High employee turnover remains a key problem facing many organizations across a broad range of industries. For example, despite making high salaries, 66% of senior product managers and 58% of IT program managers say they are planning to quit their jobs. Meanwhile, 60% of emergency room nurses and 58% of critical care nurses report that they are planning to quit, too—despite having already invested a huge amount of time, effort, and money on specialized training for their current job. Or, consider teachers. Nearly half of new teachers (44%) completely quit teaching within 5 years after their first day in the classroom.
Turnover is costly because, when workers quit, it can be difficult to replace them. Therefore, it’s essential to understand why workers quit, especially when it can help organizations find effective ways to reduce turnover.
While workers decide to quit their jobs for a variety of reasons, our new research has identified one trigger of quitting that seems to be a mistake on the part of workers. Intuitively, it seems like being assigned to do many “hard tasks” should make a worker much more likely to quit. Surprisingly, however, this isn’t really the case. Instead, it was being assigned to do a streak of many hard tasks in a row that really made workers quit. In other words, quitting was mostly driven not by hard tasks, but by streaks of hard tasks. This means that managers can reduce turnover by a substantial amount by simply re-ordering their workers’ tasks, so as to break up hard streaks. We call this strategy “task sequencing.”
As management scholars, the most common motivational tool that we see organizations using is monetary incentives. However, monetary incentives are not as effective as people think, and they are also expensive. In contrast to monetary incentives, task sequencing offers a powerful way to boost motivation at virtually zero cost. In our recent research published in Proceedings of the National Academy of Sciences, we found that a task sequencing intervention could dramatically reduce the likelihood that workers irrationally quit forever (in our data, by 22%). This conclusion was based on our analysis of five years of real-world data involving over 14,000 workers who were volunteer crisis counselors at a large organization. Workers in that organization were repeatedly and randomly assigned to tasks that were either hard tasks or (relatively) easy ones. A typical worker did hundreds of tasks over the course of the study, each of which was randomly assigned. This randomization allowed us to go beyond correlation and show causation (which is not the same thing). Assigning hard streaks truly causes workers to quit.
Read More: Forget ‘Quiet Quitting.’ Here’s How to Actually Set Boundaries at Work
We found that when workers had been previously assigned to do streak of multiple hard tasks “in a row” (rather than “not in a row”), it made them much more likely to quit going forward. For example, workers became 22% more likely to quit if (at some point) they had been previously assigned to do an “easy task, hard task, hard task” pattern (which contains a hard streak) rather than a “hard task, easy task, hard task” pattern (which does not contain a hard streak). Note that this behavior goes against logic: because workers knew that their tasks were randomly assigned, whether or not a worker’s hard tasks came in a streak should have been totally irrelevant for deciding whether or not to quit.
When hard streaks were longer, they caused even more quitting: for example, workers became 110% more likely to quit forever if they were assigned to do a streak of 8 hard tasks “in a row” rather than 8 hard tasks “not in a row.”
Our research builds on the “peak-end rule,” discovered by the late Nobel prizewinner Daniel Kahneman and his colleagues. They discovered that people’s evaluations of their past experiences tend to heavily overweight two moments that are psychologically special: the “peak” (the best or worst moment) and the “end” (the final moment).
Building on the peak-end rule, we proposed a new idea called the “streak-end rule.” Our new insight was that long streaks of many hard tasks in a row can create a “peak” moment in terms of a worker’s psychological experience, which is likely to have an outsized impact when workers are thinking back on their job tasks while deciding whether to quit. For example, suppose a worker was assigned to do one easy task and two hard tasks in the following order: “easy task, hard task, hard task.” Here, the second and third tasks will be overweighted because they form a hard “streak,” creating a negative “peak” or worst moment. Separately, the third task will also be overweighted because it is a negative “end” moment. In sum, workers will psychologically overweight the two hard tasks, which means that they will psychologically underweight the one easy task. As a result, the worker is likely to perceive their job as much harder than it really is, leading them to be much more likely to quit by mistake.
By applying the insights of the streak-end rule, organizations can greatly reduce their turnover rates at essentially zero cost. If organizations implemented a “task re-sequencing” intervention that avoided assigning “streaks” of multiple hard tasks in a row to any one worker—either by changing the order of certain tasks, or by reassigning certain tasks to different workers—then it would dramatically reduce the risk that their valuable workers might quit.