Remote meetings take more time

Recent posts haven’t been the usual posts to help people googling for how to do or fix certain things. I don’t know if this is to vent, practice writing, or practice thinking (writing to think!) but hey let’s just run with it.

Today’s musing: remote meetings are longer.


My online meetings feel like they take up more time than in-person meetings, no matter whether video is on or off. My gut feels says it’s maybe 30% more time, when not timeboxed to the same duration.

There are lots of words written recently about why video calls seem to consume more energy, but why do online meetings naturally take more time than offline meetings?

Why does that happen?

I can think of at least a few contributing factors off the top of my head:

  • People talk over each other a lot more so time is spent on random backoffs and “you go ahead” negotiations. This is often attributed to fewer nonverbal cues, and perhaps remote latency too. After all, there’s more body language in person that can convey whether people are engaged and want to speak – people looking up from their laptop, turning their head, frowning at the ceiling (or even at the speaker!) that may not be transmitted or noticed with a video call.

  • It can take so much effort to get a word in that sometimes people don’t talk at all, and so someone else might explicitly invite them to talk. Yes, it’s something that you might do in person as well, but it happens more often now so that’s more time consumed by people gathering their thoughts when polled.

  • Muting your mic is a common recommendation now, so that can result in more delays as people forget about unmuting or fumble with unmuting their mic. They might even miss a speaking opportunity entirely (“anyone else has comments? …no? ok let’s move on”) and then everyone has to take extra time and energy to ‘circle back’ and reload context about a topic when they do manage to get air time later.

  • Time spent checking connectivity and troubleshooting it. This is a problem as long as the service/software/hardware involved is even a tiny bit unreliable. People have creative solutions like having a host play music from their computer to test at least one direction of the audio, but if the computer audio is a separate stream that can fail independently of mic audio…

I’m sure there are more factors, but even from just these it seems obvious that meetings will naturally take longer, all other things held constant.

If you think meetings are taking the same time, could there be a trade-off being silently made somewhere – are fewer people getting the chance to put their 2 cents in?

What can we do about it?

  • If video calls are the norm, you could develop your own conventions with some seemingly exaggerated gestures to convey nonverbal cues. I can imagine people doing hand gestures to the camera.

  • Some software have nonverbal signalling options, which you can use to collect feedback/votes out of band. But these can feel super lo-fi, especially if there’s only 2 (‘raise hand’, ‘thumbs up’) or even 5 reaction options. There are other limitations too, like Slack calls’ chat/reactions just fly by and aren’t useful at all for counting votes or ensuring that everyone is ok with something; you might have to use a regular Slack channel alongside it for proper reactjis, which is kinda annoying. Zoom has a participant list that lets you see the reaction next to each name, but the reaction choices are limited and probably work well for limited use cases.

  • You could do the facilitator thing with varying levels of process, from someone ‘going round the room’ to poll for responses, to doing the ‘raise hand’ and only speak when the chairman calls on you thing (but who really wants to do that??), just as it could be done in-person. If this wasn’t already an explicit role, then it can feel like adding an extra job for the meeting, which is not great.

  • You could encourage people to interrupt, instead of waiting for a suitable opportunity. That is a really big change though, so you’d probably have to develop other guidelines around it to really make it work – how often have you heard the “interrupt me anytime” spiel and not had it work out? (I’m not sure if I’m entirely serious about this being an option, because there are already problems around some people being interrupted/interrupting more than others, but it was interesting to see this idea being mentioned in GitLab’s company handbook.)

  • You might want people to run unmuted to reduce the barrier to getting a word in. This does depend on the physical environment though. Echo from using regular speaker/mic (as separate devices) can be dealt with by dedicated speakerphone hardware, but avoiding other types of background noise (kids in the room, noisy neighbours) is a slightly different problem. A good headset with a boom mic and/or active noise-cancellation (on the input) might be useful for those situations.

  • You could also equip everyone with hardware that has a physical mute button, to make muting easier and more reliable than fiddling with the conferencing software. There are some OS-specific software solutions for a hotkey/menubar button for mute, or a global hotkey for push-to-talk/push-to-mute.

  • Set up a process to test audio/video reliably. It could simply be having each person coming in talk to the host and ensure 2-way communication is working before they mute themselves. The most cut down “Hi” and people ack with a “Hi” has edge cases though – someone else might join and say hi simultaneously, but your mic isn’t working…

  • If you want people to come in 5 minutes early to test and troubleshoot calls, then you have to end your meetings at least 5 minutes before the top/bottom of the hour too. In a world where people don’t routinely do that, it might be better to designate the first 5 minutes of all scheduled meetings for this, regardless of how you feel about being “on time”.

  • You can have fewer meetings, that’s most likely going to reduce time spent on meetings. But this is probably a pipe dream for most teams.

I feel like these are mostly marginal improvements though, other than encouraging interruptions.

How is this going to play out?

The basic software problems can be ironed out, especially if you are allowed to change the software involved, and the communication norms will slowly develop as people try to reduce the frustrations.

But I think it will take more time to sort out hardware issues like poor or no wifi at home, or improving the mic situation. There have been supply shortages for some of these WFH-related equipment, but the larger issue is probably that almost nobody (employer or employee) really wants to invest hundreds of dollars on a situation that very optimistically can blow over in a month or two.

It’ll be interesting to see whether this situation will drive lasting changes in how we work. It does feel way too early to claim that office space is going to be cut down dramatically because teams will permanently move to 100% WFH or anything like that. After all, many homes here aren’t exactly sized with spare capacity to suddenly add multiple home-office workspaces.