How to find the best time to schedule a meeting across time zones

Posted 24 August 2026.

Start by writing down everyone’s working hours in one zone

The reason cross-timezone scheduling feels harder than it is: people try to do the conversion in their heads, one pair at a time. For three or more zones that breaks down quickly, because each comparison is fine on its own and the group answer never appears.

Do it once, in one reference zone instead. UTC is the usual choice because it does not shift for daylight saving.

For a team in London, New York and Bangalore, each working 9am to 5pm locally:

• London 9:00 to 17:00 local is 09:00 to 17:00 UTC in winter, 08:00 to 16:00 UTC in summer.
• New York 9:00 to 17:00 local is 14:00 to 22:00 UTC in winter, 13:00 to 21:00 UTC in summer.
• Bangalore 9:00 to 17:00 local is 03:30 to 11:30 UTC all year, since India does not observe daylight saving.

Written out this way the problem stops being about time zones and becomes ordinary arithmetic on three ranges.

How do you calculate a meeting overlap window across time zones?

Take the latest start and the earliest finish across everyone. If the latest start falls before the earliest finish, everything between them is your overlap window. If it does not, there is no window at all.

Using the winter figures above: the latest start is New York at 14:00 UTC, and the earliest finish is Bangalore at 11:30 UTC. The latest start is after the earliest finish, so the overlap is empty. There is no hour of the day when all three are inside their normal working hours.

Drop Bangalore and the same arithmetic gives a real answer: latest start 14:00 UTC (New York), earliest finish 17:00 UTC (London), so 14:00 to 17:00 UTC works. That is 9am to noon in New York and 2pm to 5pm in London, which is why so many transatlantic teams converge on that band without ever working out why.

What if there is no overlap between time zones?

An empty overlap is the normal case past roughly a ten-hour spread, not a sign anyone has set their hours unreasonably. It means the honest question changes. It is no longer “when can everyone make it?” but “who steps outside their working hours, and how often?”

Three approaches work in practice:

• Rotate the imposition. Alternate which region takes the early or late slot, and keep a record of who has taken it, because memory alone reliably favours whoever complains least.
• Split the meeting. Two smaller regional calls plus a written summary often beat one call nobody can attend properly.
• Make it asynchronous. Status, updates and anything that is really a broadcast rarely need everyone in a room at the same moment. There is a short test for which meetings those are.

The failure mode to avoid is picking whichever slot the most people find convenient every time. Each individual decision looks reasonable, and the aggregate lands the same person outside their hours indefinitely. That is the pattern covered in why the same person always takes the early meeting.

What hours should count as a reasonable meeting time?

Most calendar tools answer a narrower question than the one you are asking. They show whether an hour is unbooked, which is not the same as whether it is a reasonable hour to ask someone to work.

A more useful set of bands:

• Working hours. Inside the person’s normal day. No imposition.
• Edge hours. An hour or two either side. Usually workable, and worth counting because it accumulates.
• Evening and early morning. Outside the working day and inside personal time. A real ask, and one people frequently absorb without saying so.
• Overnight. Treat as unavailable rather than as a difficult slot.

Grading the hours rather than treating them as free or busy is what makes a rotation possible: you cannot spread something evenly until you can count it.

A worked example: three zones, one recurring meeting

Say you need a weekly 30-minute sync across London, New York and Bangalore, and the overlap arithmetic above has already told you no common working hour exists.

The candidates are:

• 13:00 UTC. 1pm London, 9am New York, 6:30pm Bangalore. Inside working hours for two, early evening for one.
• 11:00 UTC. 11am London, 7am New York, 4:30pm Bangalore. Working hours for two, early for New York.
• 03:30 UTC. 3:30am London, 11:30pm New York, 9am Bangalore. Overnight for two. Not a real candidate.

So the practical choice is between an early start for New York and an early evening for Bangalore. Neither is free, and the deciding factor should be who has taken the last few, not who happens to raise it.

How to keep it fair once it is recurring

A one-off meeting at an awkward hour is a minor imposition. A recurring meeting at an awkward hour is the same imposition every week, indefinitely, and the person absorbing it is usually the only person with a reason to bring it up.

The lightweight version, no tooling required: list every recurring meeting, write down what local time each one is for each member, count how many fall outside each person’s working hours, and look at the counts. Most teams find one or two people carrying a disproportionate share, and the conversation is much easier once it is a number rather than an impression.

If you want that visibility to persist rather than being a quarterly exercise, that is the gap FairlyRemote fills: it shows every member’s local hours on one timeline, grades each candidate slot by who it actually affects, and keeps a record so the next awkward hour can land somewhere new. The fairness scoring explainer covers how that measurement works.

A note on what this isn’t

None of this makes an inconvenient meeting convenient. Distance across time zones is a real constraint and no scheduling method removes it. What is available is a choice about distribution: whether the constraint is absorbed by the same person every week, or shared across the team on purpose.