This post is a commentary on a Harvard Business School Case Study about performance evaluations at Arrow Electronics.

On a quick, personal, third-degree-of-separation type of note: while I have no inside knowledge of Arrow Electronics, I have had the pleasure of designing their facility here in Reno, Nevada. The Case study indicates that, at the time of the study, Arrow had “no world class manufacturing facilities” but they do now have, among office and warehouse/distribution space, 60,000 sq ft of world class manufacturing space here in Reno.

Performance evaluations are a ubiquitous presence in nearly any business environment. We have probably all received and maybe even given them. Undoubtedly and, as the case study shows, oftentimes performance evals can have unintended effects and can be devoid of real, actionable information. They may actually be counter-productive to the all-important climate of continuous improvement and teamwork in an organization.

Arrow discovered that ‘evaluation inflation’ was at play in their evals. Consequently, the CEO sent the evals back to re-done and subsequently received back new and “improved” with an artificial, forced, normal-curve distribution. I the new evals there ended up still being no differentiation of the quality of employees (which was what the CEO was looking for, he just wanted the evals to contain some useful information and be a valid and reliable tool); there was just a lowered cluster of scores. The scores were now clustered around “average” but still mostly grouped in a cluster. This made the evals essentially a useless tool devoid of any real information.

Lessons:

• Performance evals cannot be mandated to meet an artificial normal distribution, else all useful information contained in them shall evaporate.
• Managers and employees can and will learn to manipulate the system and will begin to behave in a less then genuine manner which will impede learning, growth and teamwork; this will impede an atmosphere of continuous improvement, which is necessary for businesses to function and survive long term. Performance evals can lead to much energy being spent on “impression management” rather than the real business of the business, whatever it may be.

We have all probably experienced some of the dark side of performance evals: seen them used in negative-energy generating ways such a being a tool to engender and reinforce the power of the manager over the employee as well as a tool to justify little or no annual raise (justify labor cost goals of company). This approach totally misses the point and the opportunity of some positive-energy generation that can come from an effective performance eval.

Performance evaluations should effectively steer employees toward learning and growth, toward continuous improvement.

It might be a good idea to separate compensation and performance evaluations: compensation might better be based on market values. Performance evaluations might be reserved to help to foster honest discussion, teamwork and an atmosphere of continuous improvement. There would be more time for learning from mistakes and less energy spent managing perceptions and worrying about how one mistake might look on a review that affects one’s compensation.

A question for further thought: Is it really possible to institute a performance evaluation that is entirely objective, valid and reliable or is what the performance evaluation trying to accomplish necessarily subjective and subject to human relationships?