From £1.1bn Headline to Contractor Demand: Reading the AI Data-Centre Boom

On 10 June 2026 the government announced a £1.1bn AI hardware plan, headlined by Apatura's plan to convert the former Ravenscraig steelworks in Scotland into one of the UK's largest green AI data centres. If you run an electrical or M&E business, an announcement like that is worth more than a passing glance, because the demand behind it feeds directly into what a buyer will pay for your business.

This is one of the four forces I set out in our 2026 Electrical Exit Window briefing. Here I want to look at it on its own: what the numbers say, what they do not say, and how an acquirer reads them.

The Numbers, Read Honestly

The £1.1bn plan does not sit in isolation. The wider announced pipeline runs to roughly £14bn of data-centre projects tied to the national AI action plan, with named commitments from Nscale, Vantage and a £22bn Microsoft investment under the UK-US technology partnership. Taken together they describe a serious, multi-year build-out of power-hungry infrastructure that someone has to wire, test and maintain.

Here is the honest caveat, and it matters. These are announcements, not delivered spend. At least one industry analysis has flagged the gap between what gets announced and what actually breaks ground, and any electrical contractor who has waited on a stalled scheme knows the difference. The right way to read the figures is as a direction of travel for demand, not a guaranteed order book.

The Grid Behind the Data Centres

Data centres are only half the story. They cannot run without grid capacity, and the grid itself is being reordered. In April 2025 Ofgem approved a connection reform package that replaced the old first-come-first-served queue with a "first ready, first needed, first connected" model. The queue held around 770GW of capacity as of January 2025, far more than a 2030 clean-power system needs, and the network operator aimed to issue its 2030 connection offers by early 2026.

The effect on the ground is that high-voltage and grid-connection capability is being repriced as the queue clears for projects that will genuinely build. You can watch that happening in the deal market. In May 2026, Triton-backed OCU Group acquired Athena PTS, a Cumbrian firm delivering power-systems engineering and contestable infrastructure up to 132kV. OCU turned over £886.5m in the year to April 2025 and is mapping a path to £1bn.

When an infrastructure-backed platform buys 132kV capability rather than building it, that tells you exactly what the market values: certificated, deliverable capacity against pipelines a buyer has already won.

What This Means If You Are Thinking About Selling

The temptation is to file all this under "good for the big grid players, irrelevant to a regional contractor". That would be a misreading. Buyers of electrical businesses are not buying your current order book; they are buying a position in a sector, and they price that position on the strength of sector-wide demand. Every announcement that hardens the infrastructure story strengthens the investment case behind every electrical acquisition in the country.

What converts that demand into value is compliant, certificated delivery. A business with a live NICEIC or NAPIT registration, an EICR and fixed-wire testing book, framework or repeat-client positions, and a retained team of qualified electricians is selling exposure to structural demand in a form a buyer can bank. Project turnover on its own has never commanded the same premium, and in a market this alert to delivery capacity, the gap is only widening.

You do not need a data-centre contract to benefit from the data-centre boom. You need a business a buyer can see plugging into it: certificated, recurring, and not dependent on you being in the room.