Nano GCC vs Traditional GCC vs Captive Center: The Real Differences
Almost every conversation about building capability in India starts in the wrong place. It starts with a question about cost, when the question that actually determines the outcome is about structure: who employs the team, who carries the compliance risk, who owns what the team builds, and how long before any of it produces value.
Three models dominate that decision. The traditional Global Capability Center, built at scale over a year or more. The captive center, which is often used interchangeably with traditional GCC but carries a specific legal meaning. And the Nano GCC, a smaller, faster, mandate-specific model that has grown quickly because the constraints that made the other two necessary no longer apply to most companies.
This article compares all three properly, on the dimensions that decide whether the decision succeeds, and sets out how to tell which one your situation actually calls for.
Key takeaways
- A captive center is a legal structure: you own the Indian entity and employ the team directly.
- A traditional GCC is a scale model: several hundred to several thousand people, built over nine to eighteen months.
- A Nano GCC is a mandate model: 10 to 40 people at launch, operational in three to four months, with a partner carrying the entity.
- The models are not competitors in a single ranking. They solve different problems and fail in different ways.
- Structure, not day rate, is what determines cost, speed and risk over a twelve month horizon.
Getting the definitions right first
Much of the confusion in this space comes from three terms being used as if they were interchangeable. They are not, and the differences are consequential.
A captive center describes ownership. The parent company incorporates an Indian legal entity, becomes the employer of record, and carries all statutory obligations directly. Captive is a statement about who holds the liability, not about how many people work there.
A traditional GCC describes scale and scope. It is typically a captive center, but the defining feature is that it runs broad functions at significant headcount, often across IT, engineering, finance operations and shared services simultaneously.
A Nano GCC describes mandate and speed. It is a small team, built around one named capability, where a partner operates the entity, compliance, payroll and facilities so the parent company can direct outcomes without becoming an employer in a jurisdiction it does not know.
The practical distinction. Captive answers “who is legally responsible”. Traditional answers “how large and how broad”. Nano answers “how fast and how focused”. A company can move between them over time, and many do, starting with a Nano GCC and converting to a captive structure once scale justifies carrying the entity in house.
The three models compared
The table below sets the models against the dimensions that consistently determine whether a capability center succeeds. Cost per hour is deliberately not the first row, because it is rarely the variable that decides the outcome.
| Dimension | Nano GCC | Traditional GCC | Captive Center |
|---|---|---|---|
| Typical launch size | 10 to 40 people | 300 to 5,000 people | 100 to 5,000 people |
| Time to operational | 3 to 4 months | 9 to 18 months | 9 to 18 months |
| Who employs the team | Partner entity | Parent company | Parent company |
| Who carries compliance risk | Partner | Parent company | Parent company |
| Scope | One named mandate | Broad, multi-function | Broad, multi-function |
| Setup capital required | Low | High | High |
| IP ownership | Assigned to parent outright | Parent | Parent |
| Exit or wind-down cost | Low | High | High |
| Measured on | Value created | Cost and throughput | Cost and throughput |
| Best suited to | A specific capability, quickly | Large steady state functions | Long term strategic scale |
Ranges reflect typical engagements rather than absolute limits. A Nano GCC can grow well past 40 people; the figure describes launch size.
Why scale used to be mandatory
The traditional model was not built large because large was better. It was built large because the fixed costs made anything smaller uneconomic.
Incorporating an Indian entity, maintaining statutory compliance, running payroll, leasing Grade A office space and staffing an HR function all carry a floor cost that exists whether the center employs fifty people or five hundred. Spread across five hundred, that overhead disappears into the unit economics. Spread across fifteen, it dominates them.
That arithmetic is the entire reason capability centers were historically the preserve of large enterprises. It was never about whether a mid-market company would benefit from a dedicated team in India. It was about whether it could absorb the fixed cost of getting one.
What changed
Two things. First, partners emerged that operate the entity, compliance and facilities layer across many clients simultaneously, which amortises the fixed cost the same way scale used to, without requiring any single client to provide that scale. Second, AI and modern engineering tooling reduced the headcount a given mandate requires, so the teams companies actually need got smaller.
Together these mean a fifteen person team can now access enterprise grade infrastructure and compliance without needing enterprise scale to pay for it. That is the structural change underneath the growth of the Nano GCC model, and it is worth understanding because it also tells you when the model does not apply.
Months from decision to a team at full velocity
The compression comes from running compliance, facilities and hiring in parallel against pre-existing infrastructure, rather than sequentially from zero.
How the cost comparison actually works
The most common analytical error in this decision is comparing a Nano GCC day rate against a captive center salary line. These are not comparable numbers, because they cover different things.
A captive center salary excludes the entity, the compliance function, the facilities lease, the HR apparatus and the internal management time spent operating all of it. A partner-operated rate typically includes most or all of that. Comparing them directly makes the captive look cheaper by excluding the costs that make it expensive.
The honest comparison models both on an all-in basis over a defined horizon, usually twelve to twenty four months. Our True-Up Cost Methodology sets out the four categories that most models omit: benefits and statutory contributions, attrition backfill, facilities and IT overhead, and compliance exposure.
Fixed cost carried
In a captive, entity, compliance and facilities are your fixed cost regardless of team size. In a Nano GCC they are the partner’s, spread across their client base.
Time cost of the build
Nine to eighteen months of delayed capability is a real cost that almost never appears in a spreadsheet, even though it frequently exceeds the rate differential.
Risk cost
Getting entity setup, labour law or tax registration wrong is low probability with a mature operator and high impact without one. It belongs in the model.
Governance, control and IP
A persistent objection to the partner-operated model is loss of control. In practice, control divides into three separate things, and they behave differently.
Direction stays with you in all three models. You define the mandate, set priorities, run planning and make architectural decisions. A well structured Nano GCC works inside your sprint process with direct access to your product owner, not through account managers.
Employment is where the models genuinely differ. In a captive you are the employer with all that entails, including termination obligations and statutory liability. In a Nano GCC the partner carries that, which is a transfer of risk rather than a loss of control.
Ownership of output should be identical across all three. In a properly structured Nano GCC, all IP is assigned to your entity outright by contract before the first engineer starts. If a provider is ambiguous on this point, that ambiguity is the actual risk, not the model.
How to tell which model fits
The decision is more tractable than it looks, because the three models suit genuinely different situations. Four questions usually resolve it.
Is the scope one capability or many functions?
If you can name the mandate in a sentence, a product engineering pod or an agentic AI team, that points to a Nano GCC. If you are moving several broad functions simultaneously, the traditional model is the honest answer.
What is the acceptable time to capability?
If the answer is this quarter, only one model delivers it. If you can genuinely absorb a year, the captive route becomes viable and may be preferable at sufficient scale.
Do you want to be an employer in India?
This is a strategic question, not an administrative one. Carrying an entity means carrying statutory, tax and termination obligations in a jurisdiction with its own labour law regime.
What headcount does the mandate genuinely require?
Below roughly one hundred people, the fixed cost of a captive rarely amortises well. Above several hundred, carrying it in house starts to make economic sense.
The path most companies actually take
A pattern worth naming: many companies start with a Nano GCC around one mandate, expand it as the mandate proves out, and only convert to a captive structure once headcount reaches a level where carrying the entity in house is economically rational.
That sequence has a significant advantage. It defers the largest, least reversible commitment until after the capability has demonstrated value, rather than requiring the commitment as a precondition of finding out. See Micro GCC vs Nano GCC for how the intermediate stage typically works.
Three mistakes that recur
Across these decisions, the same three errors appear repeatedly, and all three are avoidable.
Choosing scale by default
Selecting the traditional model because it is the familiar one, when the actual requirement is a fifteen person team owning one thing. The result is a year lost and an overbuilt structure.
Comparing incomparable numbers
Setting a partner-inclusive rate against a bare captive salary line. This reliably makes the more expensive option look cheaper and is the single most common modelling error.
Leaving IP until later
Treating intellectual property assignment as a detail to settle after launch. It is the one term that becomes materially harder to negotiate once the team has built something valuable.
Frequently asked questions
What is the difference between a captive center and a GCC?
Captive describes ownership: the parent company owns the Indian entity and employs the team directly. GCC describes function: a center delivering capability for the parent business. Most traditional GCCs are captive centers, but a Nano GCC delivers the same function without the parent carrying the entity.
Is a Nano GCC just a smaller traditional GCC?
No. The difference is structural rather than only numerical. In a Nano GCC a partner operates the entity, compliance, payroll and facilities, which is what allows it to be small and still economic. A traditional GCC at fifteen people would carry unamortised fixed costs.
Which model is cheapest?
Over a twelve month horizon and at small to mid scale, a partner-operated Nano GCC is usually cheapest on an all-in basis, because the fixed cost of the entity is shared. At several hundred people and above, a captive structure typically becomes more economic.
Do we lose control with a Nano GCC?
You retain direction, priorities and architectural decisions. What transfers is employment risk and statutory obligation. A well structured engagement gives your product owner direct access to the team with no account management layer.
Who owns the IP in each model?
You should, in all three. In a Nano GCC this must be contractual and assigned outright to your entity before work begins. Any ambiguity on this point should be resolved before signing rather than afterwards.
Can we convert a Nano GCC into a captive center later?
Yes, and this is a common path. Companies frequently start with a partner-operated team, prove the mandate, then incorporate and transition once headcount justifies carrying the entity directly.
How long does each model take to launch?
A Nano GCC is typically operational in three to four months. A captive center or traditional GCC generally takes nine to eighteen months, driven by entity incorporation, facilities and large-scale hiring running sequentially.
What size should we start at?
Most Nano GCCs launch at 10 to 40 people sized to a specific mandate. Starting smaller and expanding on demonstrated value is almost always a stronger position with a board than committing to scale in advance.
Conclusion
The three models are not points on a single scale from worse to better. They are answers to different questions, and the reason the Nano GCC has grown quickly is not that the others stopped working. It is that the fixed cost barrier which made them the only options has been removed, so companies that were previously excluded from the decision now have a route in.
If you need broad functions at significant scale and can absorb a long build, the traditional and captive models remain appropriate and well understood. If you need one capability owned properly and you need it this year, the Nano GCC exists precisely for that case.
The decision that matters is not which model is best in the abstract. It is which one matches your mandate, your timeline and your appetite for carrying employment risk in a jurisdiction you do not operate in. Answer those three honestly and the choice usually makes itself.
Not sure which model fits your mandate?
Tell us the mandate you need owned. We will come back with a team shape, a realistic timeline and an all in cost model you can take to your board.