The short answer
Setting up a GCC in India takes six steps: define the charter and business case, choose the legal structure and city, incorporate the entity and complete registrations, hire the site leader before the team, secure workspace and IT, then run payroll and operations while documenting everything for eventual handover. A pilot of 15 to 25 people can be live in 10 to 14 weeks. A 100-person centre with full leadership typically reaches steady state in 9 to 12 months.
What this covers
- Step 1: Define the charter before the headcount
- Step 2: Build a business case your CFO will defend
- Step 3: Choose the legal structure
- Step 4: Incorporate and register
- Step 5: Hire the leader, then the team
- Step 6: Workspace, IT and security
- Step 7: Operate, measure, document
- What actually controls your timeline
Step 1: Define the charter before the headcount
Almost every stalled India programme we see began with a headcount target instead of a charter. Someone decided that 80 people should move offshore, and the organisation spent the next year arguing about which 80.
Start the other way round. Name the capability the centre will own end to end — a product area, a platform, a finance process, a customer segment. Ownership is what makes the centre worth having. A team that only executes tickets produced elsewhere will hire people who behave accordingly, and you will spend the next three years wondering why the quality gap never closes.
A practical test: can you name a business outcome the India centre will be accountable for, with a metric attached, that no team in headquarters is also accountable for? If not, the charter is not written yet.
Step 2: Build a business case your CFO will defend
A credible case has four components. First, a ramp curve rather than a headcount — 20 people in quarter one behaves very differently from 20 people in quarter four. Second, fully loaded cost per role, not salary. Third, an attrition assumption that matches the city and function you chose, applied to the ramp. Fourth, the one-time costs: incorporation, legal, recruitment, fit-out, devices and the parallel-running period when both teams exist.
The common error is modelling savings against current salaries instead of against the roles you would otherwise hire. If the alternative was not hiring at all, the saving is zero and the case has to rest on capacity or coverage instead. That is a perfectly good case. It is just a different one, and it needs different evidence.
Step 3: Choose the legal structure
For a captive centre, a private limited company is the standard vehicle. It permits 100% foreign direct investment under the automatic route for most service activities, gives limited liability, and is what candidates and landlords expect to see.
An LLP is occasionally used for very small or advisory-only setups, but it complicates employee stock arrangements and is unfamiliar to most global finance teams. A branch or liaison office is generally the wrong tool for a GCC: the permitted activities are narrow and the permanent establishment exposure is worse.
Decide your transfer pricing method at the same time. A captive service provider is usually remunerated on a cost-plus basis, with the margin benchmarked to comparable Indian service companies. That choice drives the intercompany agreement, the annual documentation burden, and how your Indian tax filings will look for the next decade. Settling it before incorporation avoids restructuring later.
Step 4: Incorporate and register
Incorporation itself is fast — often two to three weeks once documentation from the parent is apostilled and available. The registrations that follow are what determine when you can legally make an offer and run a payroll.
| Item | Typical time | Blocks hiring? |
|---|---|---|
| Digital signatures and director identification | 3–7 days | Precursor |
| Name approval and incorporation (SPICe+) | 2–3 weeks | Yes |
| PAN and TAN | Issued with incorporation | Yes |
| Bank account and FDI remittance reporting | 2–4 weeks | Yes, for payroll |
| GST registration | 2–3 weeks | For invoicing |
| Shops and Establishment registration | 1–3 weeks | Yes |
| Provident Fund and ESI registration | 2–4 weeks | Yes, for payroll |
| Professional tax (state-dependent) | 1–2 weeks | For payroll |
| Intercompany agreement and transfer pricing file | Parallel | No |
The sequence matters because several of these depend on a registered office address. Committing to a managed workspace early, even a small one, unblocks the whole chain.
Step 5: Hire the leader, then the team
The site leader is the highest-leverage decision in the programme. This person will set the hiring bar, negotiate with your global functions, represent the centre to candidates, and absorb the friction that a new site generates. Hiring them after the team is assembled is the single most common sequencing error.
Look for someone who has built rather than only run. Operating an established 400-person centre is a different job from opening one, and the skills do not transfer cleanly. Expect a 10 to 14 week search including notice period, which is why it starts in parallel with incorporation rather than after it.
For the team itself, decide your compensation percentile and hold it. Candidates talk to each other, and inconsistency across your first twenty hires creates internal equity problems that take years to resolve. Run your own interview panel from the first hire — outsourcing the hiring decision is how centres end up with people the parent organisation does not trust.
Step 6: Workspace, IT and security
Stage the real estate. Managed or serviced seats let you be operational within weeks and let you leave if the plan changes, at a per-seat premium that is trivial next to the cost of a wrong nine-year lease. Move to a dedicated floor when you have enough headcount certainty to justify the fit-out, usually somewhere past 80 to 120 people.
Bring your CISO in during week two. Data residency, endpoint management, identity federation, network segmentation and physical access will all need their sign-off, and every one of those is cheap to design in and expensive to retrofit. A security review that arrives in week sixteen, when the floor is ready and the offers are out, is how go-live dates slip.
Step 7: Operate, measure, document
Two things matter in the first year beyond delivery. The first is payroll and compliance accuracy: a wrong salary credit in month two costs you internal credibility that takes a year to rebuild. The second is documentation. Whether or not you plan a formal transfer, write the runbooks as you go — payroll process, compliance calendar, vendor contracts, IT configuration, onboarding. Assembling them retroactively is roughly three times the effort and always incomplete.
Measure the things that predict trouble: offer-to-join drop-off, ninety-day attrition, time-to-productivity by role, and the proportion of work the centre owns versus supports. The last one is the leading indicator for everything else.
What actually controls your timeline
Four dependencies decide whether you are live in three months or nine, and none of them is incorporation speed:
- Parent-company documentation. Apostilled board resolutions, charter documents and director KYC from the parent routinely take longer than the Indian filings they support.
- Your decision cadence. Programmes with a named decision-maker who can approve within 48 hours run roughly a third faster than those routing everything through a steering committee.
- The leadership search. Notice periods of 60 to 90 days are normal in India. Starting the search in week one rather than week eight is worth a full quarter.
- Security sign-off. Involve information security early or accept that they will become the critical path at the worst moment.