When something doesn’t work, we usually want to fix it. Performance appraisals are an excellent example. For many organizations and HR professionals, fixing appraisals has become an enduring crusade. The hope is that somewhere out there exists the right model or formula to make performance appraisals work.

As Peter Scholtes put it: “If something is demonstrably the wrong thing to do ... you don’t necessarily need an alternative in order to cease doing it. In order to stop beating your head against the wall, you don’t need something else to beat your head against.”¹

Performance appraisals presume a fair and consistent process for evaluating and documenting worker performance. But in practice, do they measure up? Are appraisals a help or a hindrance? Do they facilitate employee motivation, or just create obstacles? Can the many concerns that exist be adequately fixed? Should the process itself be scraped? If so, what are better alternatives?

Because the process is human, it’s also imperfect. The process is about perception—a supervisor’s perception of an employee’s performance and the employee’s perception of the accuracy and fairness of the process. Folded into this process are three significant variables:

Many factors contribute to this perceptual overlay:

The performance appraisal process generally results in an overall rating or score. But there are multiple opportunities to err that can undermine this score. Unless the appraisal criteria are somewhat simplistic, such as a “go/no-go” approach to management by objectives, a narrative is needed to justify the performance assessment or score. But too often this narrative does not result in a consistent quantitative translation. If, for example, a supervisor’s appraisal narrative were to be shared with perhaps a half-dozen HR professionals, the outcome is likely to result in different scores. Is it a 3.1, 3.3, or only a 2.7 level of performance? Supporting documentation may clearly differentiate a level 5 performance from a level 2, but a more definitive graduation for grading more tightly is usually not available and, if so, hardly consistent.

Although meant to be a fair representative of an employee’s overall performance, susceptibility to one or several human errors puts the outcome of the entire appraisal process at risk. In fact, one of the major sources of errors is mistakes made by the rater.³ As described in Chapter 6, there are at least 16 rating errors. Most well-intended supervisors probably don’t know they are even making errors. Whether positive or negative, supervisors can also be perceiving performance rating as a means to justify prior interaction with employees instead of being a more objective reflection of performance over an entire rating cycle.

Just as in Garrison Keillor’s fictional Lake Wobegon, where all children are deemed to be above average, some supervisors perceive goodness in every cubicle. This tendency to subjectively give employees high performance ratings is another form of rating error. Since favorable appraisals often lead to pay increases, and there is generally a limited pool of funds available for this purpose, organizations may impose a structure to limit the number of high ratings a supervisor may award. One example is a forced distribution (10 percent at outstanding; 40 percent at above average, etc.) that limits high ratings. Another is a zero sum game with a target performance increase of perhaps four percent. Every five percent increase must be balanced by a three percent increase. Such force-fit systems hinder a supervisor’s understandable desire to reward loyal workers.

A related problem is that workers often believe they merit high ratings. In one study, 98 percent of workers believed they were above average compared to their coworkers. Another study revealed that 80 percent of workers thought they were in the top quarter. A broad study of workers across a variety of jobs showed that almost 60 percent objected to any rating that was less then the highest possible score.⁴ Thus a lower than expected rating can be destructive to an employee’s self esteem.

If a performance appraisal system were truly unbiased, accurate in measuring performance and statistically correct, 50 percent of employees would learn they are below average. Will this help them? Will this improve anything? Some who are classified as below average may resign themselves to their fate. Others will see such a classification as clear evidence that managers are hopelessly incompetent. [But perhaps the below average person will be motivated to improve performance in the next year.] Maybe they will be lucky, appear to do better, and be ranked above average, in which case their below average slot will open up and someone else will fill it.⁵

New roles for supervisors and changing workplace circumstances can make it even more difficult for supervisors to observe and accurately assess worker performance. Yet we continue to expect a great deal from supervisors, holding them responsible for administering an overloaded appraisal process that rolls together ratings, feedback, improvement goals, development, training plans, pay-raise decisions and the triggering of disciplinary measures.⁶

In the past, supervisors tended to have fewer direct reports, subordinates were at the same work location, and the supervisor’s primary responsibility was to supervise. Supervisors also tended to rise through the ranks, giving them the knowledge and experience to understand what their subordinates did and what it took to accomplish it. For budgetary and productive reasons, there has been a trend to reduce management layers. The result is often an increased burden on the first line supervisor. Consequently, the supervisor has responsibility over more people, employees may be supervised from afar (telecommuting or remote work locations), the supervisor may have little or no experience or expertise in the subordinate’s work, and the primary role is often not one of supervision but as the emerging “working” supervisor responsible for piles of administrative and related details.

Like disciplinary action and involuntary termination, the anticipation and preparation of a performance appraisal produces an emotional anguish or “angst-level.” Performance appraisal causes negative emotional states such as worry, depression, stress, and anguish on the part of those giving as well as those receiving appraisals.⁷ Writing the narrative can be an onerous chore. How can it be instructive without being too critical? Should there be as many compliments as criticisms? And what if the employee disagrees with an observation or conclusion?

As a technique to minimize surprises, the appraised employee is often asked to first complete a self-appraisal identifying accomplishments and problems of the past year. This puts the worker in the awkward position of deciding how boastful or critical to be, and how much is in his or her best interest to reveal.

Often an employee’s primary goal is to just get through the process while maintaining one’s dignity. For the more enterprising employee, he or she strives to maximize the carrots and minimize the stick and, occasionally, at the expense of other employees, use the old blame game method of deflecting criticism.

“[The performance appraisal system] devours staggering amounts of time and energy; it depresses and demotivates people; it destroys trust and teamwork and, adding insult to injury, it delivers little demonstrable value at great cost.”⁸

Organizations have to determine how frequently to conduct performance reviews. Should the review be distributed throughout the year, such as on the employee’s anniversary date? Or should it be concentrated to a few weeks each year when everyone throughout the organization is reviewed?

The anniversary date approach spreads the angst, in lower doses, throughout the year. It also allows for chronic tardiness in completing the appraisal. This necessitates retroactive pay adjustments for performance systems linked to pay-for-performance.

As an organization-wide activity requiring everyone’s attention, the annual review facilitates timely completion of reviews. It is argued that this concentrated method yields more accurate performance comparisons because all reviews are done at approximately the same time. However, the organization may be so preoccupied during the weeks of performance review that other mission-critical activities receive less timely attention. An annual review also gives supervisors an excuse not to provide timely feedback. Too often the supervisor will put off communicating a criticism until review time, either to delay an undesirable chore or to accumulate topics for the yearly appraisal session. Eliminating the appraisal takes away the excuse to procrastinate.

Proponents of performance appraisals cite the legal necessities of documenting performance. They underscore the critical need to safeguard against lawsuits, determine whom to discipline, and decide which employees to include in a reduction-in-force. They argue that the performance appraisal can provide the documentation necessary to adequately address such serious situations. However, because of problems with distorted ratings cited above and in Chapter 6, the performance appraisal itself may be incomplete or erroneous and, as a result, be contested. As detailed in Chapter 9, it may even prove to be harmful evidence. Secondly, maintaining an appraisal process for all workers in order to address the potential problems of just a few does not make much sense. Poor performers probably constitute less than 10 percent of the work force, so why incur the expense of keeping book on the entire work force?⁹

Appraisals persist as management tools for several reasons. They feed us the illusion that we are tangibly taking action to institute control, focus energies, measure performance, and bring about accountability in accomplishing desired results. Appraisal gives us documentation of people being encouraged to improve. This feels good because, by holding conversations about improvement and filling out forms, we believe we are making people accountable and getting improvement. This alluring but false impression has enabled appraisals to survive despite alternative attempts to apply new philosophical approaches to the management of workers.¹⁰

Top management wants performance appraisals to be imposed throughout the organization because it demonstrates that there is some direction and control over employee performance. Because top management is physically removed from real work and real workers, the only forms of motivation and control they have to offer anyone are externally imposed motivation and control, namely performance appraisals.¹¹

Furthermore, management believes that appraisals offer a perceived linear way to control performance. Because there often is a fundamental distrust of workers, management needs some objectives to ensure that work is being done. They have an implicit notion that they must try to fit people into their system, department, or company.¹²

The old system of offering carrots and brandishing sticks lends itself to a managerial “formula” that requires little knowledge of human behavior and less challenge than time-consuming employee one-on-one communication, interactive discussion of system improvements, and basic understanding of motivation and employee recognition. While many organizations are reaching out and stretching hard to foster the best possible climate for their employees, including assessing and reassessing how performance appraisals are conducted, many others stay stuck in conducting business as usual. In many of these organizations, performance appraisals also stay stuck, as perfunctory exercises that serve minimal value or, in fact, undercut their good intent.

One key is to establish feedback without judgment. “Judgment is the explicit or implicit attribution that the other person is right, wrong, good or bad.”¹⁴ Traditional performance appraisal is about judgment—a supervisor’s judgment of the successes and failures of a worker. Judgment is typically a one-way initiation that is often interpreted by the receiver as an act of control over the person being evaluated rather than a welcome suggestion of behavior improvement. There is a presumption that the “problem” is with the person being evaluated, not with the system that may be the root cause of the poor performance. In contrast to judgment, feedback’s purpose is to improve the system. Feedback must be in the form of data. W. Edwards Deming observed that performance appraisals became popular because it was easier for supervisors to rate workers than discover the cause(s) of the “problem.” Deming encourages supervisors to conduct a long interview with each worker “at least once a year, not for criticism, but for help and better understanding on the part of everybody.”¹⁵

“The semi-annual or annual appraisal is not a particularly effective stimulus to learning. It provides ‘feedback’ about behavior at a time remote from the behavior itself. People do learn and can change as a result of feedback. But the most effective feedback occurs immediately after the behavior.”¹⁶ Behavioral feedback (negative and positive) should occur as soon as a supervisor is aware of it.

We need to rethink and reengineer the employee’s role in the system. The goal is to shift (or at least share) responsibility and authority so that the employee is vested in his or her own professional development. This stimulates a more meaningful and accepting feedback process.

The General Motors Powertrain division (26,000 employees) created a new feedback process where the worker would “choose feedback givers from a group that included supervisors, coworkers, and subordinates.” Forms were completed by the selected feedback group and summarized by the employee’s immediate supervisor; no ratings, no scores, just narrative. Employees liked the idea of getting feedback from peers and subordinates. And it shifted responsibility to the employee to decide when the feedback would take place and from whom it would come.²⁰

“Lasting change will come only when employees take on a new view of themselves, accepting that they are adults responsible for their own growth, development, and self-worth. The ultimate goal is to help people attain an authentic and energized commitment to the organization and its goal of efficiently providing quality service and products.”²¹

We need to develop workplace methodologies and conditions (employee feedback, involvement, development, and job structure) that focus on the higher-level needs identified by Abraham Maslow. These needs, in ascending order, identify unfulfilled worker needs in most organizations.

Because it is often difficult for organizations to develop mechanisms to address these needs, they are seldom satisfied and go unfulfilled, leaving a huge void in the needs of workers. But some companies seize the opportunity to meet these needs. Entre Computer Systems designs and implements information system networks in Lansing, Michigan. In the early 1990s, it eliminated performance appraisal and pay-for-performance plans. The company reserves an amount equal to 10 percent of each of its 60 employees’ salaries for training and development. On a voluntary basis, workers meet with their supervisors each year to discuss development and how those training dollars will be used. Through this commitment to employee development, Entre is devoting resources to meet the higher-level needs of its workers.²³

There is evidence that, under certain circumstances, pay incentives do motivate improved performance. But just in the short run. Soon the effect of the incentive wears off and employee performance returns to the old “unmotivated” ways. Money can motivate when work is of little importance and dollars are an effective replacement of value, and when work is quantity driven and a good job consists of producing more of something or completing it faster.²⁶

In some situations, the traditional performance appraisal may also be appropriate. Typically, these are temporary or one-time circumstances, such as the following:

According to Alfie Kohn, “not a single controlled study has ever found that the use of rewards produces a long-term improvement in the quality of the work. In fact, experimental simulations continue to suggest that the opposite is true. Psychologists have discovered, for example, that supervisors tended to provide less informational feedback to employees and were ‘more controlling in their style of supervision when their job included administering rewards.’”²⁷

Research has demonstrated that rewards can actually have a detrimental effect on productivity, constrain quality, stifle creativity, and undermine intrinsic motivation. Kohn cites scores of studies that substantiate these conclusions in his book, Punished by Rewards: The Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other Bribes.

In the mid-1980s, the Milwaukee-based Falk Corporation, in the context of evaluating its policies and practices, decided to determine what portion of its employees were trustworthy and what portion were untrustworthy. After developing criteria for both categories, the managers assessed each employee. The result: at least 95 percent of its employees were considered trustworthy. Perhaps five percent of the workforce was considered untrustworthy. Organizational leaders concluded that their policies, practices, and procedures were written for the five percent that were considered untrustworthy. Among other changes, its 175-word bereavement leave policy was reduced to the following: “If you require time off due to a death of a friend or family member, make arrangements with your supervisor.” The offshoot: under the new policy, the total number of days used for bereavement leave was just 47 percent of the days used under the old policy.²⁸

To dump performance appraisals is a daring move. The corporate establishment is married to them, supervisors are hardened by them, and employees are conditioned to having them. But “there is no valid research to demonstrate that an organization is better off for having used performance evaluation.”³¹ “Appraisals continue because everyone else does it—it is part of the landscape of management. And it fits our linear way of thinking. In Western society, we have difficulty in looking at the circularity of things, indirect causes, and unintended consequences—we look for the direct line. If there is a problem, we tell the person he screwed up, instead of thinking of the circular and contextual causes and the design of the system. Poor performance issues are often about fit. Most performance problems are not about people who are duds, but about people who are in the wrong job or with the wrong company.”³²

“To break away from appraisals, first, companies must have the guts to break away from what everyone else is doing. Second, they must create alternative ways to deal with performance and understand performance. Foremost, they must focus on high-performance systems—the organizational culture, what the customer wants, goals, people having a voice, units having information and autonomy, and feedback from actions and the results of work units and the company. These actions will help people be more performance-oriented and help them care about performance and high standards. These strategies will make appraisals unnecessary.”³³

If organizations redirected just a portion of the time spent on “fixing” hopelessly broken appraisal systems to meeting higher-level needs, they would have much happier employees, committed to improving systems, and producing higher quality products and services. If a practice appears anti-productive, and organizations believe it’s not working, why not just stop practicing it? It’s essential to be clear about needs—and alternatives that will genuinely fill them.