GCC Governance: Who Decides What Between HQ and India

GCC governance is, underneath the org charts and steering committees, a decision-rights problem: who between headquarters and the India center actually decides each thing, and where the honest answer is "both," how that shared decision is made. A Global Capability Center (GCC) that never writes this down does not avoid the question, it just answers it slowly and painfully, one disagreement at a time, usually around month six. This is a practical map of which decisions sit where, the cadences that keep the two sides aligned, and the friction to expect before it settles.

I set up and repair delivery governance for centers, and the pattern is consistent: the centers that run smoothly are not the ones with the most process, they are the ones where everyone knows who decides what. So let me start with why the absence of that clarity is the real failure mode.

Why does GCC governance fail without clear decision rights?

GCC governance fails when decision rights are implicit, because implicit rights get interpreted differently on each side of the ocean, and the gap only surfaces when a real decision is contested. Everything feels fine while the center is small and deferential; it stops feeling fine the moment the center is competent enough to have opinions.

The classic failure is the center that was told it was "empowered" but discovers, the first time it makes a real call, that headquarters expected to be asked. Now every decision carries a hidden question of whether it was theirs to make, momentum drains into permission-seeking, and the strong hires who joined to own something start to wonder why they cannot. The opposite failure exists too: a center handed decisions it lacks the context to make well, producing choices headquarters then quietly overrides, which teaches the center that its authority is fictional. Both come from the same root, which is that nobody wrote down who decides what. The fix is not more governance meetings; it is a decision-rights map that both sides have actually read and agreed to. Writing it down is not bureaucracy, it is the cheapest insurance a center can buy, because the alternative is learning where the boundary sits by crossing it in front of an unhappy stakeholder. And the document itself is not really the point; the shared act of agreeing it is, so that both sides carry the same map instead of two different ones they each assume the other shares.

Who decides what between HQ and the India center?

The cleanest way to answer this is a decision-rights matrix that names each significant decision and its owner, distinguishing what HQ owns, what the center owns, and what is genuinely shared with one side leading. The table below is a sane default; the point is less the exact allocation than that you make one explicit and revisit it as the center matures.

Two things make a matrix like this work. First, "shared" is not a dodge; each shared row names who leads, because a shared decision with no lead is just a fight waiting to happen. Second, it is written where both sides can see it and is reviewed on a schedule, because a decision-rights map that lives in one person's head is no map at all.

What governance cadences keep the two sides aligned?

Decision rights tell you who decides; cadences tell you when and how the two sides talk, and without them even a perfect matrix drifts. The right rhythm is a small number of standing forums, each with a clear purpose, rather than a swarm of status meetings.

Three cadences carry most of the load. A strategic review, perhaps quarterly, where HQ and center leadership align on direction, priorities, and the decision-rights map itself as things change. An operational review, weekly or fortnightly, where delivery, risks, and blockers are worked, and where most day-to-day decisions are simply reported rather than sought. And a clear escalation path, defined in advance, so that when something genuinely contested arises, everyone knows exactly who it goes to and how fast, instead of it festering in email across time zones. The discipline is to keep these few and real. Centers drown in governance when every uncertainty spawns a new committee; they stay aligned when a handful of forums, with honest agendas, do the work. Who sits in each forum matters as much as how often it meets: a strategic review without a real decision-maker from headquarters just produces recommendations nobody can approve, and an operational review packed with executives produces theater instead of delivery. Staff each forum with the people who actually hold the decision rights for what it covers, and keep everyone else out of it.

Where does the month-six friction come from?

Almost every center hits a friction point around month six, and it comes from the same place: the center has become competent enough to want real ownership at exactly the moment headquarters has not yet let go. The honeymoon of the early build is over, the team is good, and the decision-rights map that was fine for a dependent center now chafes.

This is not a sign that something is broken; it is a predictable stage, and naming it in advance defuses most of the damage. The center reads continued HQ control as distrust; HQ reads the center's push for autonomy as overreach; and because nobody planned for the transition, it plays out as tension rather than as a scheduled handover of decisions. The centers that pass through it well are the ones that treated the decision-rights map as a living thing from the start, with an explicit expectation that ownership would migrate toward the center as it earned it. The ones that struggle are those that framed the initial allocation as permanent and then had to renegotiate it under stress. A useful move is to schedule a real governance review at around the six-month mark before the friction shows up, framed not as a reaction to tension but as a planned checkpoint to hand the center the decisions it has earned. Naming the moment in advance turns a brewing conflict into an ordinary agenda item, which is a much better place for it to live. If you are standing a center up, our guide to setting up a GCC in India covers the build, but plan the month-six shift into your governance from day one.

How do decision rights shift as the center matures?

Decision rights should migrate from HQ toward the center over time, deliberately and visibly, as the center accumulates context and a track record. A governance model that is static is a governance model that will be wrong within a year, because the whole point of a capability center is that it grows into more capability.

The trajectory is predictable enough to plan. Early on, HQ holds more, because the center lacks the context to make certain calls well, and that is appropriate. As the center proves itself, ownership of hiring, day-to-day architecture, and delivery priorities shifts toward it, with HQ moving from deciding to calibrating and setting guardrails. In a Build-Operate-Transfer arrangement this migration is even more pronounced, since the entire point is that ownership ends up with you; our write-up on what actually transfers in a BOT covers how the decision rights, not just the people, have to move for a transfer to be real. The governing idea is simple: decide who decides, write it down, and schedule the moment you will move the line. A center whose autonomy grows on a plan stays motivated and aligned; a center whose autonomy is negotiated crisis by crisis burns trust it did not need to spend. The healthiest centers treat the decision-rights map like a roadmap of their own autonomy: a visible sequence of decisions they will own by a named milestone, each tied to a capability they will have demonstrated to earn it. That reframes control from a standing grievance into a shared plan with a direction everyone can see and work toward. If you want help designing an operating model that ages well, our Global Capability Center work starts exactly here.