Based on reinforcement theory, a quasi-experimental design was used to evaluate the effect of (a) feedback obtained from (b) a relatively neutral third party (namely, mystery shoppers) that was obtained on a (c) variable interval schedule for managers to use to (d) coach their employees. An interrupted time-series design showed that both employee and organizational performance increased as a result of this intervention. Performance dropped when this intervention was cut back and, subsequently, discontinued. These results were replicated in two additional restaurants. © 2012 Wiley Periodicals, Inc.