In January, Shopify deleted 12,000 recurring meetings from its staff’s calendars. The e-commerce firm also reinstated a no-meeting Wednesday policy. The idea wasn’t to prevent meetings from happening, but for staff to be intentional about them. In addition, it sent a clear message that it was OK to protect one’s time.

When I ask managers what their biggest operational frustration in their job is, they typically say: “We have too many meetings.” Before the pandemic, I interviewed product managers to have a clearer sense of their roles. They described their days as running from one meeting to the next and being interrupted the rest of the time. It’s a familiar experience. You can probably relate.

Indeed, ethnographic studies on software developers, for instance, revealed how chaotic and distressing work can feel. In 1999, Harvard’s Leslie Perlow coined the term “time famine” to describe the feeling of having too much to do and not enough time to do it, largely because of constant interruptions at work.

There is a fundamental trade-off between the time that we could spend working – being productive and adding value – and the time that we spend in meetings. However, it’s not that meetings are completely useless. Teams do need them to resolve issues, coordinate work, convey information and even socialise workers. There’s a lot of research on these benefits.

However, as UCLA’s Charles J. Corbett and I found, little academic work has been done on the art and science of scheduling team meetings. We felt the subject was relevant to the field of operations management: With a long tradition of scheduling machine work, its principles could be applied to finding optimal rules for scheduling meetings.

For this purpose, we used stylised game theory to generate preliminary insights. As described in our paper forthcoming in Management Science, a starting point of our model is the need to move towards a more collective perspective. People often complain about having too many meetings, but that's because they only consider their individual perspective. While a meeting may not be useful for them – for instance if they are the ones sharing information – it may prevent a co-worker from working in vain for a week.

Our stylised model characterised the best meeting scheduling rule as a function of two dimensions: the team size and how homogenous its members are. In our context, homogeneous means that team members are similar both in terms of their needs for coordination and the value they produce.

If the team is small and workers are relatively homogeneous

Let’s imagine a team of two developers working on the same app – one is developing the Android version and the other, the iOS one. They have the same skills and, by and large, encounter similar issues. In this case, meetings should be triggered based on the workers’ needs – as opposed to being based on a set frequency, e.g. weekly.

This means that as soon as one team member wants to coordinate, they can take the liberty to interrupt the other team member(s), which is akin to an open-door policy. Alternatively, they could decide that meetings should be triggered only if both say they are available (e.g. in their status on Microsoft Teams), which is akin to a closed-door policy. This is the most fluid approach, but it can get problematic when the team grows larger or when workers are heterogeneous.

With a bigger team, the needs of its members often become dissimilar. In practical terms, the team can soon face a “squeaky wheel” problem. There will inevitably be one worker who would need help, and thus ask for meetings or interrupt their colleagues’ flow of work; and conversely, there will inevitably be one productive worker who prefers not to be interrupted, even if their input may benefit their colleagues.

When the team is of intermediate size

A few options are possible for teams of intermediate size. They can adopt an open-door policy – where everyone can interrupt anyone anytime – but it should include a provision for the minimum amount of time to be blocked off for individual work thereafter.  For example, if the team meets now, a minimum of X days should be reserved for quiet time thereafter. After that quiet time, the workers should feel free to interrupt each other.

Alternatively, they can adopt a closed-door policy where, by default, time is reserved for quiet work (production), but up to a certain limit. That is, after a predefined maximum number of days of production, the workers should avail themselves in case a coworker needs help and would like to coordinate.

Both approaches can be complicated to implement as the time safeguards would need to be calibrated to the team. But at least these options create ways to manage the squeaky wheel problem that can sap the productivity of the team as a whole.

If these options don’t work, the next best thing is to use a hierarchical structure to schedule meetings. This involves designating a worker – ideally, the one most representative of the team’s needs – as the only person who can call meetings.

What to do with larger teams

When the team is large, these need-based policies with time safeguards remain very effective, but we were also surprised to discover that a simple fixed-interval scheduling rule works really well. This means that irrespective of whether the team needs to meet or not, it should stick to a set meeting time, be it weekly, biweekly or monthly. This rigid fixed-interval scheduling rule works well because it averts the squeaky wheel issue which is more pronounced in large teams.

In our simulations, we were surprised to see that teams of ten members were already large enough for a fixed-interval rule to perform well.

Reducing the chaos with robust theory and data

When Perlow did her ethnographic work in a high-tech software firm some 24 years ago, the rules for various forms of “quiet time” did not endure after she concluded her study. But today, firms have technological solutions that enable them to track how people work and coordinate. Firms can now track workers throughout the day, see how often they check their emails, and with whom they tend to have the most meetings. That’s a lot of data that can be leveraged.

Our study allows us to build the theoretical background and make predictions about which rules would perform best under what situations. Right now, in practice, most of us work in an unguided way and tend to feel overwhelmed. Some people block time on their calendar to prevent meetings from popping up and ensure they can get some work done. This is the most fundamental time management technique. But it’s also chaotic.

Instead of setting quiet slots on an individual basis, we should synchronise these slots and those available for meetings at a collective level. With a little more data and intentionality, we can work better and reduce our individual and collective time famine.