If you work in Indian financial services technology, you already deal with global capability centres, whether or not you call them that. A large and growing share of the engineering, data, risk, and operations work done for global banks, insurers, asset managers, and payment networks now sits inside centres they own and run in India. This guide is about what those centres actually are, where they cluster, and why hiring for one is a different problem from hiring for a product company or a domestic financial services firm.
What a GCC or captive actually is
A global capability centre, or GCC, is an entity that a parent company owns and operates directly, offshore from its home market, to do work for itself. The defining feature is ownership. The people in a GCC are employees of the group, working on the group's own systems and products, not a vendor's staff sold back to it under a services contract. That is the line that separates a GCC from third-party outsourcing, and it is the whole point of the model.
The terminology has shifted over the years and the words get used loosely. "Captive" is the older term and is still common, especially in financial services. "GIC", for global in-house centre, had a run. "GCC", for global capability centre, is the label most of the industry has settled on now, and the change from "captive" to "capability" is not only marketing. It marks a real shift in what these centres are asked to do, which we come back to below. For most practical purposes you can treat captive, GIC, and GCC as the same thing.
What they are not is a call centre or a back office in the old sense. A modern financial services GCC in India runs core engineering, builds and owns products, does quantitative and data science work, runs cyber and cloud, and increasingly holds global process ownership rather than just executing tasks defined elsewhere.
Why India, and where the centres cluster
India is the largest location in the world for GCCs, and financial services is one of the deepest sectors within that. The reasons are the familiar ones: a large pool of engineering and analytics talent, an established operating environment that global firms already understand, and a time-zone and language fit that works for firms headquartered in the United States and the United Kingdom. None of that is new. What has changed is the seniority and the mandate of the work, which has pulled the centres up the value chain and made the talent they need scarcer.
The centres concentrate in a handful of cities, and the concentration matters for hiring because it sets who you are competing with for any given candidate. Bengaluru is the largest and deepest market, and the default location for a technology-heavy build. Hyderabad has grown into a major centre and competes hard for engineering and data talent. Pune carries a strong base of engineering and delivery talent, often at a slightly lower cost point than Bengaluru. Delhi NCR, particularly Gurugram and Noida, holds a large share of the more senior and leadership-weighted roles. Chennai and Mumbai both host significant financial services centres, with Mumbai's proximity to the domestic banking and markets industry giving it a particular character.
The practical consequence is that a captive's hiring difficulty is largely set by its city. A firm building in Bengaluru or Hyderabad is hiring into the most contested engineering markets in the country, against product companies and other captives paying at the top of the range. The same role in Pune or Chennai is a different, and often more winnable, search.
What makes a financial services captive different
Two things separate a financial services GCC from a generic technology captive, and both change how you hire.
The first is domain. A payments platform, a trading system, a credit-decisioning engine, or a regulatory-reporting pipeline is not a generic piece of software, and the engineer who has built one inside a regulated firm carries context that does not show up on a CV as a skill. It shows up as time served under audit, under reconciliation discipline, under the knowledge that a defect in the wrong place is a reportable event rather than a bug for the next sprint. A captive that hires purely on stack, ignoring whether the person has worked inside financial services, tends to spend the saving twice over on the ramp. We have written more on why this matters in BFSI technology hiring in India.
The second is the shift from captive to capability. The older captive was a cost-arbitrage play: move a defined body of work offshore, run it cheaply, measure it on efficiency. The centres that matter now are measured on capability, on owning products and outcomes rather than executing instructions from headquarters. That shift changes the hire. A cost centre needs competent executors. A capability centre needs people who can hold a product decision, argue with the parent, and run something end to end. Those are more expensive, harder to find, and much harder to assess, and a hiring process built for the first kind of centre quietly fails at the second.
Why hiring for a captive is its own problem
A captive occupies an unusual position in the talent market, and candidates read it in ways that cut both directions.
On one side, a GCC offers things a startup cannot: stability, the scale and rigour of a global institution, exposure to problems at real size, and a brand that travels. For many strong engineers, especially those a few years in who want depth and stability over equity and chaos, that is genuinely attractive.
On the other, a captive competes for exactly the same people as the product companies and the other captives in the same city, and it often competes with one hand tied. Compensation bands are set with reference to a global structure and a comp committee, which makes them slower to move than a well-funded fintech that can decide on an exceptional offer in a single conversation. A captive that runs its offers against an internal band without checking what the specific candidate is actually being paid, and what they expect to move, loses hires it should have won. That failure gets recorded as a compensation problem when it was really a diligence problem, and it is the single most common way a good captive search stalls. Our guide to technology compensation benchmarks covers how to read the current market before you build an offer.
There is also a rhythm problem specific to captives. Hiring during a build-out phase, when a centre is standing up a new function and needs volume fast, is a different exercise from steady-state hiring into an established team. The build-out phase rewards a partner who can move at pace across many roles at once and hold quality while doing it. Steady-state hiring is more about the occasional senior or genuinely scarce role. Firms that use the same approach for both tend to under-serve whichever phase they are not currently in.
The roles that are hardest to fill
Across the financial services GCCs we work with, the difficulty concentrates in a few predictable places.
- Leadership for a new build. The person who will stand up a function, hire the first cohort, and represent the centre to a sceptical parent is a rare profile, and getting it wrong sets the whole build back. This is often a confidential search.
- Domain-plus-technology intersections. Risk technology, quantitative development, core-banking and payments engineering, and regulatory-reporting platforms all sit where a small specialist pool meets sharp demand, and the bidding is correspondingly hard.
- Roles that compete head-on with product firms. Machine-learning and data science talent, in particular, is chased by every product company and captive at once, and a captive's slower comp process is most exposed here.
- Anything scarce in a contested city. The same role that fills in weeks in one market can stay open for a quarter in Bengaluru or Hyderabad, purely because of who else is hiring.
How to hire well for a captive
None of this is unfixable. The captives that hire well tend to do a few things deliberately.
They brief for the domain, not just the stack, so the shortlist is judged on whether a candidate has worked inside financial services rather than on keyword overlap. They find out a candidate's real compensation and expectation early, before the offer is built, so the offer that eventually goes out is not fifteen per cent into a conversation the candidate thought started somewhere else. They separate build-out hiring from steady-state hiring and resource each properly. And they treat the confidential and leadership searches as a different exercise from volume hiring, because they are.
Most of this is the ordinary discipline of good hiring, applied in a setting where the cost of getting it wrong is unusually high because the parent is watching and the market is contested. A specialist partner earns its place here by knowing the specific market, the specific city, and the specific comp reality, rather than working the brief as a generic technology search.
If you are building or scaling a financial services captive in India, that is the work we do. Our GCCs and captives page sets out how we engage across build-out and steady-state hiring, and the talent acquisition page covers how a mandate runs end to end.