Back to back meetings: where a day loses its edges

A day of back to back meetings does not feel like a day of work. It feels like a day of arriving. Each call ends, the next window opens, and the ten seconds in between are spent finding the right tab rather than thinking about what was just agreed. By four o'clock the list of things that were supposed to happen between the meetings is untouched, and it is not obvious where the time went, because on paper every hour was accounted for.

Most writing on this topic argues that this is bad. That part is settled. The more useful question is mechanical: what in the calendar produces the pattern, and which specific settings stop producing it. A day packs itself because the defaults make packing the path of least resistance, not because anybody chose it.

What the research measured, and what it did not

The most cited study here is worth reading precisely, because it is narrower than the headlines suggest and more useful for it.

Microsoft's Human Factors Lab put fourteen people in electroencephalogram caps and had each of them sit through two different arrangements on consecutive Mondays. One arrangement was four half hour meetings back to back. The other was the same four meetings with ten minute breaks between them, during which participants meditated so the downtime activity would be identical across subjects.

The write up of that work, published as Research Proves Your Brain Needs Breaks in April 2021, states the conclusion flatly: back to back virtual meetings are stressful, and short breaks are the remedy.

Three findings came out of it. Beta wave activity, associated with stress, rose steadily across two straight hours of meetings and stayed flat when breaks were inserted. Frontal alpha asymmetry, a marker associated with engagement, was positive in the break condition and negative without breaks, which points at withdrawal rather than attention. And the transition between calls produced its own spike, which the researchers attributed to knowing another meeting was imminent and having to switch subject matter immediately.

Fourteen participants is a small sample and meditation is not what most people do with ten minutes. Neither weakens the mechanical point: the cost is not in the meetings, it is in the absence of a boundary between them. That is a scheduling problem, and scheduling problems have settings.

Why the calendar packs itself

Nobody sits down and decides to book six hours solid. Four defaults do it on their own.

Events default to a round hour or half hour. A conversation that needs twenty five minutes becomes thirty, and thirty minutes ending at the top of the hour leaves a gap of exactly zero. Round numbers tile perfectly, which is the problem.

Free time looks free. A gap on a calendar reads as availability to anyone with permission to see it, including yourself. There is no visual difference between an hour reserved for thinking and an hour nobody has claimed yet, so the hour gets claimed.

Invitations arrive already scheduled. An invitation is not a request for time, it is a notification of time already chosen. Accepting is one click and declining requires a sentence, so the asymmetry pushes toward acceptance.

Travel and preparation are invisible. A meeting at an address and a meeting at the desk look the same on the grid. So does a review that needs an hour of reading beforehand and a status call that needs nothing.

Each of those is addressable. None of them is addressed by resolving to be more disciplined.

Find out where the week actually goes first

Before changing anything, get the number. Intuition about meeting load is unreliable in both directions.

On a work or school Google account, Time Insights produces a breakdown for whatever date range the calendar is showing. It is worth knowing what it counts, because the definition determines whether the number is believable. An event counts toward Time Insights if you responded Yes, it has at least one other guest, it is marked Busy, and it runs less than eight hours. Events where you are the only person invited do not count, and neither do events you did not accept.

That definition is reasonable and it also excludes exactly the category most people want counted: the solo blocks that get eaten. So read the meeting figure as meetings and keep a separate eye on whether your own blocks survived the week.

If the account is a personal one, Time Insights is unavailable, and the crude version works: count the events with other guests for one week and add up the hours. The crude version is enough, because the decision you are about to make does not need a precise figure. It needs to know whether the answer is nine hours or twenty nine.

The settings that put the edges back

Change	Where it lives	What it actually stops
Shorter default event duration	Calendar event settings	Meetings that end exactly when the next begins
Working hours	Google Calendar settings, work or school account	Bookings outside the hours you work
Blocks marked Busy	The event itself	Your own time reading as free to others
Travel blocked as an event	A calendar app or an add on	Impossible transitions between locations
Invitation handling set deliberately	Event settings	Unvetted events landing automatically

Default duration. Both Google Calendar and Outlook can shorten the meetings you schedule by default, so a half hour call ends five minutes early and an hour call ends ten minutes early. Microsoft's own write up of the EEG study points at exactly this control, noting that individuals can set scheduling defaults that automatically shorten the meetings they schedule, and that organisations can set the same default for everyone. Every meeting then ends with a gap built in, and nobody has to remember to create one. This single change does more than any resolution, because it works by default rather than by effort.

Working hours. Setting working hours on a work or school account tells people trying to book with you when you are unavailable, and the hours can be split across a day so a genuine lunch shows as unavailable rather than as a convenient slot. Calendar can suggest hours based on time zone and local work patterns, and those suggestions can be edited. On a personal account this option is absent, and the equivalent is a recurring block marked Busy.

Invitation handling. How invitations reach the calendar is configurable. The options are to add everything automatically, to add only from senders who are known, or to add only when you respond to the email. Moving off the first option converts every invitation from a fact into a decision, which is the entire point.

Travel as an event. A calendar that measures the journey between two addresses and blocks it turns an impossible day into a visibly impossible day, at the point when it can still be changed. This is the one change on the list that cannot be made inside Google Calendar's own settings, because the location field is not converted into time. The travel time and time to leave behaviour in a desktop calendar exists for this, and the value is not the estimate itself but that the gap stops being bookable.

Recovery time is a real cost and it is never scheduled

Separate from the transition spike is the work a meeting creates. A thirty minute call that produces three follow up messages and one document edit is not a thirty minute commitment. It is closer to forty five.

The practical consequence is that a calendar showing five hours of meetings in an eight hour day is not describing three hours of other work. It is describing something closer to one. This is why days that looked feasible on Monday morning are underwater by Wednesday, and why the answer is rarely to work through the gaps faster.

Two habits address it without requiring anybody else to change behaviour. Schedule the follow up work at the same moment you accept the meeting, in the fifteen minutes immediately after it, rather than assuming it will fit somewhere. And when declining is not possible, shorten instead. A forty five minute meeting cut to thirty rarely loses anything except the part where everyone repeats the summary.

The three kinds of packed day, which need different answers

Treating back to back meetings as one problem is why the usual advice bounces off. There are three distinct versions and they have almost nothing in common.

The self inflicted version. You schedule the meetings, and they tile because the defaults tile. This is the easy case and the settings above solve it outright. Anyone who controls when their own meetings start can fix this in an afternoon and never think about it again.

The imposed version. Other people book the time, and the calendar is a record of decisions made elsewhere. Settings help at the margins: working hours narrow the window, invitation handling turns arrivals into decisions, and blocks marked Busy remove slots. None of it changes the volume. The volume is a conversation with whoever is generating the invitations, and the honest preparation for that conversation is the count from Time Insights rather than a description of how it feels.

The geographic version. The meetings are individually reasonable and collectively impossible because two of them are in different places. This looks like a scheduling problem and is actually a measurement problem. The calendar does not know that the 11:00 is forty minutes from the 10:30, so it accepted both, and nobody noticed until the morning of. No amount of buffer discipline fixes this, because the discipline would have to know the distance.

Deciding which version you have determines where to spend effort. Someone with the imposed version who spends a week tuning defaults will find nothing changed. Someone with the geographic version who negotiates meeting volume will still be late.

Protecting a block so it stays protected

Blocking time only works if the block behaves like a meeting, which means three things.

It has to be marked Busy, or it is invisible to anyone checking availability. It has to have a name that reads as a commitment rather than as a placeholder, because Focus is easier to override than Draft Q4 plan. And it has to be visually distinct, since a block that looks identical to the six meetings around it will get moved by the same reflex that moves meetings.

The recurring version is more durable than the ad hoc version. A block that exists every Tuesday morning becomes a known property of your week that other people schedule around. A block created on Monday for that Thursday is understood by everyone, including you, as provisional.

One caveat worth stating: a calendar full of defensive blocks and no real availability creates its own problem, because colleagues stop consulting the calendar and start asking directly. The goal is a week with visible edges, not a week with no openings.

What to change first

Turn on the setting that shortens the meetings you schedule today, because it costs one minute and applies to every meeting you create afterwards. Then look at whether the calendar knows about the trips in your week, since those are the gaps no setting can protect. Seeing how desktop calendars handle this is the next step if the day involves going places, and Caltimate blocks the journey and tells you when to leave rather than leaving the gap to chance.