A recent issue of Knowledge@Wharton indicates, that while 91% of companies worldwide have performance appraisals only 35% to 40% do performance reviews well.  The question remains, what does doing them well mean?  What are the criteria for the performance of performance reviews?  Can performance reviews (as we’ve come to know and love them) really be done well?  Should they be done at all?  If not what should be put in their place?

Before we address these let’s understand the basics.  Since performance is an emergent property of a system, it is a function of the interaction of the many constituent components of the system as well as the components themselves.  So to argue that maximizing everyone’s performance leads to maximum organizational performance is wrong. To illustrate the performance of person A in a two-employee company would be equal to A’s contribution to performance plus the contribution from the interaction between person A and the system.  So you basically have one equation and two unknowns; and this can’t be solved.  In other words, an individual’s performance (algebraically speaking) cannot be separated from the system itself—check with a high school algebra book for confirmation.

As you increase the number of employees the complexity of this increases almost exponentially.  The point of this is though you may have a number you claim reflects someone’s individual performance it is a false claim.  I am sure many have observed or even experienced firsthand situations wherein a person is given either credit or blame for something he/she had nothing to do with, other than being proximate to the event.

Making the practice of performance reviews even worse, many rank order people (best to worst) as a means (cloaked in the presumption of numeric objectivity) of doling out merit increases.  Many know the practice that managers often use when they are told (in so many ways if not explicitly) that they have say a fixed 3% merit budget that they are to distribute according to the best to worst rankings.  So this necessitates a forced distribution allowing managers to only score a small percent of employees at the top rating.  With the absurdity of it all evident, most employees know that no matter what they do they are up against it.  Hence the forced distribution decreases commitment to learning/improvement and diminishes cooperation among employees. Moreover the interaction term in the above equation will likely be negative—decreasing overall performance—the more there is fear and competition within the system.

Making the performance review practice worse, people are ranked within a department based upon a performance metric and for the unlucky bottom 10% either notice is served or they are fired straight away.  After all who wants bottom dwellers in their department/organization!  If the same thinking is applied to the collection of all managers in an organization—any organization—unavoidably there would be a bottom 10% and about half would be below average!  What’s being done about this?

With only a modicum of understanding of the theory of variation almost everyone will see that within a given group of people there will always be a bottom 10% and half of them will be below average—these are statistical facts that are just as true as the force of gravity.  Oh, what a basic understanding of the theory of variation would do to rid us of these misguided managers and their common practices.

Clearly with this way of managing feedback isn’t really feedback that is part of a process of learning in support of helping people improve it is a justification process for misguided practices. That is, feedback will tend to be what the manager needs to say to justify the rating and ranking given consistent with the forced distribution.  Such a practice does more to identify those in management who haven’t a clue, than it does to accurately identify people in need of help.  These managers believe it is possible to (eventually through enough iterations of the annual review process) to rid the organization of below average people.

So with all the negative experiences people have with performance reviews as well as the fact that they really don’t help people and they don’t afford an assessment of performance, why is the practice so widespread?  The usual answer is we need a means of awarding merit increases and identifying those who should be let go.  No wonder people abhor the performance review!   No wonder reviews are absent of honest, helpful and productive feedback.  As professor Samuel Culbert, of UCLA’s Anderson School of Management, asserts, “they destroy the trust between the boss and the employee, and cost the company enormous amounts of money in terms of time and wasted effort.  The people being reviews worry about pleasing their boss before they concern themselves with delivering results to the company.”  When the review process is about ranking, sorting, labeling and rewarding employees—and not about facilitating the actualizing of potential, looking good trumps doing well.

As Deming asserted the annual performance review must be replaced by leadership.  If a person sees that his/her manager actually cares and is most interested in coaching for improvement, then why would anyone not commit to learning toward actualizing his/her potential and improving his/her performance?  When the feedback is for coaching and is facilitative toward learning to improve and not used for ranking, categorizing, labeling and reward/punishment—a means of accountability—then it turns from being hurtful to helpful.

When feedback becomes an essential part of the plan-do-study-act continuous improvement cycle—which must be part of everyone’s job, even management—then you have the making of a system for improvement of quality.  Isn’t this what managers should be focused on and seek to create, rather than measuring, ranking and sorting people for merit and demerit?