Summary
- Apatura claims almost 2,100 permanent on-site roles across five proposed Scottish campuses.
- The figures sit alongside much larger estimates for construction, indirect employment, and economic output that remain modelling rather than existing jobs.
- Scotland's current policy debate is placing greater scrutiny on how hyperscale employment and infrastructure benefits are calculated.
Apatura says five of its proposed Scottish data centre campuses would employ almost 2,100 people directly on site once operational, publishing a site-level breakdown as the economic case for hyperscale development comes under increasing scrutiny.
The developer’s reported figures include 567 on-site staff at a proposed 550MW Ravenscraig campus, 501 at Westerhill, 399 at Freeport West Lothian, 332 at Wester Hermiston near Edinburgh, and 297 at Larbert.
Those figures are projections rather than current jobs. They also need to be kept separate from wider estimates that include construction activity, indirect employment, and roles supported elsewhere in supply chains.
That distinction has become politically significant because large job numbers attached to data centre projects can combine very different categories of work. A four-year construction programme may sustain thousands of job-years without producing the same number of permanent operational positions once the facility is complete.
Apatura’s site-by-site on-site headcount at least provides a more specific number against which later planning, construction, and operating outcomes can be compared.
Ravenscraig carries the largest claim
The largest individual proposal is Ravenscraig in North Lanarkshire, where Apatura is promoting a 550MW data centre campus on the former steelworks site.
The company has previously published broader economic modelling for Ravenscraig, including direct, indirect, and induced employment and gross-value-added estimates. Those calculations depend on assumptions about construction spending, procurement, wages, and multiplier effects, so they should not be treated as equivalent to a payroll count inside the data centre fence.
The new on-site figures narrow the question but do not settle it. Staffing requirements differ substantially between facilities depending on operating model, security arrangements, maintenance strategy, customer mix, automation, and whether functions are performed by employees, tenants, or contractors.
Power density can also change the relationship between megawatts and headcount. A highly automated hyperscale facility does not necessarily employ proportionately more people simply because its IT load is larger.
DataCentral has already examined the employment debate around the UK pipeline, including the gap between large headline job forecasts and the smaller number of roles that may be permanently located inside operating facilities. It has also covered the planning process around Apatura’s Larbert proposal.
Infrastructure benefits face a harder test
Employment is only one element of the economic case for a data centre, but it is among the easiest claims for communities and planning authorities to understand. Other claimed benefits — business rates, construction spending, renewable-energy investment, supply-chain activity, and connectivity — are harder to compare across projects.
Apatura is also proposing community-benefit funding based on £6,000 per MW of data centre capacity each year, applying a benchmark derived from the Scottish Government’s proposed approach for onshore wind.
That creates an unusual link between electrical capacity and local payments. It may make benefit calculations more transparent, but it also means the eventual sum depends on what capacity is actually built rather than the maximum size proposed during development.
The current portfolio remains a pipeline, not an operating estate. Projects are at different points in planning, consultation, and development, and the infrastructure needed to serve them — particularly electricity — would be substantial.
The employment figures therefore provide a clearer developer claim rather than proof of delivered economic impact. If the campuses progress, the most useful comparison will come later: projected on-site staffing against actual operating headcount, and promised wider benefits against what reaches host communities once the facilities are energised.
