Build, operate, transfer or outsourcing: how a small company should choose
Outsourcing a project buys a result. Build, operate, transfer buys a team you will own later. How to tell which one fits, and what the transfer has to say to be worth anything.
Outsourcing a project buys you a result. Build, operate, transfer buys you an operation you will own later, with the team and the knowledge that come with it. For a company with a small technology team or none at all, the choice turns on one question: will you employ people to run this yourself within the next couple of years? If the answer is no, a defined project or a service arrangement is simpler and you avoid paying for an exit you will never take. If the answer is yes, this model was built for that, and the part that decides whether it works is the transfer document.
What is build, operate, transfer?
A provider builds the thing, runs it for an agreed stretch, then hands it to you. The idea is old and did not start in software. It has long been used both by businesses setting up in a country where they have no presence and as a way of delivering public projects such as a hospital or a toll road, with the provider contracted to build the asset, operate it, and transfer it on an agreed trigger (Morgan Lewis on build, operate, transfer models).
In software it usually looks like this. In the build phase the provider designs and builds the product, sets up hosting and working processes, writes the documentation, and helps you hire the people who will eventually own it. Hiring early is the point, because those people need to learn the system while it is being made. In the operate phase the provider runs it to an agreed service standard while your new team works alongside and takes on more each month. At transfer, the code, hosting, guides, accounts and day-to-day ownership move to you.
The commercial shape follows the phases. Our own build, operate, transfer arrangement has a fee for each of the three phases, a build fee, a monthly operate fee and a transfer fee, all agreed before work starts, so the cost of the handover is known at the beginning rather than discovered at the end.
How is it different from outsourcing a project?
Outsourcing a project is a trade of scope for a result. You describe what you need, the provider builds it, you accept it, and the relationship either ends or continues as maintenance. The provider keeps the way it works. You keep the output.
Build, operate, transfer trades differently. You are buying the capability as well as the output: the team, the processes, the documentation and the running knowledge. That is why it costs more than a project of the same size and why it takes longer to finish. You are paying to stage a commitment you are not ready to make all at once.
Three practical differences matter more than the definitions.
- Knowledge. In a project, it stays with whoever built the thing unless the handover was specified in detail. Here, moving it across is the deliverable rather than a courtesy at the end.
- Hiring. A project needs nobody on your payroll. This model needs real people named during the build phase, because somebody has to receive the transfer.
- What the contract spends its words on. A project contract is mostly scope and acceptance. A build and operate contract lives or dies on the third phase, which is the phase providers tend to leave vague.
When does each model suit a small company?
Most of the time the honest answer is one of the simpler arrangements.
Your situation | Usually fits | Why | What to watch |
|---|---|---|---|
You want a working product now and will hire people to run it within a couple of years | Build, operate, transfer | The handover is the purpose of the arrangement, not an afterthought | The transfer terms, and whether anyone will be ready to receive it when the date comes |
The work has a clear end: a website, an app, a system with a defined scope | A fixed-scope project | Ownership moves at the end with the source files and logins, and there is no operate phase to fund | That the handover list is written into the contract, not promised in a call |
You want the thing run, watched and kept current, and you never want to hold it yourself | A managed service or care retainer | Nobody has to be hired or trained, and small fixes stop being events | What happens to your code, data and accounts if you change provider |
You already have in-house people and need more hands on a roadmap you own | A dedicated team | Your side keeps the knowledge and the direction | Whether you are buying capacity or decisions, because the two are priced and managed differently |
You are not yet sure what to build, or who should build it | A second opinion or a part-time technology lead first | Deciding wrongly costs more than deciding slowly | That whoever advises you can be kept separate from whoever delivers |
You want an operation in another country and are not ready to register a company there | Build, operate, transfer | Staging the commitment before you incorporate is the original use of the model | Entity, employment and tax questions, which belong with counsel in that country |
Nobody in the business is ready to take anything over, now or later | Not this model | A transfer needs a receiver, and without one you pay for the option and never use it | Being sold the model without a real handover in it |
You can compare the arrangements side by side on our ways to work page. If your situation points at a smaller model than the one you came in asking about, that is worth hearing early.
What must the contract say about the transfer?
This is the part that decides whether the whole arrangement was worth anything, and most proposals leave it thin. A provider has little commercial reason to make its own exit easy, so vagueness tends to collect in the third phase.
Start with ownership, because the default position is probably not the one you assume. Under Chapter IV of the Copyright Act 1957, the author of a work is the first owner of the copyright in it, and an employer becomes the first owner only where the work was made in the course of employment under a contract of service. A provider building your product is not your employee. So the copyright in the code sits with them until it is assigned, and an assignment is valid only if it is in writing and signed by the assignor.
Writing it down is not enough on its own. The same chapter requires an assignment to identify the work and to specify the rights assigned, the duration and the territory. Silence has consequences that are set by the Act rather than by either party:
- If the period of the assignment is not stated, it is deemed to be five years from the date of assignment (Copyright Act 1957, Chapter IV).
- If the territory is not specified, it is presumed to extend within India only.
- If the person receiving the rights does not exercise them within one year, the assignment in respect of those rights is deemed to have lapsed, unless the document says otherwise.
So a company can pay a transfer fee, believe it has bought its product outright, and hold an assignment that runs out without anyone noticing. Ask for the period to be stated as the full term of copyright, ask for worldwide territory, and have the provider sign the assignment rather than describe it in a covering email.
With ownership settled, the remaining transfer terms are a short list. Agree all of them before the build phase starts, while you still have something to negotiate with.
- When the transfer can happen. Whether you can call it inside a window, or whether a fixed date decides for you. The window is the version you want.
- What the transfer costs. Fixed at signing. A price agreed later is a price agreed when you have no alternative left.
- Which people come across. Whether you can name individuals or have to take the whole team, and what happens to anyone who stays behind.
- Which charges stop. Put it in writing that the operate fee ends on the transfer date, then look for anything that would keep billing you in respect of staff you now employ.
- What the handover contains. Write the list down: runbooks, architecture notes, deployment steps, an inventory of accounts, and the reasoning behind the decisions that are hard to reverse.
- What the transfer moves. Assets, or a company. Where it is an asset transfer, the operating agreement needs exit obligations to carry the service across, and a transitional services arrangement to cover the gap afterwards (Morgan Lewis on build, operate, transfer models).
- What happens if you walk away. What is payable and what you keep if the transfer never happens. This clause gets read once, at the worst possible moment.
We implement arrangements like these and we will tell you where the terms look thin, but the legal interpretation of an agreement belongs with counsel, and anything touching a company, employment or tax in another country belongs with counsel there.
What goes wrong at handover?
Failures at transfer are rarely technical. Four patterns account for most of them.
The first is that nobody was hired. The build and operate phases run, the date arrives, and there is no team to hand to. Hiring slipped because the product was working and the gap never felt urgent. Naming the receiving people during the build phase and giving them real work, while the provider still carries responsibility for the outcome, is what prevents it.
The second is knowledge that lived in conversations. Everything worked while the people who built the system were still answering questions, and ordinary tasks became research once they stopped. Documentation written at the end is a summary of what someone remembers. Written as the work happens, it is a record.
The third is assumed ownership. The accounts are in the provider's name, the domain renews on their card, the repository sits in their organisation, and the copyright was never assigned in writing. Each one is easy to sort out early and awkward to sort out while a relationship is ending.
The fourth is treating the handover as a single day: a meeting, a folder of files, a signature. Transfers that hold up happen gradually. Your team picks up tasks during the operate phase, breaks things while someone experienced is still on hand, and is already running most of the system before the formal date. If you are inheriting something that was handed over badly, our guide on how to take over a stalled software build covers what to check first.
What are the alternatives, such as a dedicated team or a fractional CTO?
If your situation points somewhere other than build, operate, transfer, these are the arrangements worth knowing.
- A fixed-scope project suits work that has an end. Insist that the handover includes every source file, account and credential, and that it says so in the contract rather than resting on custom.
- A dedicated team is a group working only on your roadmap, managed by you and employed by the provider. It suits a company that already holds technical direction in-house and needs capacity.
- A managed service or care retainer keeps the thing running, updated and monitored, for a business that wants the result without the responsibility.
- A part-time technology lead, or fractional CTO, works on your side of the table: deciding what to build, judging whether the result is any good, managing providers and helping you hire. Our guide to what a fractional CTO does in the first 90 days sets out the early months, and the fractional CTO page explains how the arrangement runs. It is often the right first step for a company that cannot yet tell which delivery model it needs, because the decision matters more than the model.
A reasonable sequence for a company with no in-house team is to get advice first, run one defined project to see how the working relationship holds, and only then agree a build and operate arrangement for the stage that has to end in your own hands.
If you are weighing these up for something specific, tell us what you are building and who you expect to run it, and see how we work. We will say plainly which arrangement fits, including when it is a smaller one than you asked about.
Questions
Is build, operate, transfer worth it for a small team?
Only if you intend to employ people to run the thing yourself, and reasonably soon. The arrangement exists so that ownership moves to your side at the end, and you pay for that option across all three phases. If nobody in the business will be hired or trained to take it over, a fixed-scope project or a care retainer gives you the same working product without the cost of an exit you never use.
Who owns the code during the build and operate stages?
Whoever the contract says, and if the contract is silent the answer is probably not you. Under the Copyright Act the author is the first owner, and an employer becomes first owner only for work made under a contract of service, which a vendor relationship is not. So the code a vendor writes stays the vendor's until there is a written assignment signed by them. Settle it at signing rather than at handover.
What happens if we do not take the transfer?
That depends entirely on the exit terms, which is why they are worth reading before the first phase starts. A good agreement says what you owe if you walk away, what you keep, and whether you can carry on as an ordinary service arrangement instead. An agreement that says nothing leaves you negotiating from a weak position, because by then the operation is running and you are not running it.
Can we switch from outsourcing to build, operate, transfer later?
Often, yes, and it can be the sensible order: run a defined project first, see how the working relationship holds up, then agree a build and operate arrangement for the next stage. The thing to check is that the first project handed over everything, source files, accounts and documentation, so you are starting the second stage on your own property.
Filed under Tech Consultancy. The service behind it: Tech Consultancy.

