One of the most common questions we hear from companies exploring India is also the most expensive to get wrong: should we launch the GCC at full size, or start with a pilot and scale?
The answer is almost always the same. The most successful GCCs we've helped build — and the ones we've watched others build well — started with 10 to 30 people. Not 200. The "Pilot + Scale" model is, in our experience, the single biggest predictor of whether an India operation matures into a real capability engine or becomes an expensive lesson.
Why big-bang launches go wrong
It's easy to see the appeal of launching at full size. You've made the business case. You've raised the budget. You want results in this fiscal year, not next. So you green-light a 200-person ramp from day one.
The problem is that everything you don't know yet — and there's a lot you don't know yet — gets baked into the foundation while it's still wet. Hiring decisions get rushed. Reporting lines get drawn before you understand the actual work. Office space gets leased for a team shape that will look different in six months. And nobody has time to embed culture, because everyone is buried in execution.
The foundation matters more than the speed. Every GCC we've seen unravel was rushed into existence.
What a pilot actually tests
A pilot isn't just "the first phase before the real launch." It's a deliberately undersized team designed to surface answers to five questions that you can't answer from outside India:
- Talent availability for specialized skills. The headline "India has 1.5 million engineers a year" doesn't tell you whether you can hire 8 senior Rust engineers in Bangalore in 90 days. Find out at 20 people, not 200.
- Productivity metrics. Throughput, defect rates, sprint velocity, cycle time — measured against your US team's baseline. Is the work actually equivalent quality?
- Leadership bandwidth. Does your US leadership have the time and time-zone tolerance to manage a remote team, or do you need a senior India lead from day one?
- Cultural alignment. How do feedback loops work? Are decisions getting made or escalated? Does the team feel like an extension or a vendor?
- Cost-to-value ratios. Not just "is it cheaper" — but "is the work delivered worth the total cost of ownership, including management overhead?"
Six months at 20 people gives you data on all five. Six months at 200 people gives you a problem.
The common mistakes we see in rushed setups
When companies skip the pilot phase, the same four problems show up:
- No clarity on roles and ownership. Teams are hired before anyone has decided who owns what. The result: duplicate work, missing work, and lots of awkward calls.
- Unclear reporting structures. Does the India lead report to the US functional head, the India MD, or both? Get this wrong and decisions stall for weeks.
- An operating model designed for the US team, not the joint team. Standups at 9am US time work for the US team. They're at 7:30pm in Bangalore.
- Delivery expectations that haven't been negotiated. The US team assumes India is operating at full velocity from week one. India is still finding the bathroom.
Each of these failures is cheaper to fix at 20 people than at 200. Even better: most of them are easier to prevent when the pilot team's first job is to design the operating model alongside you.
What "Pilot + Scale" looks like in practice
A well-run pilot has three phases:
Months 1–3: Build the founding team
Hire your India lead first. Then 8–15 people in one or two functions — typically engineering or operations — that the US team understands well enough to manage closely. Set up the entity, the office, and the basic operating cadence.
Months 4–6: Operate and measure
Let the team run real work. Track the five questions above. Adjust roles, processes, and reporting lines as you learn. Resist the urge to keep hiring just because budget allows it.
Months 7–12: Scale with conviction
Now you know what works. You have a leader. You have an operating model. You have proof points. Now you ramp — and the scaling is fast because the foundation is real.
The mindset shift
The Pilot + Scale model only works if you stop thinking of India as a way to get work done cheaper, and start thinking of it as a long-term capability engine. Engines need to be tuned before you put them under load.
That mindset is the single biggest gift you can give a new GCC — and the one most companies skip when they're under pressure to show fast results.
Where are you in your setup journey — planning, hiring, or scaling? Drop us a line at hello@globalcapabilitypartners.com and we'll share what we've seen at your stage.
In the next post, we'll cover the five most common mistakes global firms make during India expansion — and how to spot them before they cost you a year.