Low Grades for Performance Reviews

Some years ago, a human resources manager at Apple Computer (AAPL) got her managers to finish their performance reviews by bribing them with free tickets to San Francisco Giants games. At North Carolina software company SAS Institute, David Russo, then head of HR, got cheers from employees in the mid-1990s for a bonfire celebration that fed appraisal forms into the flames.

Nothing has changed much since then. Managers don't like doing appraisals. Employees don't like getting them. Perhaps that's because they all suspect what the evidence shows: Such performance reviews don't work.

With the dreaded midyear review season upon us, now is as good a time as any to think about giving up this bankrupt process. The main beneficiaries of traditional employee evaluations are the consulting firms that try to improve the fatally flawed practice and the software providers that make it more efficient.

Why don't these kinds of appraisals accurately assess employees? There are so many reasons. Research conducted by Purdue University business school professor F. David Schoorman in the 1980s showed that employees hired by the person conducting a review got higher scores than employees who came to the company another way. (Managers have a greater commitment to people they have picked.)

And there are other kinds of rater bias: The evidence shows that gender and race affect reviews, with evaluations more positive for underlings whose managers share their social demographic. Ingrained expectations, about what types of people perform better, also subtly influence the process.

What's more, in jobs where work is difficult to assess objectively—in research and development, where outcomes can take a long time to appear, or in managerial positions where there is so much interdependence with others that one person's contribution is tough to discern—performance reviews mostly reflect the ability of employees to ingratiate themselves with the boss. Not surprisingly, research shows that political skill—the ability to understand others and use that knowledge to influence them— helps individuals put a gloss on their performance that ensures a higher rating.

Nor does the traditional review process help employees learn something about a better way to work. The reviews occur too infrequently to provide useful feedback. And the peer comparisons invariably create competition and discourage collaboration—a big problem in transferring knowledge in the workplace.

Besides, many of us believe we are above average and resist being told that we aren't. Numerous studies show most people think they are better than the majority of folks when it comes to skills, negotiating ability, even sense of humor. Since performance assessments often require half the staff to be rated below average, they can pose a threat to people's self-esteem. As a result, employees often discount them.

Possibly the biggest issue, however, is that performance appraisals focus on precisely the wrong thing: individuals. As W. Edwards Deming, the father of the quality work movement, taught us a long time ago, a company's performance arises more from variations in systems than from differences among individuals. One reason Toyota Motor (TM) has been so successful, as its leaders have come and gone, is that its quality management system gives it an edge.

By focusing instead on the presumed deficiencies or strengths of individuals, performance reviews divert attention from the systemic reasons, such as inferior technology, that may be behind poor results.

What should you do if you're locked into performance appraisals for now? To reduce supervisor bias, make evaluation criteria more explicit and objective and involve more people in each review. Encourage managers to have frequent, ongoing conversations with their staff about performance (as SAS's Russo did after he threw the forms into the fire). Annual reviews rely on hazy recall, with managers remembering recent events and overlooking what was done earlier in a review cycle. And don't force comparisons among people.

A recession is a good time for managers to focus more on evidence and less on received wisdom or old habits. Asking hard questions about performance management would be a good place to start.

    Before it's here, it's on the Bloomberg Terminal.