9 April 20266 min read
When a team first works across locations, the instinct is to add meetings, because the informal contact has disappeared and a meeting is the obvious replacement. Six weeks later there is a daily call, a weekly planning call, a fortnightly review, a monthly retrospective and a standing thirty minutes that nobody remembers creating.
The problem is not that meetings are bad. It is that in a distributed engagement they consume the one genuinely scarce resource: the hours when everyone is awake at the same time. Every hour spent in a call is an hour nobody can use for a quick question, and quick questions are what the overlap is actually for.
What overlap hours are worth
Between Beirut and Western Europe the offset is an hour or two, so most of the working day is shared. That is unusually generous, and it is easy to spend carelessly precisely because it does not feel scarce. Teams working across a bigger gap ration their overlap ruthlessly because they have two hours; teams with seven hours fill them.
The discipline is the same either way. Before scheduling anything recurring, ask what it does that a written note could not. If the answer is that it keeps everyone informed, delete it: that is what writing is for. If the answer is that people need to disagree, decide, or work something out together, keep it.
The cadence that is usually enough
- One planning conversation a week, timeboxed, covering what is being picked up and any ambiguity in it. This is the meeting that prevents a week of wrong work.
- One demo or walkthrough a week or fortnight, where working software is shown rather than described. Short, and the point is that it is real, not that it is polished.
- A retrospective every month or so, with both sides present and someone empowered to change something afterwards.
- Ad hoc calls whenever a written exchange has gone two rounds without converging. No approval required; the rule is that going two rounds is itself the trigger.
That is roughly ninety minutes of recurring commitment a week for most engagements. Anything substantially above that, sustained, is worth examining rather than defending.
The daily stand-up question
The daily call is the most contested item and the answer genuinely depends. It earns its place in the first month of an engagement, when calibration matters more than anything and misunderstandings are frequent. It earns its place when the work is exploratory and changes direction often.
It stops earning its place when it becomes serial status reporting: six people each saying what they did, addressed to a manager, while the other five wait. That version is a written note with extra steps, and it costs a slot of everyone's shared hours every single day.
A reasonable rule: run it daily for the first month, then ask openly whether it is still doing anything. If the honest answer is that it is a check-in ritual, replace it with the end-of-day written note and keep the slot free for real conversations.
The meeting most teams are missing
The demo. Not a formal presentation, just someone sharing a screen and showing the thing working, with the client actually clicking through it. It takes twenty minutes and it is the single most effective way to catch the failure where something was built exactly to specification and is nevertheless wrong.
Written status reports cannot do this, because they describe the work in the same words the ticket used, which is precisely the vocabulary that carried the misunderstanding in the first place. Seeing it removes that.
Meetings to delete first
Three candidates, in order. The status meeting where nothing is decided: replace with writing. The meeting with more than about six people, of whom two talk: split it, or send a note and invite the two. The recurring call whose agenda is written in the first two minutes, which is a strong sign that it exists because it is in the calendar rather than because there is anything to discuss.
One warning. Do not delete the meetings where the relationship gets built and keep only the ones where work is inspected. An engagement made entirely of scrutiny and no conversation has a predictable trajectory, and it will not show up in any metric until someone declines to renew.
Protect the hours you have left
Once the cadence is small, defend the remaining overlap actively. Two things to try: mark a block of shared hours as a no-meeting window on both sides, and put recurring calls at the edges of the overlap rather than the middle, so they do not fragment the day into unusable pieces.
The test of whether it worked is not how many meetings you removed. It is whether a blocking question asked at eleven in the morning is answered before lunch. If it is, the cadence is right regardless of what the calendar looks like. If it is not, you have quiet calendars and a stalled engagement, which is the worse of the two outcomes.