Four things changed, and the threshold fell
Small owned capability centres were not possible for most US companies five years ago. The reason they are now has less to do with strategy than with four costs that fell at roughly the same time.
The entry fee fell, and the reason to pay it changed
Two independent shifts. Building an owned team in India got much cheaper to start, and the reason companies want one moved from cost to access.
For twenty years the capability centre conversation in the US had a fixed shape. It was a cost conversation, it applied to large enterprises, and it required a commitment measured in years and hundreds of people. Companies below a certain size looked at it, concluded correctly that the arithmetic did not work, and stopped looking.
Both halves of that have changed. The fixed cost base that forced the scale threshold has fallen substantially, which makes a team of eight to fifteen viable. And the reason companies want offshore capability has shifted from reducing cost to acquiring seniority they cannot hire at home. Those two changes together are what produced the Nano GCC as a category.
Four costs that fell at roughly the same time
None of these are about capability centres specifically, which is why the shift happened across very different companies almost simultaneously.
Entity and compliance became routine
Company formation, payroll, statutory registration and the annual filing cycle in India are now well trodden and can be stood up in weeks. This used to be a large share of the fixed cost that demanded scale.
Infrastructure stopped being capital
Cloud, managed services and remote first tooling removed the data centre, the long lease and most of the internal IT function that a large centre had to carry.
Distributed work became normal
The practices that make a small remote team effective are now standard rather than experimental, which removed the argument that a team had to be large enough to be self sufficient in one building.
Senior talent became directly reachable
A company of two hundred people can now hire a principal engineer in Bengaluru directly and credibly. In the previous era that access was largely mediated by scale and brand.
Why companies want one has also changed
This is the part that decides most engagements. The cost case is still true and it is no longer what starts the conversation.
When senior engineering roles stay open for two or three quarters, the question stops being how to do work more cheaply and becomes whether the capability can be acquired at all in a timeframe that keeps the plan viable. That is an access problem, and it responds to a different solution than a cost problem does.
The distinction matters practically because the two produce different centres. A centre built for cost is optimised for a low average per head, which means a thin seniority mix and specified work. A centre built for access is small, senior heavy and scoped around ownership. Building the first while intending the second is the most common way these arrangements disappoint.
| Decision | Built for cost | Built for access |
|---|---|---|
| Objective | Reduce cost of known work | Acquire capability that cannot be hired locally |
| Seniority mix | Minimised | Deliberately heavy |
| Size | Large enough to amortise fixed costs | As small as the capability requires |
| Work assigned | Well specified | Ambiguous, requiring judgement |
| Success measure | Cost per FTE | Whether the roadmap moves |
| Competing against | Other offshore providers | Every company hiring that seniority band |
The last row is the one most often missed. A centre built for access competes for candidates against every well funded company in that market, not against other outsourcing options.
The companies where this is now live
The model became available to a group that had correctly ruled it out before, and remains the wrong answer for some of them.
Companies of 50 to 500 people
Previously below the viable threshold by a wide margin. This is the group for whom the arithmetic changed most, and the group least likely to have revisited the decision.
Companies with stalled senior hiring
Where roles have been open for more than two quarters, the cost of not hiring usually exceeds the entire salary difference. That changes which structure makes sense.
Companies with permanent core systems
Product surfaces, platforms and data systems that will exist in five years. Permanent work benefits from permanent owners in a way bounded projects do not.
Not: companies with bounded projects
A migration or a defined integration programme has a real end date. A vendor is usually the better answer and we will say so.
What the conversation actually looks like
In practice four questions settle it, and none of them is about rates.
- Will these systems still exist and need improving in five years?
- Has senior hiring genuinely stalled, or is the budget the constraint?
- Can the team be given a named system to own, with the decisions attached?
- Is there an executive sponsor who will protect it through a difficult quarter?
- What would have to be rebuilt if the arrangement ended?
The question that most often ends the conversation. If the answer to the third question is that everything would still need onshore approval, the model will not work regardless of who is hired or how much is spent. That is worth establishing in the first call rather than the second year.
Where the entry threshold actually sits now
Fixed costs still hit small centres hardest. What changed is how large that fixed base is, not the shape of the problem.
Illustrative pattern showing fixed cost as a share of total cost by centre size, for planning rather than measured research. Model your own fixed base explicitly rather than blending it into a per head average.
The practical implication is that the marginal economics improve fast in the early range and then flatten. Going from eight people to twenty improves per head economics substantially; going from eighty to a hundred barely moves them. That curve is why the viable floor moved rather than disappeared.
It also means the second team is much cheaper than the first, because most of the fixed base is incurred once. That optionality is real value and it is routinely left out of the business case for the first team.
What a first engagement looks like
Entity and the search
Company formation, payroll and statutory setup run in parallel with the senior pod lead search, which opens on day one because it is the critical path.
Senior core in place
Pod lead first, then senior engineers and the decision maker. The lead interviews every later hire, which is what makes it a team.
First productive output
Environments, access and domain context complete. Real work ships. Readiness is what gets measured until this point.
Ownership transfers
A named system changes hands on a named date with the onshore approval gate removed rather than relabelled.
The companies revisiting this decision are mostly not doing it because their strategy changed. They are doing it because a role has been open since last spring.
Frequently asked questions
Is this just a new name for offshore development?
No. Offshore development centres are typically larger, more junior and scoped to execute work specified elsewhere. A Nano GCC is small, senior heavy and scoped around owning named systems including the decisions about them. The difference shows up in who can make a call without asking.
Why now rather than five years ago?
Because four costs fell at roughly the same time: entity and compliance, infrastructure, the overhead of distributed working, and the difficulty of reaching senior talent directly. Together they lowered the entry threshold far enough that small teams became viable.
Are large companies doing this too?
Yes, often alongside an existing large centre. A small senior team scoped around a specific outcome is a useful structure even for a company that already operates at scale, particularly for capability that does not fit the existing centre s operating model.
Is the cost saving still the main driver?
It is still real and usually substantial, and it is increasingly not what starts the conversation. Most engagements we see begin with senior hiring having stalled rather than with a cost reduction target.
What size company does this suit?
Companies of roughly fifty to five hundred people are where the change is most consequential, because that group was previously excluded by the scale threshold. Larger companies use it too, generally for capability that sits outside an existing centre s remit.
How is this different from an employer of record?
An employer of record gives you employees without an entity, which suits testing a market or a very small first team. It costs more at scale, produces a less clean IP chain and creates a dependency on the provider, so it is often a bridge to an entity rather than a permanent structure.
What is the biggest risk?
Underinvesting in seniority. The model only works if judgement is present locally, and the most common failure is applying a cost per head target that thins the seniority mix until the team can no longer decide anything.
How do we know if we are ready?
The readiness assessment is the cheaper first step and it is designed to produce an honest answer including a negative one. A capability centre built for the wrong problem is expensive in a way that takes about two years to become visible.
Related reading
Find out whether the arithmetic has changed for you
If you ruled this out more than three years ago, the numbers you ruled it out on have moved. Tell us the roles you cannot fill and we will model it honestly.