ManageTime

Article

You are taking the minutes: keeping the record without losing the meeting

Type the minutes, and keep four or five lines by hand for yourself. A 2024 meta-analysis of 24 studies found that typing captures far more of what was said (Hedges g of 0.919), while handwriting produced higher achievement afterwards, but only a small amount (g of 0.248) and only for people who reread what they wrote. This page is about the meeting where you were handed the record, not the one where a bot keeps it. It covers what the scribe role actually costs you, which format serves which job, and what to do when you are always the one asked.

How do you take minutes and still take part in the meeting?

Two minutes before the call starts, the chair asks who is writing this down, and the pause runs just long enough that you volunteer. Now you have two jobs. One of them is visible to everyone in the room, and the other one is the reason you were invited in the first place.

Those two jobs pull against each other, and only one of them gets looked at. The minutes are a product for other people: what was decided, who owns it, by when. Your own grasp of the discussion is a separate product, it has no deadline, and nobody will ever ask to see it. That is exactly why it is the one that quietly loses.

The research that prices this split is not about meetings at all. Flanigan, Wheeler, Colliot, Lu and Kiewra pooled 24 studies from 21 articles in Educational Psychology Review in 2024, comparing college students who typed their lecture notes against students who wrote them by hand.

The two headline results point in opposite directions.

Typing captured far more of what was said, and that effect was large (Hedges g of 0.919). Handwriting produced higher achievement afterwards, but the effect was small (g of 0.248), and it held for students who both wrote their notes and went back and reviewed them. The authors also translated that gap into grades, which is where a small effect becomes legible: on their reading, students who wrote by hand would be expected to earn somewhat better course grades than students who typed. Small and real is the honest summary.

Read those two numbers against your two jobs. Coverage is what the room asked you for, and typing is what the evidence says delivers it. Retention is what you need for your own work, and handwriting buys some of that, modestly, and only if you reread the page. The scribe is the one person at the table who has to serve both.

None of it was measured in a meeting. Every study behind those numbers is a lecture: college students, teaching recorded or live, a test afterwards. Nobody scored the minutes. Achievement was the outcome, which means this literature measures what the note-taker learned and never whether the document was any good for the eleven people who were not writing. Take the direction and leave the size alone.

The record is the easy half.

Should you type the minutes or write them by hand?

Type them. The interesting question is what you do with the hand you just freed up.

The record is a coverage problem, and coverage is the thing the evidence is loudest about. A keyboard keeps up with a room that talks over itself. A pen does not, and the gap is not close. If you are producing the document eleven other people will rely on, you are optimizing for how much of the hour survives into text, and that argues for typing with no real counterweight.

The best-known counterweight has weakened, too. Urry and colleagues ran a direct replication in Psychological Science in 2021, with 74 people taking notes on laptops and 68 writing longhand. The first half of the original finding held: laptop users wrote more words, and more of them word for word. The second half did not hold. The quiz afterwards did not separate the two groups, and mini meta-analyses the team ran across eight similar studies showed the same flat pattern. Their conclusion was that the evidence does not support longhand improving what you take in straight away.

Word count pointed the same way in both runs: people who wrote more words scored better afterwards. What hurt, inconsistently, was copying phrases down word for word, and even that link did not survive the second run. For someone whose assigned job is volume, that is about as comfortable as the evidence gets.

The keyboard wins that half outright.

So type the record, and keep something small and separate by hand. Four or five lines on the back page of a notebook, written while the discussion is still running: what was actually decided and why, what you personally now owe, and the thing you disagreed with and did not say. That strip is not minutes. It is the part of the meeting the group document cannot carry out for you, and it is the part the 2024 meta-analysis says rereading pays off on.

How the page is laid out matters more than it sounds. Bellinger and DeCaro gave undergraduates four different note-taking guides for the same recorded lecture, in the Quarterly Journal of Experimental Psychology in 2019. Two were outlines, offering a partial skeleton the listener had to fill in; two were cloze sheets, carrying the whole lecture with words removed. Students rated the outlines harder, and an objective measure of mental effort agreed with them. The outlines also produced better free recall and better inference. On questions that only asked what was said, the two formats tied. The authors read this through what they call material-appropriate difficulty: the extra work pays off, but only when the work it costs you is the kind of thinking the material actually needs.

The template that asks least of you is doing least for you. An agenda with three empty headings under it will leave you knowing more about the meeting than a form with every field pre-labeled, and the advantage in that study was largest where the guide supplied the least. That was a lecture and not a project review, with four guides and one audio recording behind it, so take it as a reason to prefer a thin template rather than proof that thick ones fail.

What do you do when you are always the one asked to take the minutes?

Nobody decided this. It accumulated. You did it well once, the chair remembered, and the role attached itself to you across a standing series without ever being named as a job.

It is a job, and the twenty minutes of writing up afterwards is the cheap part. The expensive part is attention, taken out of the hour you were invited to contribute to. Most people handed the record find the same thing happens: they stop talking. You cannot draft an accurate sentence about what somebody just said and work out whether you agree with it at the same time.

The record gets written and the contribution does not.

Ask to rotate it, and ask during the meeting rather than afterwards. One sentence does the work: you have taken the minutes for the last several of these, you would like to contribute properly to the next one, and you suggest it moves round the group. Almost nobody argues with that, partly because the logic is visible to everyone and partly because the role was never anybody's idea. If the series has a standing admin item, put it there.

Where the role cannot move, change the brief instead of the person. Minutes that record the discussion are a transcript nobody reads, and they take much longer to write. Minutes that record decisions, owners and dates are a page people actually open, and they are far easier to take while still thinking. Say which of the two you are producing before the meeting starts, and your load drops without anyone giving anything up. In practice that brief is four columns and nothing else: the decision, the person who owns it, the date, and whether anybody in the room said out loud that they disagreed. The last column is the one most templates leave out, and it is the one that saves the most argument three weeks later.

Send them the same day, and the reason is sitting in your own handwriting. A strip of shorthand written at speed stays readable for a few hours and then stops being readable at all. So does your memory of which of two people said the thing you are about to attribute. No study on this page measured send-out timing, so treat that as practice rather than evidence, though the practice is close to unanimous among people who have tried to decode a four-day-old note.

A recording does not replace the minutes, whatever the person who set it running believes. The recording holds everything, which is exactly why nobody opens it; the minutes hold the four things that mattered, which is why they get read.

Write for the colleague who missed it.

If a bot is producing a summary alongside you, that changes what you write for yourself rather than what you owe the group, and this site covers that trade on its own page. What happens to the action items over the following fortnight belongs to the meetings pillar. Anything on that list which is both yours and small is better done than minuted, which is the entire argument of the two-minute rule.

How does your working style change the answer?

Which half of the job slips depends on how you already work. The four names below are this site's own, from the working-style self-check on the home page.

Readers closest to the Architect style produce excellent minutes and the most expensive silence in the room. The structure is already in their heads, the document comes out clean, and they contribute almost nothing to the discussion they were invited for. Thinning the template helps them more than polishing it.

Sprinters have the reverse problem.

They half-type, trust that they will remember the rest, and write up two days later from a page of fragments that no longer mean anything; for them the same-day rule is the whole intervention.

Visionary-leaning readers lose the reasoning, which is the first thing a decision list drops and the thing they most need out of the hour. The handwritten strip exists for them. Improvisers are usually fine during the meeting and absent afterwards, so the write-up has to happen in the ten minutes straight after the call or it will not happen at all. Block those ten minutes when you accept the invitation, not when the call ends.

Two situations sit outside all of this. Where minutes are formal, the official record is governed by rules that outrank anything here: a works council, a board, a regulated committee, anywhere a near-verbatim account is the point. Follow the format you were handed and take the attention cost, because accuracy is the deliverable. And where a meeting decides nothing that touches your own work, the scribe role costs you nothing worth protecting, so write the minutes and skip the rest of this page.

For everything in between, the next meeting you are asked to minute is the test. Type the record, keep five lines by hand, send the decisions out before you go home. Then check whether you said anything at all during the hour. If you did not, the rotation conversation is the one worth having, and it goes better with a month of minutes behind you.

Frequently asked questions

How do you take minutes and still take part in the meeting?

Type the record for the group, and keep four or five lines by hand for yourself. The two jobs need different tools. A 2024 meta-analysis of 24 studies found typing captures far more of what was said, while handwriting produced a small advantage on what the writer retained, and only among people who reread their notes. Those studies are lectures rather than meetings, so the direction is the finding and the size is not.

Should you type the minutes or write them by hand?

Type them, because the record is a coverage problem and typing is what covers. The handwriting advantage is real but small, conditional on going back to reread, and measured on what the note-taker learned rather than on how good the document was. A 2021 direct replication with 142 participants found no quiz difference between the laptop and longhand groups at all.

How soon after a meeting should the minutes go out?

The same day, while your own shorthand is still readable and you can still tell who said what. No research on this page measured send-out timing, so this is practice rather than evidence. What the note-taking literature does show is that fast, partial notes lean on the writer to reconstruct the rest, and that is the part which fades first. Waiting until Friday means writing up a meeting you no longer remember from notes that assumed you would.

Do minutes still need writing up when the meeting was recorded?

Yes, because a recording holds everything and a decision list holds what people came back for. Nobody searches an hour of video for an owner and a date. The recording is an archive; the minutes are the usable version. An AI summary shifts what you write for yourself rather than what the group needs, which is a separate question with its own page on this site.

Sources