A leading supermarket chain in an eastern European Union country feared an 8-percent drop in sales as discounting giant Lidl was about to enter its market. So, in collaboration with researchers, it decided to run a randomized controlled experiment. The goal was to reduce its costly personnel turnover problem in a bid to improve quality and operational efficiency.

Selected store managers received a letter from top management, encouraging them to do something about the 90-percent yearly staff turnover. It worked: During the next three quarters, the monthly quit rate fell by 20 to 30 percent. However, surprisingly, this vast improvement led to no discernible effect on the predefined performance metrics (sales and value of perished food). In interviews, the researchers found the explanation. As store managers focused more on HR issues, they spent less time interacting with customers (to increase sales) and dealing with the flow of goods (to reduce food wastage).

 …