Research reveals advantages of scheduling retail employees with higher performers

One barely has to wade into a Reddit thread in which workers whine about their employer before encountering something related to shifts. This often focuses on with whom a worker is scheduled and how that makes things better or worse.

“Shift work is prevalent in most businesses, and yet it doesn’t really fit into any of our big pillars of HR research,” said Patrick Downes, associate professor of management & entrepreneurship at the University of Kansas School of Business.

That void in the literature led to his article, “A Relational View of Shiftwork: Co-scheduling With Higher Performers” published in Human Resource Management. The study examines how sales employees’ social contexts—particularly their being co-scheduled with higher performers—relate to changes in performance over time. Results show co-scheduling with higher performers has an immediate negative effect but is positively associated with performance over time.

Co-written with KU doctoral candidate Ella Lee, the article finds that management often views shift work only from an operations perspective.

“They think, ‘If I’m going to have 1,000 customers, then I need 10 employees to service those 1,000 customers across a certain period of time. Therefore, we just need 10 bodies scheduled on Thursday,'” Downes said.

“Our point is that when you put those 10 people in there, now you’ve created all these social interactions—actually 45 possible pairs of social interactions. So how does that affect the business?”

The study determined that if a sales employee works one week with a higher performer, then the employee’s performance will initially be slightly worse. But in the following month, the employee’s performance goes up about 2% per week.

“So if you work a full month with someone better than you, then the next month you’re going to be almost 8% higher in terms of your sales,” Downes said.

Conversely, if the same employee works with a lower performer, the results do not decline.

“One of the reasons is you don’t learn from lower performers quite in the same way as higher performers. If someone is not as good, it’s not like you can learn something from them that can be put into practice,” he said.

This study applies primarily to sales-related jobs. More interdependent shift work where employees have to rely on each other—line cooks in a restaurant, for example—does not offer the same competitive component.

“Because if you’re really good, it’s not going to hurt me if we have to team up to serve this customer,” Downes said.

His theory was tested using archival records of a large U.S. consumer retail organization. Records included monthly sales metrics and shifts worked in a single year by 7,893 retail sales representatives.

“It’s a big dataset involving several hundred stores,” he said. “We looked at every shift that every employee co-worked with somebody else. Then we went back and looked at, ‘OK, this person was a better performer than you in the past, and you spent four hours with them on this day. So how did that predict your performance a month later?'”

Downes is currently in his third year at KU. He teaches topics in HR, and his research focuses on how employees’ social contexts affect their motivation. Human Resource Management awarded Downes and Lee the 2023 Best Article Award for their study.

If the professor suddenly found himself in charge of an HR department, how would he implement these findings?

“I think particularly at scale, one of the things happening now in organizations is a big push to use data to make people-related decisions. And this is a space where the big companies and medium-sized companies can do this,” Downes said.

“If you have a few hundred shift workers, it’s worth investing in some analytical models to decide who should be working with who, and when and where.”