A large share of leadership roles in this atlas are in India, running teams whose stakeholders are elsewhere. The difficulties are structural rather than cultural, and most are addressable by design.

You get a few hours a day of overlap with headquarters. What goes into it determines what your teams are unblocked on for the next twenty-four hours. So: decisions and unblocking in the overlap, status in writing outside it. A synchronous meeting spent on updates that could have been read is the most expensive thing that happens in a distributed organisation.

Centres are frequently described as execution arms, and leaders respond by trying to build trust. That rarely works, because the cause is usually structural rather than attitudinal. The questions that diagnose it:

• Does anyone here own a product outcome end to end, or do we own components of outcomes owned elsewhere? If the second, no amount of relationship-building changes the perception, because the org chart says what it says.
• Is someone from here in the room when the roadmap is set — at the point of decision, not at the readout? Being consulted is not deciding.
• Where do our strongest people go when they leave? If they leave at the same level for product companies, we are not offering scope, and that is measurable.
• Who presents our work upward? If it is always someone from headquarters, the centre is invisible regardless of what it delivers.

Leaders arriving from pure engineering backgrounds are consistently underprepared for the statutory side: entity governance, labour law, data residency, and in regulated industries a local board or regulator. It is real work, it is not delegable in the way people assume, and having a credible answer about it separates candidates at site level.

Distributed teams fail on decision latency and context, not on communication tooling. The tooling has been adequate for a decade; what has not changed is that a decision requiring three time zones to agree takes days rather than minutes, and that the people who were not in the room do not know why something was decided.

• Write decisions down, with the reasoning. In a co-located team the reasoning diffuses through overheard conversation. Distributed, it does not exist unless someone wrote it, and its absence is why remote teams re-litigate decisions the centre considers closed.
• Push decision authority to where the work is. Every decision that must travel to another time zone costs at least a day. If a site cannot decide anything, it is an execution function regardless of what the org chart says — and its best people will work that out and leave.
• Overlap hours are a budget, not a schedule. Two or three hours of overlap is the scarcest resource in a distributed organisation. Spending it on status meetings is the most common waste; it should go to the things that genuinely need synchronous discussion — disagreement, design and difficult feedback.
• Rotate the inconvenience. A recurring meeting that is always early morning for one site and always mid-afternoon for another says something about whose time counts, and everyone at the inconvenienced site knows exactly what it says.
• And name the work, not just the people. A site given ‘support for’ another site's roadmap has no way to demonstrate impact and no path to owning anything.