Unreliable is better: theoretical and practical impulses for performance management

Abstract

This review aims to stimulate discussion about a comprehensive understanding of performance evaluation – namely, the taken-for granted benefit of maximal reliable performance evaluation, where employee performance is evaluated with high levels of reliability (i.e., large samples of performance observations). So far, the management discipline has ignored the evidence-based view that one’s performance is better under unreliable performance evaluation compared to reliable performance evaluation. Drawing on tournament theory, behavioral research, and real-world sports data, we argue that while reliable performance evaluation boosts only superior employees, unreliable performance evaluation boosts all employees. The mechanisms that drive inferior and superior employees to perform better when evaluated unreliably substantiate that psychological insight is essential for efficient performance management. Overall, we complement the predominant thinking of performance management by offering innovative insights and implications that are significant for academics, employees, and employers.