Distributed engineering is usually treated as a synchronisation problem: find the overlapping hours, protect them, schedule the important conversations inside them.
That approach has a ceiling. Overlap is a fixed, small resource, and every process that depends on it competes for the same few hours — which is why the standard experience of distributed work is a compressed block of back-to-back calls followed by a long stretch where you are blocked on someone asleep.
The organizations that handle it well spend their effort somewhere else: reducing how much overlap they need in the first place.
What consumes overlap unnecessarily
Decisions that were never written down. If the reasoning lives only in a call, anyone absent must ask — and asking costs a round trip of up to a day. Writing decisions down is not documentation overhead; it is overlap you get back.
Ambiguous ownership. When it is unclear who decides, the resolution is a meeting. Clear ownership converts many meetings into a message.
Status as a synchronous activity. Standups exist to transmit information one person at a time to people who mostly do not need it. Asynchronous written status returns the most overlap for the least change, and is the easiest thing to try.
Reviews that block. If a change needs approval from one specific person in another time zone, you have built a daily stall into your pipeline. Multiple qualified reviewers per area is a scheduling fix disguised as a quality practice.
Overlap should be spent on the things that genuinely need a room: disagreement, ambiguity, and anything with feelings in it. Everything else is a message.
What overlap is actually for
Three things, in my experience. Disagreement — text makes conflict worse, reliably. Genuine ambiguity, where the shape of the problem is not yet clear and exploring it in writing is painfully slow. And anything personal — feedback, career conversations, difficulties.
Everything else can be asynchronous, and treating it that way protects the overlap for the cases that need it.
Handover as a deliberate practice
The teams that do this well treat the end of a working day as a handover, not a stop. A short written note: what moved, what is blocked, what the next person could usefully pick up.
It takes a few minutes and converts time-zone spread from a pure cost into something closer to a relay. Without it, distributed teams do not get follow-the-sun delivery — they get follow-the-sun waiting.
The signal to watch
Track how often work sits blocked waiting for a person rather than a system. If that number is high, the constraint is decision ownership and written context — not the number of overlapping hours, and no amount of calendar engineering will fix it.