Build a 15 to 50 Seat Shared Services Center in India

Planning a shared services center in India at 20 or 30 seats rather than 500? The published playbooks are all written for the large build. They assume a dedicated site, a captive entity and a management layer, and if you are a US or UK finance or operations leader moving your first processes across, none of that exists yet.

So let's take the band on its own terms: what moves first, what it costs at 15, 25 and 50 seats, how to hire and employ the team, and when your own entity starts to pay off.

Location selection, the compliance stack and the full step by step setup are in the complete guide to building a shared services center in India. This page stays on the seat count.

What counts as a 15 to 50 seat shared services center in India?

A single owned team running defined back office processes for a foreign parent, usually finance and HR operations. At this size it is one accountable lead with named process owners rather than a multi-tower organization, and it can run without an Indian entity of its own.

The work that moves first is the recurring, rules-based part of the cycle, because that survives a time zone gap without a conversation.

• Transaction processing: payables, receivables and expense workflows running on a daily or weekly cadence.
• The recurring close: reconciliations, journal preparation and the schedules a group controller signs off each month.
• Payroll and HR operations: the pay run, onboarding, offboarding and the employee service desk, which offshore HR shared services teams in India usually absorb first.
• Reporting support: management packs, variance commentary and the data pulls behind them.
• What does not move yet: judgment that needs a room, and anything still being defined. An undocumented process cannot be handed over.

The label matters less than the scope. Where the boundaries sit between a GBS center, a GCC and a shared services center is settled in a GBS center, a GCC and a shared services center, and it does not change what a 25 seat team does on a Tuesday.

The band is not an arbitrary slice either. Wisemonk's India GCC landscape report splits India teams into three size tiers, and 15 to 50 seats straddles the first two.

Tiers, models and timelines are current as of September 2026. The build operate transfer model fits 30 to 100 employees, which starts inside this range too.

Two segmentations both break inside 15 to 50 seats. The operating model choices above the seat count are in setting up a GBS center in India. Which raises the obvious question: why not just scale the large playbook down?

Why does the 500 seat playbook not transfer to this size?

Because three of the four things a large build plans around do not exist yet. There is no dedicated site, no captive entity on day one and no management layer, so the decisions that dominate a 500 seat plan are decisions you have not had to make.

• No dedicated site to plan around: on a 50 person build, lease and fit-out runs $200,000 to $250,000 one time, with IT and cybersecurity at $75,000 to $150,000. At 15 seats none of that is bought yet.
• The city still moves pay: where the team sits changes the salary line even with no lease to sign. That choice is worked through in the best Indian cities for offshore finance operations.
• No captive entity on day one: an Employer of Record suits 1 to 50 hires, goes live in 1 to 5 days and needs nothing upfront. The whole band fits inside it, so hiring employees in India without an entity is the practical start.
• One accountable lead, not a management layer: sourcing an India Head takes 30 to 45 days, against 1 to 2 weeks to a first working employee. That gap is the sequencing problem in one line, and the hiring timeline in India shows where the rest goes.
• Tooling is bought per seat, not per licence tier: the software line moves with headcount rather than sitting flat. That helps at 15 seats and quietly stops helping, because nobody renegotiates at seat 40.

None of that changes why you are doing it. The reasons companies set up a GCC in India hold at 25 seats as much as at 500.

What does it cost at 15, 25 and 50 seats?

There is no reliable per seat price at this size, and anyone quoting one is modeling. What you can pin down is the layers: pay and statutory contributions, the employment route, tooling per seat, and annual HR and compliance once the center is large enough to carry it.

Attributes reflect how builds at this size actually run, as of September 2026. There is no reliable per seat figure at 15 or 25 seats, so those cells describe behavior rather than carry a number.

• At 15 seats: the cost is almost entirely people. Salary plus statutory contributions is the number, and what the statutory side adds is in the cost of employment in India.
• At 25 seats: the entity conversation starts, but the cost structure does not change. A second function usually arrives, and the cost of an offshore finance team in India is the comparison for running two together.
• At 50 seats: real numbers start to exist. A 50 to 100 person center runs $500,000 to $3 million to set up, with 40 to 60% total operating savings against a US team. The detail sits on our page for global capability centers in India.
• Annual HR and compliance: $50,000 to $100,000 at that scale as of September 2026, and fixed overhead plateaus after about 30 people.

The pay and statutory layer carries the most weight at every seat count, and the employee cost calculator runs it against a real salary.

Where a build does reach that size, the cost of setting up a GCC in India carries the full layer by layer model.

Cost settled, the next question is who runs it.

How should a center of this size be staffed and led?

As one accountable lead with a named owner for each process, and nothing above that. A second line of management adds a reporting layer to a team small enough to be managed directly.

• Hire the lead early, but do not wait for it: it is the slowest hire in the plan, and that clock is set by seniority rather than by the size of what it runs.
• Give ownership to whatever runs on a fixed cadence: a payables cycle or a pay run has a defined output and a defined date, so an owner can be measured from week one.
• Hold back anything still being defined: moving an undocumented process early is how a small center ends up holding work it cannot finish.
• Skip the second manager: it costs a senior salary that would otherwise fund two more process owners, and below 50 seats the owners usually remove more risk.

How the staging plays out in practice is shown in a Singapore company scaling finance shared services into India.

How do you hire and employ 15 to 50 people in India?

Post the roles, screen against a written scorecard, and put the hires on an Employer of Record so nobody waits on an incorporation. At this size the constraint is rarely the talent pool. It is how fast an approved seat becomes somebody working.

• Write the scorecard before the job description: at this size every hire is visible, so define the skills, seniority and constraints that matter and screen against those rather than against a CV.
• Expect volume, not signal: a finance or operations role in India draws hundreds of applications, and the person you needed is rarely on page one. TalentScout ranks every applicant against your scorecard as it arrives and shows the reasoning behind each score.
• Check the salary band before you post: an offer built on stale pay data sits unfilled for four quiet weeks, and that is how a small center misses its ramp.
• Verify before anyone owns a process: identity, education, employment and court record checks matter more here than at 500 seats, because one person is the whole process.
• Leave the employment route until offer stage: employ on your own entity if you have one, or have an EOR hold the contract. That decision does not have to gate the hiring.

Which brings the entity question back, this time with a real headcount attached to it.

When does an Indian entity start to earn its keep?

Somewhere inside this seat band. Per-head EOR fees stop making sense past 25 to 30 employees, while the EOR route itself suits 1 to 50 hires. Both land inside 15 to 50, which is why the decision belongs to this page.

• What the EOR carries until then: the legal employment, payroll, statutory filings and contracts, which is the infrastructure a center this size has not built. See EOR services in India.
• What incorporating changes: employment, filings and liability move onto your own company. Expect 3 to 6 months with experienced local help, closer to 6 to 12 assembled in-house.
• What it costs upfront: entity setup runs $15,000 to $25,000, against nothing upfront and 1 to 5 days for an EOR. The ongoing compliance staff is the bigger number.
• How to decide: entity overhead is largely fixed, so cost per head falls as you add people, while an EOR fee is flat per head. EOR vs entity in India sets out both sides.

Run it against your own headcount and salary mix with the EOR vs entity calculator, which is usually what settles it.

When you do switch, the transition from EOR to a legal entity covers moving people across without resetting anyone's service.

What changes as the center crosses 30 seats?

Three things at once. Fixed overhead stops growing, space stops being deferrable, and the entity question moves from theoretical to a line somebody has to own in next year's budget.

• The per head number improves: the total keeps rising with headcount, but the share that is not pay stops rising. The center does not get cheaper. The next seat does.
• Space stops being deferrable: a Grade A or SEZ commitment becomes worth making once the team crosses 30 to 50 people, and an SEZ cuts effective total cost by 15 to 30%.
• The entity question gets real: at 30 seats the crossover band is the range you are standing in. How that call gets made is in a US product startup weighing an India EOR against a captive entity.

The two clocks do not have to run in sequence either, and hiring via an EOR while your India GCC is being set up runs both at once.

How can Wisemonk help you build a shared services center in India?

Wisemonk is an India-native Employer of Record (EOR) that helps global companies hire, pay, and manage talent in India without setting up a local entity.

For a center in this range, that means named process owners working your close, your payables and your HR operations within weeks, on compliant Indian contracts, without registering a company first.

And the entity decision stays open until the headcount argues for it, rather than being forced in month one.

We support 300+ global clients and more than 2,000 employees across India, process $20M+ in payroll, and hold a 4.8/5 rating on G2 across 261+ reviews. Pricing starts from $99 per employee per month as of September 2026.

Here is how we help:

• PEO: once your entity is live, payroll, Provident Fund, professional tax and TDS filings run under your own registration numbers, with equipment procurement alongside. From $49 per employee per month.
• Managed payroll: the India pay run and its monthly filings executed while your team keeps the entity and the HR function.
• Entity setup: incorporation and the tax and employer registrations, filed in your own company's name from the first form onward.
• Background verification: identity, education, employment and court record checks before somebody joins a team that will own one of your processes.
• Contractor of Record: engage independent contractors compliantly where a process is genuinely project-based rather than a standing seat.

GCC setup, managed payroll and entity setup are quoted to scope rather than listed at a price.