Build cost

Is offshore development actually cheaper?

Last updated 12 October 2026

Start with the finding, because it determines how much to trust everything else: no official or academic dataset comparing offshore and UK software development rates appears to exist. Search for one and you get outsourcing agencies and brokers — thescalers, devoxsoftware, uvik, yourteaminindia and a dozen more. Not one statistical agency, trade body or academic source.

The most carefully presented of them states its method, which is more than most: a survey of 60 software development partners, published November 2025, by a company whose business is introducing clients to offshore suppliers. Sixty self-selected vendors, collected by a party with a commercial interest in the answer, is not a market measurement. If an article quotes those ranges, the sample size and the publisher’s role belong in the same sentence.

The one like-for-like comparison that is public

There is a genuinely clean source, and almost nobody uses it. Suppliers on the UK government’s G-Cloud framework must publish a pricing document, and the larger ones price both onshore and offshore delivery in the same table, for the same roles, on the same day, under the same commercial terms. That is a controlled comparison, published, and dated January 2026.

One supplier, maximum rates on a government framework, 7.5-hour day, excluding VAT, January 2026. A second supplier on the same framework publishes software development as a band: £250 to £1,100 onshore against £60 to £855 offshore — and note that the top of that offshore band overlaps most of the onshore one. Senior offshore people are not cheap. Junior offshore people are very cheap.

So the published ratio is roughly 2.3 to 1, consistently, across grades. That is a real saving. It is also about half the gap implied by the headline numbers in vendor rate guides, and it is a gap on rates, before anything is said about how many hours each side needs.

What the rate gap does not include

The costs that erode an offshore saving are real but they are not measured anywhere credible, so this section is reasoning rather than data and is labelled as such. They land on your side: writing a specification precise enough to be built by people who cannot tap you on the shoulder; reviewing what comes back; the latency on every decision that needs an answer from you and arrives while you are asleep; and rework when an assumption was wrong.

The nearest thing to evidence is general rather than offshore-specific. A study of 1,471 IT projects published in the Harvard Business Review in September 2011 found an average cost overrun of 27 per cent and one project in six running far beyond that. The sample was 83 per cent US-based and 92 per cent public sector, so it is a caution about large projects rather than a measurement of offshore delivery. Anyone telling you offshore projects overrun more, or less, with a number attached, is making it up.

The comparison worth making instead

Rate per hour is the wrong unit, because it prices the input rather than the result, and the side quoting the lower rate is not the side carrying the estimating risk. A fixed price per stage moves that risk to whoever did the estimating, which is the right place for it and the reason we quote that way.

If you want to compare properly, get a scoped, fixed-price figure from each side for the same defined outcome, and ask both what happens when the estimate is wrong. The answer to the second question is usually more informative than the first.

Sources: Crown Commercial Service G-Cloud 15 supplier pricing documents, January 2026, both quoted suppliers publishing onshore and offshore rates in one table · Accelerance, “2026 Outsourcing Rates”, 24 November 2025, n=60 self-reported supplier partners, publisher is an outsourcing broker · Flyvbjerg and Budzier, “Why Your IT Project May Be Riskier than You Think”, Harvard Business Review 89(9), September 2011, n=1,471, 83% US, 92% public sector. No official or academic dataset comparing offshore and UK development rates was found; that absence is reported rather than filled.