Default lengths are habits, not estimates. Thirty minutes because the calendar snaps there. Sixty because that is what last week’s recurring used. The honest question is: how long should this meeting be given the objectives and the blocks you actually planned? Blackhatter compares agenda duration to the hold so that gap is visible before anyone sits down.
Time is also money when you count the people in the room. A bloated status hour with eight salaries in it is an expensive way to read a document aloud. Duration analytics exist to make that uncomfortable early, while you can still cut a block or shrink the guest list.
Add the blocks, then pick the hold
Work backwards from the agenda, not forwards from a round number. List the objectives, give each a block with a duration you would defend in front of the group, and sum them. That sum is the meeting. Round to a calendar-friendly length only after you have a real total — and only upward if you need a buffer, never downward to “make it fit.”
If the sum is eighteen minutes, book twenty-five or send a prompt instead. If the sum is seventy, do not book sixty and talk faster. Split the meeting, move Inform into a pre-read, or drop an objective. Optimism is not a facilitation technique.
Buffers are for transitions and the one decision that always runs hot, not for mystery content. Five minutes on a forty-minute Decide is reasonable. Twenty minutes of unnamed “discussion” is how Status colonizes the hour.
Match the length to the objective
Inform can often leave the room entirely. If it must stay, keep it short and pointed at a question, not a tour of slides. Status belongs in a tight loop: what moved, what is stuck, what you need. Ideate needs more time than people admit, and it fails if you also expect a decision in the same half hour.
Decide needs a packet and a framed choice. With that, twenty-five focused minutes can beat a meandering sixty. Without it, sixty minutes still will not decide; it will discover that you were not ready. Align is similar: alignment on a written draft is faster than alignment on vibes.
Problem-solve and Plan usually need more wall clock than a standup, and fewer people. Review can be short if the work is already in a place attendees could have opened yesterday. When objectives mix, the longest honest block sets a floor. Do not staple Ideate onto Decide and keep the original thirty-minute hold.
Build those blocks from aims first. The companion guide how to build a meeting agenda from objectives is the other half of this math.
Watch the hold, not just the clock
The booked hold is a contract with everyone who accepted. If your agenda overruns it, you are spending time they allocated to something else. If your agenda is far under it, you trained them that this meeting will fill itself with leftovers. Either mismatch is a design bug, visible in coverage and duration before the start.
Recurring meetings rot here. The hold stays sixty minutes after the work shrank to twenty, or the work grew and nobody changed the invite. Re-sum the blocks when the objectives change. Export a new pre-read and a new .ics when the duration changes. The artifacts should match the meeting you intend to run this week, not the one you ran in March.
Cost is a duration check in disguise
Once you know how long the meeting is, the next honest question is who needs to be there for that whole time. A ninety-minute Plan with a dozen people is a different object from a twenty-minute Decide with three. Blackhatter can estimate cost from headcount and rate so “we always invite the whole team” has a number next to it. You do not need a finance model to use the instinct: more minutes times more people is more expensive, and most Inform does not need the expensive version.
Cut length or cut the list. Both are legitimate. What is not legitimate is a hold that cannot cover the agenda you wrote. Pressure-test duration the same way you pressure-test objectives: on the page, before anyone sits down, while you can still change the design.