit.

Assuming the organization wants to stay with a system – and moving away from a routinized approach can trigger its own unfairness – McKinsey & Co. offers some broad principles to follow. Consultants Bryan Hancock, Elizabeth Hioe, and Bill Schaninger looked at a host of factors that may affect employee perceptions of fairness and found three stood out:

  • Linking performance goals to business priorities: The system must be clear about what is expected from employees and specific about how their work ultimately fits into the larger picture of what the company is trying to accomplish. Employees should be given a say and senior management must be flexible. “Connecting the dots starts with making employees at all levels feel personally involved in shaping their own goals. Mandating goals from the top down rarely generates the kind of employee engagement companies strive for,” they write on the corporate website. I’d add: In an agile world, this raises questions about time frame that you must consider – an annual system may just lead to end-of-the year evaluations based on goals that are outdated.
  • Coaching by managers: Managers must be taught how to be coaches. Less than 30 per cent of the McKinsey survey respondents said their managers are good coaches. Only 15 per cent of respondents with poor coaches reported the performance-management system was effective. Before the company evaluates its staff it must improve the managers. This will be difficult, people being people. But junking the system and opting for a more informal, managers-as-coaches approach, which some organizations are trying, runs up against the same obstacle.
  • Differentiating compensation: People expect to be paid for better performance. But it’s hard "to find the right benchmarks or to differentiate among top, middle, and low performers when roles are interdependent, collaboration is critical, and results can’t easily be traced to individual efforts. The only way, in our experience, is to carefully tinker your way to a balanced measurement approach, however challenging that may be. Above all, keep things simple at base, so managers can clearly explain the reasons for a pay decision and employees can understand them,” the consultants advise.

They urge organizations not to eliminate rating systems. But they need to avoid sizeable differences in compensation among team members in the middle of the pack given the collaborative environment these days. It will lead to charges of unfairness otherwise.

If the company has forced rankings – a certain percentage of employees must be declared A, B or C players – get rid of it. Nobody considers that fair, outside the executive suite. Consultant David Dye, on the Let’s Grow Leaders blog, notes management creates contradictions as it hires great employees, but then advises a segment of them that they’re not great after all; the rankings create internal competition rather than outward competition; and it provides strong incentives to game the score rather than play the real game of serving your customer.

Another step toward fairness can be creating calibration committees of higher-level supervisors. These committees adjust the ratings supervisors give employees to improve consistency. “It may seem counterintuitive to allow a committee to adjust the ratings of employees they generally do not observe firsthand,” professors Will Demeré of the University of Missouri, Karen Sedatole of Emory University, and Alexander Woods of the College of William & Mary write in Harvard Business Review. “But even though supervisors may have better information than calibration committees about the performance of individual employees, they do not know how the ratings they assign compare with ratings given by other supervisors.’

Their three-year study of a multinational organization found the committee adjusted ratings 25 per cent of the time – decreasing them four times as often as they were increased, with downward adjustments larger than upward adjustments. Ratings were more likely to be adjusted downward if given by a supervisor who tended to give higher-than-average ratings, while upward adjustments tended to come for supervisors who tended to give lower-than-average ratings. So it would appear calibration committees improve fairness.

Assume your system is broken, or impaired. Use these ideas to push for improvement.


  • Try this question this week with subordinates: What do you know you can do well but haven’t done yet? And be prepared to act on what you hear, even if it makes things more complicated.
  • The 20/60/20 rule for leading change was originally stated by a new CEO as: “Twenty per cent of you know where we are going and are on board with it. Sixty per cent of you understand the need for change but are skeptical that we can really do this. My job is to win you over. And 20 per cent of you do not agree with our plan and have already made up your minds about it. My commitment is to ensure you a fast and graceful exit.”
  • Michael Bungay Stanier, CEO of Toronto-based consultancy Box of Crayons, says coaching need not take a long time – it should be fast, not adding to workload. And don’t try to turn managers into coaches; just aim to make them coach-like, which involves staying curious a little longer and delaying a bit in rushing to action and giving advice.

We’ve launched a new weekly Careers newsletter. Sign up today.

Follow related authors and topics

Authors and topics you follow will be added to your personal news feed in Following.

Interact with The Globe