Managers who want to keep employees from quitting should consider reordering their tasks, according to a new paper co-authored by Wharton management professor Maurice Schweitzer.

In the largest field study of its kind, Schweitzer and his colleagues found that people are far more likely to quit when given too many difficult assignments in a row, compared with a workflow that is balanced out with easier tasks. Breaking up long streaks of challenging assignments may be one of the simplest ways that managers can reduce employee burnout and boost retention.

“Retaining and motivating people is really hard, and there is always difficult work to be done. The insight from this research is that we don’t want to load it all at once,” Schweitzer said. “We do better when we break it up. It shouldn’t be the case where we have one horrible day and get it all over with. Lining up a bunch of difficult things in a row is exhausting and demotivating.”

The paper, which was published in the Proceedings of the National Academy of Sciences, was co-authored by Polly Kang, a recent graduate of the Wharton doctoral program, and David P. Daniels, a professor of management and organization at NUS Business School at the National University of Singapore.

The authors analyzed nearly 2 million text conversations over five years between 14,383 trained volunteers at a crisis hotline and the people who reached out to them for help. The text conversations, which were randomly assigned to the volunteers, varied in their intensity, with suicide prevention as the hardest kind of conversation.

While the content of the conversations influenced the quit rate of volunteers, the data revealed that the order of the conversations mattered even more. Volunteers who experienced long streaks of hard conversations were 22% to 110% more likely to quit. Conversely, breaking up these hard streaks by reassigning tasks to different volunteers would “reduce volunteer quitting rates by 22%, boosting prosocial behavior and likely saving lives,” the authors wrote in the paper.

“We actually don’t know as much about quitting as we’d like to,” Schweitzer said. “The broad question is, when do people quit and when do they persist? Is there something about the nature of the work that matters? What’s interesting to me is that there is this relatively simple intervention that could dramatically reduce quitting.”

“The broad question is, when do people quit and when do they persist? Is there something about the nature of the work that matters?”— Maurice Schweitzer

The authors noted that the results would likely translate to many contexts across different professions and industries.

“There are so many contexts in which quitting is a very common problem. If we could just reduce turnover, it would improve the efficiency of an organization so much,” Schweitzer said. “When organizations think about efficiency, they think about the workflow and using the physical space as efficiently as possible when, in fact, it’s the experience of the people that’s so important.”

Peaks and Streaks

In addition to offering practical guidance for managers and policymakers who want to reduce turnover and boost retention, the study makes an important theoretical contribution to psychology and behavioral science. The scholars tested the validity of a cognitive bias known as the peak-end rule, which states that when people reflect on a past experience, they tend to overweight the most extreme moment (the peak) and the most recent moment (the end) while neglecting everything else. For example, an airline passenger will often remember a flight as bad if it involved a single extremely unpleasant moment, such as a few seconds of very rough turbulence, even if the flight was good overall — an on-time departure and arrival, smooth landing, and excellent in-flight service.

The study extends the peak-end rule by showing how it can manifest when people evaluate a sequence of discrete events. Based on their findings, the authors propose what they call the streak-end rule: When people evaluate a sequence of past events, they disproportionately focus on “streaks” (long streaks of similar events in a row) and on “ends” (the most recent event). In the specific context of crisis counselors, harder tasks will lead to disproportionately more quitting if they arrive in long streaks or if they were the most recent task.

“That’s the big idea,” Schweitzer said. “It’s not that people are doing less work or less significant work, it’s the sequence in which they are doing it.”

The study also found a flipside effect in which long streaks of relatively easier tasks made people less likely to quit, although these causal effects weren’t as large.

Schweitzer noted the natural tendency for managers to turn to the same reliable employees over and over again to get things done, especially on a deadline. But he urged those bosses to throw some lighter duties into the mix to prevent burnout and bitter feelings.

“Counterintuitively, adding a bit of extra work — specifically, adding easier assignments — can keep workers more motivated, by preventing streaks of hard tasks from being created,” he said.

“Counterintuitively, adding a bit of extra work — specifically, adding easier assignments — can keep workers more motivated….”— Maurice Schweitzer

Preventing Burnout

The idea for the study began with a visit from the crisis counseling organization, which was participating in a conference at Wharton. Schweitzer said the organization had a treasure trove of data and hoped it could be used to figure out why quitting rates among volunteers were so high.

But Schweitzer said that their results have implications for many kinds of workers, beyond volunteers. For example, nurses are a group in which burnout and turnover are common. Nurses perform both harder tasks, like providing critical care to high-risk patients, and easier tasks, like providing care to low-risk patients who are in stable condition. However, if hospitals reprogrammed their scheduling software to avoid assigning streaks of harder tasks to any one individual nurse — either by reordering patient appointments or by reassigning nurses to help different patients — it’s likely that burnout and turnover would drop substantially.

“It’s so costly to attract and retain workers that this would pay for itself quickly,” Schweitzer said. “This kind of software change would be easy to implement in a hospital.”

Even without the help of software, there are things managers can do to be more mindful of burnout.

“Rather than making Monday about five difficult Zoom conversations, we want to spread things out and have things that are less difficult interspersed with things that are more taxing,” he said. “That will sustain us and give us an overall better experience and more favorable impression of our work. We should recognize that people do need these breaks.”