Too many meetings to get work done: finding the hours back

The complaint is almost never that there are too many meetings. It is that the time between them has stopped being usable. A week with fourteen hours of meetings out of forty should leave twenty six hours of work, and it does not, because those twenty six hours arrive in pieces of twenty five minutes with a context switch at each end.

That distinction matters, because it points at a different fix. Declining meetings attacks the total. Rearranging them attacks the shape. Most calendars are broken in shape, not in volume, and shape is the cheaper thing to change.

Count the hours before arguing about them

Any conversation about meeting load that starts from a feeling loses. The calendar already holds the evidence, so take a month of it and count four numbers.

Total meeting hours per week. Not events, hours. Fifteen half hour calls and five ninety minute workshops are the same number on a status report and completely different weeks.

The largest unbroken block per day. This is the number that predicts whether anything difficult got finished. A day with five hours free in one piece and a day with five hours free in eight pieces are not comparable.

The number of gaps shorter than forty minutes. These are the hours that vanish from the accounting. They are recorded as free, and nothing that requires loading a problem into memory will fit in them.

Recurring versus one-off. Recurring meetings are the ones that compound and the ones nobody reviews. A weekly hour booked eighteen months ago is now nine working days a year.

Two of those four numbers can be read directly off a calendar with a year view and a list view, which is the honest argument for having both. A year view makes recurring load visible as a pattern rather than a series of individual days, and a list view makes the gaps countable instead of eyeballed.

Do the count before the conversation. A request to move a standing meeting lands very differently when it arrives with the number of days a year it costs.

The gaps that are already unusable

Once the count exists, the obvious target is the short gap. It is tempting to treat every free half hour as capacity, and it is the reason calendars look emptier than lives feel.

Three things eat a gap before any work starts. The end of a meeting overruns, and almost all of them do, because the default duration was chosen by the invitation rather than the agenda. The beginning of the next one needs finding, opening and joining. And in between sits the switching cost, which does not appear on any calendar and is not shorter for being ignored.

Travel is the same problem in physical form, and it has the advantage of being measurable. Apple's Calendar guide documents a Travel Time menu on events and a time-to-leave alert built from the current traffic or transit conditions and the likely starting location. It also documents two limits that explain most failures. Travel time does not update automatically when locations change, and no time-to-leave alert is given at all for destinations that take more than three hours to reach.

The starting point rule is the one to understand. Apple states that Calendar looks for a location in any event up to three hours before the one in question, and failing that uses the work or home address on the user's card in Contacts depending on the time of day, and failing that the computer's current location. So a day made entirely of video calls contains no addresses, which means the first physical meeting of the afternoon is measured from home. The estimate is not wrong so much as working from the only data available.

The practical consequence is that travel has to be written into the calendar as a block, not held in the head as an intention. A block is visible to the availability grid the next person opens. An intention is not.

Fix the defaults, not the individual events

Individual meetings are negotiated one at a time and come back. Defaults change every future meeting at once, and nobody has to be argued with.

Working hours. Microsoft's documentation for Scheduling Poll notes that suggested meeting times can be restricted to working days and hours, and that the default working hours are taken from the organizer's settings or from Exchange settings, falling back to 8 AM until 5 PM when no setting is found. An unset working hours preference is an open invitation at eight in the morning. On the Mac side, Apple's Calendar has "Day starts at" and "Day ends at" in General settings, which control what the day view even shows.

Default duration. Meetings expand to fill whatever the form offers. A default of thirty minutes produces thirty minute meetings whether or not the topic needs them.

Show as. Microsoft notes the default Show As value for all meetings is Busy, and that when participants respond their status reflects whatever the organizer set. That default is why a calendar of optional informational calls reads as a solid wall to everyone looking for a slot.

Holds. Scheduling Poll can place a tentative event on the calendar for each proposed time and remove every hold once the meeting is booked or the poll is cancelled. Used deliberately, holds protect a block. Used carelessly, a few open polls can make an entire week look taken.

The pattern across all four is the same. Every one of them is a setting that someone else's scheduling tool reads when it decides whether a slot is available. Leaving them at their defaults delegates the shape of the week to whoever opens the meeting form next.

Protecting a block so it survives

A blocked hour that other people can book over is decoration. Whether it holds depends on how it appears to the tools pointed at it, and that depends on the boundary between organizations.

	Colleagues on the same system	People outside the company
Can they see a busy block	Yes, from free and busy data	No
Can they see the title	Only if sharing permits it	No
What happens to an unmarked gap	Gets booked	Gets proposed
What actually protects it	A real event marked busy	Only a published set of open slots

The right hand column is the important one. Microsoft states that availability for Scheduling Poll comes from the organizer's calendar and the calendars of attendees in the same organization, and that any calendar not associated with a Microsoft 365 account shows availability as Unknown. Outside the company, nobody can see a protected block no matter how it is marked.

That leaves one mechanism that works across the boundary: publishing the times that are open, rather than defending the times that are not. A booking page computed from the real calendar answers the availability question before it is asked, which is also the only version of the answer that cannot be argued with. A desktop calendar that can compute free slots and publish them as a page turns block protection from a negotiation into a setting.

Stack the meetings instead of spreading them

The same total spread evenly across five days destroys more capacity than the same total stacked into two. This is arithmetic rather than preference. Four meetings, one per afternoon, cut four days in half. Four meetings in one afternoon cut one day in half and leave three intact.

Nobody schedules a scattered week on purpose. It happens because each meeting is booked into the first slot that looks empty, and the first empty slot is almost always in the middle of an otherwise clear day. The tool is optimizing for finding a gap. The person needs it to optimize for preserving one.

Two habits push against that drift without requiring anyone else to cooperate.

The first is proposing times rather than accepting them. A reply offering two specific slots, both adjacent to meetings that already exist, is almost always accepted, and it moves the new meeting to the edge of an occupied block instead of the centre of a free one. This costs a sentence and is the single highest leverage habit in the whole exercise.

The second is keeping the week's shape visible while making that decision. Free and busy grids show a day at a time, which is exactly the wrong resolution for the question being asked. A month or year view answers it: which day this week is still whole. Calendar sets that separate work from everything else help for a related reason, since a day that looks free in a work calendar may already be committed elsewhere, and a decision made from half the information gets remade later.

There is a version of this that goes too far. Stacking eight meetings into one day produces a day nobody can think in and an afternoon of overruns, and the last two meetings get the worst of the person in them. The useful target is not maximum density. It is one or two protected days a week that contain nothing at all, bought by accepting that the other days are heavy.

The reason this works where willpower does not is that it changes what the calendar looks like to the next person who opens it. A week with two clear days and three busy ones offers obvious slots on the busy days. A week of alternating gaps offers a slot everywhere, and every one of those slots costs a whole day's continuity.

Declining without making it a fight

Some meetings do have to go, and the reason this is hard is rarely the decline itself. It is that a decline without information reads as a refusal, while a decline with information reads as a scheduling constraint.

Three framings do most of the work. Offering the conflict is one: a specific reason, including travel that is genuinely blocked on the calendar, converts a judgement about priorities into a fact about the day. Offering a smaller commitment is another, since attending for the first fifteen minutes of a status meeting is often the whole of the value. Asking for the artefact instead is the third, because a meeting whose output is a written decision can frequently be replaced by the written decision.

What all three have in common is that they need a calendar the other person can reason about. A wall of untitled busy blocks produces arguments. A calendar where the reason for each block is visible at the permission level that was granted produces rescheduling instead.

What to change first

Count one month before changing anything, and record the largest unbroken block per day alongside the total hours. That single number decides whether the problem is volume or shape.

Then set working hours and default duration, and start writing travel into the day as real blocks rather than intentions. Caltimate measures door to door travel time and blocks it automatically, with a free 14 day trial and no card required.