It’s well known that many of us leave our jobs in large part due to a poor relationship with our direct manager.  That relationship may have deteriorated due to conflicting personalities in some way, and so the idea of working with an unbiased machine may seem highly appealing.

A review of around 45 previous studies into algorithmic management by Curtin University suggests that this expectation might not be met, however, with many people reporting as much displeasure about working under an algorithm as under an annoying human being.

Algorithmic management

The approach has been popularized by gig economy platforms, such as Uber, where drivers are largely controlled by computer systems that monitor, collect performance data, and utilize nudges as part of its “management” toolkit.

As such an approach has begun to branch out into the wider economy, the researchers wanted to understand how people feel about such “managers”.  They identified six core functions of most algorithmic management today: scheduling, performance management, compensation, monitoring, goal setting, and termination.

The results revealed that all but 4 of the 45 studies found consistently negative effects on workers, with a particular focus on reduced job autonomy, a reduction in task variety, and increased uncertainty and insecurity.  These can often have unintended consequences for the organizations deploying the systems.

For instance, the authors cite a healthcare example in which nurses used an app to tag various activities, with pay dependent upon the number of tasks completed in such a way.  The problem was that only hard tasks, such as changing a bandage, were recognized, which meant that soft tasks, such as talking to patients, were nearly completely neglected.

A reduction in autonomy is also a pressing concern, not least as autonomy over how, when, and where one works is often a key selling point of gig platforms.  The authors are at pains to point out, however, that this is not a preserve of just the gig economy, with similar coercion found in areas such as the trucking industry.

Unsecure work

Such work is also often typified by high levels of both insecurity and intensity, both of which help to create the impression of unfulfilling work.  In many studies of gig economy workers, people report feeling on edge about the pressures they face to always be available and to deliver good work even under the most intense circumstances.

This not only leads to higher levels of stress and anxiety, but can also prompt shortcuts to be taken that are dangerous for both the worker and their clients.  Many of the studies highlight the potential for work under algorithmic bosses to be extremely unhealthy, which if such forms of management are spreading should be of grave concern.

The authors argue that while current adoption of algorithmic management tends to be driven by a desire for efficiency, a better approach may be to ensure that such systems help to provide dignified and meaningful work.  For instance, workers should feel like they understand how such “managers” work and how they can be held accountable for their actions so that they can appeal to humans if they feel they’ve been wronged.