How do we stop it becoming a black box?

Insist on your repository, your board and your CI. If progress is only visible through a supplier's portal or a weekly slide, you have already lost sight of the work. Four hours of daily overlap and access to every engineer, not through a coordinator, are the other two conditions.

Can an offshore team become ours later?

That is what build-operate-transfer exists for. The thing to check is whether the transfer terms — date, price, mechanism — are in the first contract or left to be negotiated when you want to leave. Only the first is really BOT.

Offshore team or in-house hires: which gets you engineering sooner

The comparison is usually framed as cost. It rarely is. It is about how quickly you need capability, how long you need it, and who carries the risk when someone leaves.

How we engage

The short answer

Hire in-house when the capability is permanent and central and you can wait three to six months per hire. Use an offshore or dedicated team when you need capability in weeks, when the need may not be permanent, or when you want retention risk carried by someone else. Most organisations end up with both, and the useful question is which roles belong in which column.

Offshore / dedicated team against In-house hiring.

	Offshore / dedicated team	In-house hiring
Time to productive	2–6 weeks.	3–6 months including notice periods.
Retention risk	The supplier's — replacement and overlap are their cost.	Yours, including the re-hire and the lost context.
Scaling down	Contractual notice, usually 30–90 days.	Slow, expensive and bad for morale.
Depth of context	High if the team is stable; low if the supplier rotates people.	Highest, and it compounds.
Cost shape	Operating expense, predictable monthly.	Salary plus benefits, recruitment, equipment and management overhead.
Cultural fit	Managed deliberately through overlap and process.	Native.

Day-rate comparisons flatter in-house because they omit recruitment, equipment, benefits, management time and the cost of a vacancy. Compare fully-loaded cost per productive engineer-month, including the months a seat sits empty.

When each is the right answer.

Each line below is a condition, not a preference. The matrix above scores the same choice on every axis a buyer weighs.

Choose offshore / dedicated team when

You need engineering capacity this quarter, not next year.
The workload is real but may not be permanent.
You want someone else to own recruitment, bench and replacement.
You want to test a capability before committing headcount to it.

Choose in-house hiring when

The role is central, permanent and shapes the product direction.
You are building institutional knowledge you never want to lose.
Your governance requires employees rather than contractors.
You already hire well and quickly, which is rarer than most teams believe.

Follow-ups

If cost is the whole proposition, the thing being bought is capacity rather than ownership of an outcome, and the usual consequences follow: rotation, no domain knowledge and quality you have to police yourself. What is worth buying is a team that stays long enough to learn your domain and a supplier who holds the retention risk. Those cost more than the cheapest rate and are worth it.

Other comparisons

In writing

Discovery ends in a recommendation, including the one to buy.

Where a packaged product covers the need, that is what the document says. The artefacts are yours either way, and they are written to be buildable by anyone.

Get the recommendation Send a brief instead

No sales deck
An engineer on the call, not an account manager
NDA before you share anything