Every founder building custom software in India eventually sits across from the same three quotes. A freelancer found through a referral, charging INR 1,800 an hour, available next week. An offshore agency in Bangalore or Pune with a deck full of Fortune 500 logos, a pre-built three-month plan, and a project manager already assigned. A boutique studio with five people on the wall, a six-week waitlist, and a price that's two-and-a-half times the freelancer.
The temptation is to read the three quotes as a price comparison. That's the wrong frame. They are three different products, not three prices for the same product. This post is the framework we walk through with founders and operations leaders who are trying to figure out which of the three to actually hire — and, more often than not, in what order.
It assumes you've already ruled out hiring a full in-house engineering team. That's a different decision, made on a different timeline, for a different stage. This is for the moment in between: you have a product idea or an internal ops problem worth real money, and you need someone outside the company to build it.
The three models, described as honestly as we can
Before any vendor conversation, name what each of the three is actually selling. Most disappointment in this category traces back to buying one product and expecting another.
The freelancer
A freelancer is one senior practitioner, usually working from home, juggling two to four clients, billing in hours or fixed sprints. The good ones are excellent in their lane — a senior React engineer, a Shopify developer, a data engineer, a brand designer. They will close tickets you write faster than any agency. They will not write the tickets for you, run a roadmap, hire a second person, or own a product.
The freelancer model has a hard ceiling at "one person can hold it all in their head." Above that ceiling — multiple skills, multiple weeks of continuity, real product judgment — it stops working, even if the freelancer is brilliant. People try to scale freelancers by hiring three at once. It almost always ends with the founder becoming the project manager, the QA, and the integration tester for three contractors who don't talk to each other.
The offshore agency
The offshore agency is the model most Indian software shops default to. Fifty to several thousand engineers on staff, sales-led, project-managed by a layer of account managers and PMs, billed by staffed team capacity per month. The pitch is reasonable: you get a named team, a process, a contract, and someone to escalate to when things slip.
In practice, what offshore agencies are actually good at is a narrow thing: executing a well-specified backlog at moderate cost with predictable cadence. If you have a clear product, a clear roadmap, an in-house product owner who can write good tickets and accept work, and you need three to ten engineers for six to eighteen months, this model works well. India has a deep candidate pool for this — NASSCOM's strategic review has tracked the steady-state size of the export-led services industry for two decades, and the operational playbook for staffing this kind of engagement is mature.
What they are not good at is the messy front half of building something new. The discovery is shallow because the salesperson is incentivized to close. The first sprint plan is built before anyone has thought hard about the product. The senior people in the pitch tend not to be the people on the project once the SOW is signed. The PM is usually a coordinator, not a product thinker. None of this is malicious; it's how the unit economics of a hundred-person agency force the work to happen.
The boutique studio
A boutique studio is small on purpose. Five to twenty senior people, a partner-led engagement model, working on a handful of clients at once because anyone they bill out is also someone with judgment they're spending. They scope by outcome (a launched product, a working internal system, a milestone) and not by team-month. The senior people in the pitch are the people on the project, because there are only senior people.
What you're buying from a studio is judgment in the gaps. Which feature to cut so the launch ships in twelve weeks instead of twenty. When the second integration is a mistake. When the database design will hurt you in a year. When the customer-facing copy needs a rewrite, not another iteration of the form. None of this is in the SOW, because none of it can be. It's the product of senior people having enough room to think.
The trade-off is real and worth naming: a studio cannot give you raw capacity. If you need ten engineers next month, a studio is the wrong answer. If you need the right product in six months, an offshore agency probably is.
The five questions that actually decide it
Skip the comparison spreadsheet. The decision usually collapses into five questions.
1. Do you have a product owner inside the company?
If yes — a founder who can write specs, an operations lead who knows the workflow cold, a CTO who can run sprint reviews — your bottleneck is probably execution capacity, and an offshore agency is the most cost-effective answer.
If no, and you're hoping the vendor will own the product, only a boutique studio is an honest fit. Offshore agencies will say they'll do this. They almost never do, because the senior people who could are running sales calls, not your project. A freelancer can't own a product because owning a product is more than one person's job.
2. Is the scope stable, or will it change as you learn?
Stable scope — a known set of integrations, a defined workflow, a feature list signed off by stakeholders — fits an offshore agency or a senior freelancer well. The work is execution, and execution is what those models are built for.
Changing scope — a v1 where you'll learn what to build by watching real users, an internal tool where the workflow itself will be redesigned during the build, a market category that didn't exist eighteen months ago — needs a partner who treats scope as a question, not a contract. That's a studio's whole value proposition, and it's the part of the work that an agency's change-control process actively penalizes.
3. What's your timeline, and what happens if it slips by a month?
If the answer is "we ship in twelve weeks or the round closes" or "we present to the board on April 15," your bottleneck is not capacity — it's coordination, decision-making, and the ability to cut scope on the fly without a change order. Studios are built for this. Agencies are built for the opposite: predictable cadence over predictable risk.
If the timeline is honestly flexible and the work is mostly known, an agency's predictable monthly burn-down is fine.
4. What's the long tail look like after launch?
Most software is more expensive in year two than in year one. The freelancer often disappears. The offshore agency stays, but the team rotates, and the people who built it are gone in six months. A boutique studio is the only model where the people who built it are still around to maintain it, because there are only twenty of them and turnover is low.
If the system you're building is going to be load-bearing for the company for years, weight that. Continuity is a real cost line that doesn't show up in the initial quote. A senior person who already knows the system is worth more per hour than a new senior person learning it for the third time. Our retainer model is built around exactly this: the people who built v1 are the same people maintaining and extending it eighteen months later.
5. What's your honest budget — total, not hourly?
Hourly rates are the wrong unit. Compare total cost-to-launch and total cost-of-ownership for year one.
A senior freelancer at INR 2,500/hour, working twenty hours a week for sixteen weeks, is roughly INR 8L for an initial build. The hidden cost is your time managing them.
An offshore agency staffing four engineers and a PM for sixteen weeks is roughly INR 30–50L for the same period. The hidden cost is the discovery work the salesperson promised the team would do but didn't, plus the gap-filling work you'll need to commission afterward.
A boutique studio milestone-priced for the same launch is usually INR 15–35L. The hidden cost is lower, because the model is designed around accountability for the launch working, not for hours billed.
The right comparison is "what does it cost to have something real in production in twelve to sixteen weeks." On that comparison, the three models converge much more than the hourly rates suggest, and the right choice falls out of the other four questions, not the price.
Where most founders go wrong
Three failure patterns show up over and over, almost regardless of vertical.
Hiring a freelancer for product work. A senior engineer is not a senior product person. The work ships, but the product is wrong, and the founder doesn't realize until launch. The fix is rarely to swap freelancers; it's to admit the work needed a different model.
Hiring an offshore agency without a product owner inside the company. The agency will execute whatever spec you hand them. If the spec is bad, the product is bad, and the agency is correct to hand it back to you and say "you signed off on this." Six months later you're paying the same agency to rewrite half of it.
Hiring a boutique studio for raw capacity. A studio's whole model is small senior teams. If what you actually need is ten engineers throwing themselves at a known backlog, a studio is the wrong tool, and the engagement will feel slow and expensive. The studio is being honest; you bought the wrong product.
The good news is that the right model is usually obvious once you walk through the five questions in order. Most disappointment is not because the vendors are bad. It's because the model didn't match the work.
A note on India specifically
The Indian software services market is genuinely deep, and the talent pool is real. What's harder to find — and the reason this post exists — is the boutique tier. The offshore agency category has been a global export industry for thirty years, with hundreds of mature players. The freelancer market has consolidated onto Upwork, Toptal, and a handful of Indian platforms. The boutique tier is the thinnest layer, partly because the unit economics of staying small are hard to defend against the pull to grow into an agency.
This is also where the most leverage sits for an Indian founder building for an Indian market. Global agencies don't understand UPI or GST or how Indian customers actually pay. Global freelancers don't either. A studio that builds for Indian clinics, Indian law firms, Indian field service operators, or Indian D2C founders is making different product calls in week three than a global team would. That's the part you can't outsource to a generic offshore body shop, and it's the reason the boutique model exists at all.
The shortest possible decision tree
If you only remember one thing from this:
- One person, three weeks, no judgment required: freelancer.
- You have a product owner, you know the spec, you need capacity: offshore agency.
- You don't have a product owner, the spec will change, the launch matters: boutique studio.
- Year two onwards, regardless of who built it: the team that built it, on retainer.
The mistake almost no one makes is overpaying for a studio when an agency would do. The mistake everyone makes is the other direction — hiring an agency or a freelancer because the rate is lower, then spending the savings (and more) on rework.
If you're trying to make this call right now and want to talk it through with someone who is not selling you a long contract, we're happy to walk through the five questions with you on a discovery call. Even if the right answer turns out to be a freelancer or an agency, the half-hour is worth it — and we'll tell you when the answer isn't us.