Calendar discipline: protecting time that stays protected

A block on a calendar marked 'focus time' loses almost every contest it enters, and the reason is structural rather than a failure of resolve. To anyone else looking at your calendar, that block carries no information: they cannot tell whether it represents work with a deadline tomorrow or an aspiration you set three weeks ago and have overridden twice since. Faced with a real meeting request and an unexplained block, a reasonable colleague books over the block, because the request has a named purpose and the block does not.

The second reason is that you are usually the one who moves it. A focus block has no other attendee to disappoint, which makes it the cheapest thing on the calendar to sacrifice when something arrives. Every time you move it, you also teach yourself that it is movable, and by the fourth time it has stopped being a commitment and become a note about your intentions. Protected time survives only when overriding it costs something, and a block whose only defender is your own resolve at the moment of interruption costs nothing at all.

The fix is to stop treating all blocked time as one category. Genuinely immovable time is tied to an external consequence: the filing due Friday, the proposal a client is expecting, the prep for tomorrow's board conversation. Negotiable time is real and valuable but survives being moved to Thursday — reading, thinking, general catch-up.

These should not look identical on a calendar, because they are not the same commitment. Name immovable blocks after their actual deliverable rather than the activity: 'GST return — due Friday' tells anyone reading it exactly what they are displacing, and tells you why moving it is expensive. Leave negotiable blocks named loosely, and move them without guilt when something genuinely better arrives; that is what they are for. The discipline is in being honest about the ratio. If everything on your calendar is marked immovable, the distinction has collapsed and nobody — including you — can tell which blocks matter, so all of them become negotiable again by default.

A calendar with meetings scheduled directly against each other has no capacity to absorb ordinary overrun, so the first meeting that runs five minutes long makes you late to everything after it. The visible cost is lateness. The larger and less visible cost is that no meeting in a back-to-back run gets any follow-through: the actions you agreed to are never written down, because the moment the call ends the next one has already started.

That is where most meeting output actually disappears — not in the meeting, but in the ten minutes after it that never existed. A meeting whose decisions are never recorded generates a second meeting to re-establish what was decided, which is how a full calendar reproduces itself. Booking meetings for fifty minutes rather than a full hour, or leaving one gap after any conversation that will produce work, is not a productivity flourish. It is the difference between a meeting that ends in actions and a meeting that ends in a vague memory and a follow-up invitation.

Most calendar invitations are not demands but proposals, and they can be answered as proposals. The move that preserves the relationship is to decline the format while engaging with the purpose: 'I can't make Tuesday, but if this is about the pricing question, I can answer it by email today' gives the sender their outcome faster than the meeting would have, which is difficult to be annoyed about.

When you cannot help, say so plainly and early rather than holding the invitation as tentative for a week. A prompt no lets someone plan; a slow maybe followed by a no costs them the ability to arrange an alternative. If the meeting is genuinely necessary but not now, decline with a specific date attached rather than an open-ended deferral — 'not this month, but let's do it in the first week of October' is a real answer, while 'let's find time soon' schedules nothing and quietly makes it your job to remember. Where you are declining something a senior colleague called, offer the alternative rather than only the refusal.

The audit is deliberately crude, because an elaborate one does not get done. For one week, at the end of each day, write beside every block what actually happened in it: did it occur, did it occur as scheduled, and did it produce the thing it was for. Three columns, no scoring system, five minutes a day.

At the end of the week, four things are usually visible. First, how much of your blocked time survived contact with the week, which is the number that tells you whether your calendar is a plan or a wish. Second, which specific kinds of request displaced it, because it is almost always one or two recurring sources rather than random chance. Third, how many meetings produced no recorded outcome. Fourth, how much genuinely unscheduled time existed, which is usually less than people assume and is the buffer that absorbs everything unexpected. None of this needs a tool. It needs a week of writing down what happened, which is the part that gets skipped in favour of redesigning the system again.

A calendar records intentions and appointments. It does not record capacity, and confusing the two is the most common way a technically achievable week turns out to be impossible. Nothing on a calendar shows that Wednesday's two-hour block follows a difficult negotiation and will therefore be worth substantially less than the same block on Monday morning. Nothing shows that a task estimated at an hour has never once taken an hour.

This is worth stating plainly because it bounds what any scheduling method — or any tool — can do for you. A calendar can tell you that two commitments overlap, which is a fact about the grid. It cannot tell you that six non-overlapping commitments amount to more than you can actually do, because that judgement depends on how demanding each one is, how well you are recovering, and how accurate your estimates have historically been. Those remain your assessment. The useful habit is to treat a full-but-legal calendar as a claim to check rather than a plan that has been validated by the fact that nothing collided.