Article
Multitasking in a meeting you cannot leave: what it costs and which work survives
Multitasking in a meeting costs you more in how colleagues read you than in what you finish. Two 2022 experiments found that people judged a coworker working on something unrelated as breaking a norm, and judged meeting-relevant work negatively too, just less harshly. Relevance buys cover, not neutrality. This page is for the meeting you cannot get out of. It covers which work survives a half-attended hour, how to judge from the invitation whether a meeting needs all of you, and where the evidence stops. Getting out of the meeting in the first place belongs to the meetings pillar.
What do people think when they see you working through a meeting?
Laptop open, somebody else's agenda item running, and twenty minutes of expense receipts sitting there that you will otherwise do at seven tonight. The arithmetic looks settled. What it leaves out is the only part anyone has measured, which is what the other people in the meeting make of you.
De Bruin and Barber tested that in Psychological Reports in 2022, using vignettes: participants read about a coworker using a device during a meeting and rated them on whether the behaviour violated a norm, and on how capable and how considerate the coworker seemed. The first study, with 274 participants, varied whether the second task was relevant to the meeting. The second, with 188, made the task ambiguous, so raters could not tell what it was. Multitasking on something unrelated was rated as more of a norm violation, and as less considerate, than multitasking on something relevant. Work tasks read as more capable than personal ones, and an ambiguous task read as more capable than an obviously unrelated one. Then the finding that decides the hour you are sitting in: multitasking for any purpose, relevant included, was generally judged negatively.
So relevance is worth something to you. It buys a softer judgment rather than an exemption, which is a smaller prize than the phrase "I was working on something for this meeting" suggests. Softer is not clean.
Two things about that design are worth holding onto before you carry the numbers anywhere. Raters were judging a coworker in a written scenario, not one they had to sit next to on Monday, and a rating typed into a study is cheap in a way real irritation is not. The design also measures how you get read, not what you took in. Your notes are not the finding.
The objection turns up well outside a lab. Karl, Peluchette and Aghakhani analysed 549 comments posted to a LinkedIn discussion about videoconferencing in July and August of 2020, published the analysis in Small Group Research in 2022, and sorted the comments into six themes. Multitasking sat inside the meeting-management theme, in company with late arrivals, missing agendas, meetings that ran long and days booked back to back. One contributor stated the objection flatly: a person multitasking through a call is not giving it their attention, and the organizer should rule on that in the invitation. Another stated the defence just as flatly, having gone from three or four meetings a week to three or four a day, that not multitasking meant losing the work. Both describe the same calendar. Probably yours.
People who comment on a LinkedIn thread about meetings are a sample of nobody, and the authors were reading what got written rather than counting how often it happens. It tells you which objections are in circulation, not how often anyone acts on them. They are also six years old, from the months when a whole day of video calls was still new to the people writing about it.
On video the judgment runs on thinner evidence still. The largest theme in that analysis, 130 of the comments, was the camera: whether to leave it on, and what leaving it off says about you. Plenty of contributors read a dark square as disrespect. Others listed the reasons that have nothing to do with attention, including bandwidth, fatigue, kids at home and plain discomfort at being on camera. Both groups were guessing. A muted square gives them nothing else to go on. That is the position you occupy the moment you look down at a second screen, and it is why the cheapest move on this page is a sentence said out loud.
Which work survives half your attention?
Sort the work by what a mistake costs, not by how long the task takes. Cost, not duration. Neither study above measured any of this, so what follows is how it shakes out in practice rather than something anyone has tested.
Work survives a meeting when it holds no state and forgives being dropped mid-move. Filing and deleting. Approving the thing you already decided to approve last Thursday. Picking a slot from three offered times. Adding a line to tomorrow's list so it stops circling. Skimming a document you have already booked time to read properly. What these have in common is that being interrupted at any point costs you the restart and nothing else, and that nobody downstream acts on the output before you have looked at it again.
Work that does not survive shares the opposite shape: someone acts on it directly, or it needs several constraints held at once, or the error stays invisible for a week. Anything you write to another person belongs in that group, and it is the one people misjudge most often. A message costs three minutes and feels like safe work, then lands blunt or half-informed, because the half of your attention that would have caught the tone was following the discussion about next quarter. The tone is what goes.
The Deep Work page names the far end of the range: long blocks, no interruptions, for work whose value comes from continuity. A meeting is that condition inverted, and the inversion does not soften because the meeting is dull or because it runs ninety minutes. A long meeting is not a quiet block with an audience. The Two-Minute Rule is the import that does work here, since a task that fits in two minutes and finishes when you stop is roughly the size a meeting can absorb without either half suffering.
There is a running cost worth pricing before you decide any of this is free. You lose the thread, your name gets said, and you ask for the last point again. That request is more expensive than the receipts were, and it is the moment the whole calculation gets made in public, in front of the people whose judgment the studies above are about. Two or three of those and you are no longer being read as busy. The focus pillar covers the attention side of this properly; the point here is narrower, that the meeting is where the bill arrives.
How do you tell whether a meeting needs all of you?
Read the invitation rather than the room, because by the time you are in the room the decision has already been made for you. Three questions settle most of it. What does this meeting decide, does any of what it decides bind your work, and does a written record follow it?
A meeting that decides something binding on your work, with no record afterwards, needs all of you regardless of what your inbox is doing. You will be living inside that decision for a month and reconstructing it from memory. A meeting that decides nothing and produces minutes is at the other end, and your attention is worth less there than almost anything else you could do with the hour. Most invitations sit between those two, which is why the question is worth asking in advance rather than fifteen minutes in.
When the honest answer is that none of it needs you, a laptop open under the table is the wrong repair. The right one is asking to be marked optional, or to be scheduled for your item and leave after it. That conversation belongs to the meetings pillar, which sets out how to have it without spending goodwill you need later. Multitasking is what you do when that conversation has already failed, or when the meeting is genuinely one you cannot leave.
For those, say it. One sentence at the start beats an hour of inference: you have a deadline at eleven, you will be half present until your item comes up, and someone should say your name if they need you. It costs you nothing you were not already spending, and it replaces the guess people would otherwise make. The camera theme is the evidence for how those guesses go. One contributor in that study asked colleagues to assume positive intent about a camera left off, which tells you the default assumption was running the other way.
This does not survive every room. In front of a client, in a meeting called to settle a disagreement, or where a manager has already set the norm out loud, the announcement costs more than the work is worth and the answer is simply to close the laptop. Read that before you speak. The move is cheap in a standing internal update and expensive almost everywhere that anybody is watching what you do next.
How does your working style change the answer?
Which part of this goes wrong depends on how you already work. The four names below are this site's, from the working-style self-check on the home page.
Readers closest to the Architect style sort the work correctly and then over-commit to the sorting, until a recurring meeting is quietly a planned working slot and nobody in it has your attention at all. Sprinters do best at the two-minute end, where a burst that finishes inside one agenda item leaves no state to drop. Visionary-leaning readers have the worst fit of the four, because the work they want to steal the hour for is exactly the work that needs continuity, so a half-attended thirty minutes produces a paragraph they rewrite from scratch afterwards.
Improvisers are the interesting case. Following a conversation loosely while doing something else is a genuine strength here, so the cognitive cost lands lighter than it does for anyone else. The exposure just moves. What you get judged on is the part you are least likely to notice, since the room is reading you while you are comfortably on top of both tasks.
So take the next meeting you cannot leave. Write down two things before it starts: the one decision the meeting owes, and the one task you will allow yourself during it. If the decision goes past and you missed it, you have your answer, and it cost you one meeting to get. If it does not, you have found the shape of hour your calendar can afford to spend this way.
Frequently asked questions
Should you multitask during a meeting?
Assume it will be noticed and judged, then decide whether the work is worth that price. Two 2022 experiments, with 274 and 188 participants rating a described coworker, found unrelated multitasking was judged more of a norm violation than meeting-relevant multitasking. Relevant work was still judged negatively, just less so. The studies measured how people read the behaviour, not what it did to anyone's comprehension.
What work can you actually get done during a meeting?
Work that finishes when you stop and costs nothing to redo: filing, approving a decision you already made, picking a time slot, skimming something you will read properly later. Keep out anything another person will act on directly, anything needing several constraints held at once, and anything where a mistake surfaces a week later. Messages to colleagues fail this test more often than people expect, because the attention that catches tone is the attention the meeting is using.
How do you tell whether a meeting needs your full attention?
Ask what the meeting decides, whether any decision binds your work, and whether a written record follows. A binding decision with no minutes needs all of you, since you will be reconstructing it from memory for weeks. A meeting that decides nothing and produces a summary needs least. If the honest answer is that none of it needs you, the fix is asking to be optional rather than working through it.
How obvious is it when someone is working during a call?
Colleagues cannot see your screen, so they guess from thin signals, and the guesses tend to run against you. In a 2022 analysis of 549 comments about videoconferencing, the largest theme was the camera, and many contributors read a switched-off camera as disrespect. Others named bandwidth, fatigue and a child in the room. Saying at the start that you are half present for the first ten minutes replaces that guess with a fact.
Sources
- De Bruin & Barber (2022), Psychological Reports: Is electronic multitasking always viewed as a counterproductive meeting behavior? Understanding the nature of the secondary task
- Karl, Peluchette & Aghakhani (2022), Small Group Research: Virtual work meetings during the COVID-19 pandemic: the good, bad, and ugly