Performance management is a favorite target of critics from the bookstore to the blogosphere. Calls for abandoning performance appraisals are shrill, but companies are not listening. A recent study by my colleagues Ed Lawler, George Benson, and Michael McDermott found that every one of over 100 large companies studied had a performance management system – the same result as 10 years earlier.
As long as managers have limited resources, they will need to determine who gets more of the available rewards and who gets less -- or none. There is simply no alternative to sound performance management for making personnel decisions on matters such as pay raises, promotions and developmental opportunities.
That said, performance management practices are evolving. We see three types of changes that are especially promising.
Complex rating schemes are still common, enabled by software that forces supervisors to rate employees on multiple factors and instantly tabulates results. One defense contractor uses a 150-point rating scheme (15 factors with 10-point rating scales). The system is a monument to false precision. Managers cannot defend the differences in ratings that take so much time and energy to generate. Moreover, complex ratings make little sense in an era of 3% average annual pay increases. The differences in ratings do not lead to meaningful differences in consequences, yet they foster ill will among employees.
In response, many firms are now radically simplifying their ratings. Validity increases as ratings become fewer and simpler. Firms may require only a single summary score, often on a three-point rating scale. Among the many labels for points on the scale, my favorites are “Walks on Water,” “Swims,” and “Sinking.” Supervisor-subordinate conversations can focus on what did employees did, how well did they did it and what can they do to improve next year – rather than on complex ratings. Moreover, meaningful rewards can be preserved for top performers. For example, one engineering company grants raises that automatically are twice as high for those in the top category as those in the middle category.
A supervisor with a handful of subordinates is not in a position to make valid judgments about how subordinates are performing compared to others in the firm. As a result, it has become commonplace to “calibrate” ratings across large groups, typically of 50 to 100 employees. This is done in meetings with supervisors and managers in a department or unit. Supervisors propose a rating for each subordinate, which the calibration group accepts or modifies. No ratings or consequences are announced to employees until the supervisor ratings have been reviewed systematically in a calibration session. Managers typically find that the process gives them a deeper understanding of the strengths and weaknesses their talent pool.
However, calibration sessions can be time-consuming and emotionally difficult; one manager calls them “gladiator combat.” Therefore, companies are experimenting with the process. For example, some calibration sessions focus on every employee, others only on those who may be in the top and bottom categories. Employees may be told nothing before the calibration session begins, or may receive a preview of their supervisor’s position. A few companies do calibration quarterly or semi-annually, while most do it annually in conjunction with the annual appraisal process. Some companies use calibration to determine performance ratings while leaving supervisors to determine final consequences (pay, promotion, dismissal, etc.); other companies use calibration to determine consequences. Best practices are not yet clear.
Perhaps the most innovative trend in performance management is the incorporation of real-time peer feedback. This makes use of “crowd-sourced feedback” that is built into on-line recognition systems such as those from Globoforce, Achievers, and Work.com. These systems create the social media version of “360 degree feedback” systems, permitting employees to recognize each other any time, in their own words, over any computer or mobile device. Globoforce advocates recognition systems in which 70% to 80% of employees give and receive feedback every year. The beauty of the online systems is that supervisors can tap into the database of comments and review all feedback given to a subordinate in the prior year. This helps the supervisor recall the employee’s accomplishments and rounds out the picture of how well the employee performed.
We are planning to conduct the first academic research on crowd-sourced feedback systems. These systems raise many important questions. How widely and how often is such feedback provided, given the goal of continuous, real-time feedback? Do the recipients accept the feedback as valid and meaningful? How do they react to peer feedback, both when it is consistent with feedback from the formal appraisal process and when it is not? Do these systems create the “culture of recognition” that they intend, and if so what are the consequences for organizations? What are the effects on team perception and performance?
Performance management is here to stay, but it is getting more interesting. Companies are experimenting with various ways of making it more valid, fairer, more useful and more cost-effective.
Gerald Ledford, Ph.D., is senior research scientist at the Center for Effective Organizations (CEO), Marshall School of Business, University of Southern California. He is an authority on human capital issues, with over 30 years of experience as a researcher and consultant. He received a Ph.D. and M.A. in psychology from the University of Michigan. Gerry has authored over 100 articles and 10 books. He can be reached at gledford@marshall.usc.edu.