Kill the Curve - Poor 50's HR Practice of Stack Ranking Continues

In the 1950's industry became obsessed with scientific management. People became human resources and performance management was created to "motivate" these resources. A lot has changed since then. The speed of innovation now requires collaborative employees rather than resources mechanically following process. Science has also moved on in understanding how to get the best from these people. Why does this 50's thinking prevail?

Scared or Lazy Managers

Too many executives and managers are either too lazy to learn; or choose the comfort of the status quo that gave them their power, instead of embracing the personal uncertainty that comes from adopting the new mindset of modern successful leaders. If your organisation truly values its people. If it wants to attract and keep top talent. If it wants employees to be more effective, then leadership and HR needs to kill the thinking that labels people as resources and kill one of the most destructive processes mistakenly implemented by HR. The bell curve approach to performance management.

History of Performance Management

Performance Management can be traced back to the 1950's when WD Scott an Australian management consulting firm (acquired by Coopers and Lybrand in 1985), first devised the approach. Performance Management was a personality based system for measuring a persons working performance. 

The Chartered Institute of Personnel and Development now defines  Performance Management as a:

  • communication of a vision to all employees
  • setting departmental and individual performance targets that are related to wider objectives
  • conducting formal reviews of progress towards these targets using the review process to identify training,
  • development and reward outcomes
  • evaluating the whole process to improve effectiveness
  • expressing performance targets in terms of measurable outputs, accountabilities and training/learning targets
  • using formal appraisal procedures as ways of communicating performance requirements that are set on a regular basis
  • linking performance requirements to pay, especially for senior managers

Bad Impementation

Unfortunately the implementation of this has lead to the creation of the annual appraisal. A process where you meet with your boss once a year and are "awarded" a grade from 1 to 5 describing how well you performed. The fun begins because the number of 1's through 5's that can be awarded are pre-decided by a formula used by HR. The formula used is based on the curve below, sometimes known as a bell curve because it's shaped like a bell: 

Performance Management Bell Curve

Performance Management Bell Curve

This means that your manager can give lots of 3's, then less 4's and 2's and finally even less 1's and 5's. This is all based on an old belief that a group of people's performance follows a normal distribution. The implication is that the average performance of an employee is in the middle range of this curve - the belief then being that most people are average performers.

Debunking the Myth

The HMRA Research Briefing paper describes more recent research in to why the bell curve for performance management is wrong. Showing that most people under perform the average. This is due to a very small number of "rock stars" or hyper performers making a massive difference. They shift the average performance of the organisation significantly to the left. This gives us a power curve where the majority will reach below average performance:-

Performance Management Power Curve

Performance Management Power Curve

This shows that even from a statistical perspective the bell curve approach to performance management is wrong. However it's more than wrong, it puts team members in direct competition with each other for their annual grades. This at the same time as we're saying that we want collaboration.

Where as when only a small number of people in the team can get the top grade we encourage people to:-

  • Protect knowledge rather than share, creating key person dependencies and therefore additional organisational risk
  • Avoid helping others who they'll be compared against
  • In the worst case knowingly allowing others to fail

All of these issues are even more likely in organisations that link the annual grading to a reward system. This can be brutal. I speak from experience having managed in this environment. 

What a Waste of Time

There's also the amount of time the performance management process takes. Managers going through the process have to endure calibration. For those lucky enough not to have experienced this, it's where managers get together with HR and they agree what's a grade 1 and what's a 5. They do this so that not only does each team have grades awarded using the bell curve but that across all of the teams the awards match the bell curve!

This process can take days to weeks of management time across large organisations. Then comes the job of telling people their grades and dealing with the fall out of disappointed people and  those appealing against their grade. There's considerable wasting of management and employee time in this process. Management time much better spent hiring better talent, managing risk, coaching teams and continuously improving the systemic issues.

When I worked at Skype it was acquired by Microsoft. The bell curve model was used in Microsoft and introduced in to Skype. Lisa Brummel, then Microsoft's Chief People Officer, visited the Skype offices to see how Skype worked. At the time all Skype teams were some level of Scrum. She saw the commitment that team members made to each other every day to make progress. She saw how they collaborated together to remove impediments. She saw how they didn't wait to the end of the year to address issues. Microsoft wanted to harvest the best of Skype's market responsiveness and Lisa saw how the bell curve performance management system was discouraging these exact behaviours. In 2014 Microsoft took the bold move of killing the bell curve. Although not the only reason, I believe Lisa's Skype experience was a significant factor in the decision.

In Closing

There are still a large number of organisations that continue with this 1950's bell curve performance management system. In a mistaken attempt to motivate us to be better employees. At the same time there's a mountain of modern research showing that not only is the math wrong but that it's also in direct conflict with the modern behavioural science on increasing organisational agility.

Organisations that want high performing employees should follow the science and other successful companies and kill the bell curve with immediate effect. R.I.P.