European power panels at Datacloud Global Congress 2026 delivered a stark consensus: power availability has overtaken fibre connectivity as the primary determinant of where data centres are built. For European operators in particular, the gap between grid planning timelines and construction timelines has become a structural crisis, not a temporary bottleneck.
For most of the past decade, site selection for a data centre was a connectivity question first. Find the fibre, then find the power. That logic has been inverted. In market after market, panellists at DCGC 2026 described a landscape in which power availability not only determines site selection but increasingly determines whether a project proceeds at all.
Grid connection delays in Europe now routinely run to seven to fifteen years from planning to energisation. Data centres, by contrast, can be designed and built in 18 months to three years. The gap between these timelines is not a coordination problem; it is a structural mismatch between the pace of digital infrastructure demand and the pace of energy network investment.
Compounding this, large power transformers (the long-lead critical equipment at the heart of every grid connection) now carry delivery lead times of five to six years. Operators who have not already secured manufacturing slots are, in some cases, effectively locked out of the next development cycle.
The ‘power first’ shift in site selection
The panel was unambiguous that the industry has moved to a power-first approach to site selection. Connectivity remains important, but bringing power to an established connectivity hub has become the harder problem. The practical implication is that operators are increasingly looking at sites that have available grid capacity, even if that means secondary or tertiary markets without existing data centre ecosystems.
In Europe, this is driving renewed interest in regions with decommissioned heavy industrial infrastructure, former coal and gas plant sites where grid connections already exist and may be reactivated at lower cost and on shorter timelines than greenfield connection requests.
The panel also noted that local authorities in some markets are now requiring evidence of secured power before granting planning permission. A “bring your own energy” dynamic that adds further pressure to early-stage power strategy.
Behind-the-meter and onsite generation as a structural response
With grid connections constrained, operators are investing in behind-the-meter and private power solutions at scale. The panel described onsite generation not as a transitional measure but as a permanent component of power strategy, both to bridge grid connection delays and to provide grid services that improve the commercial position of the facility.
Battery energy storage systems (BESS) were identified as a particularly important tool, enabling operators to draw on stored energy during peak demand periods and support grid stability during ramp-up. AI-driven optimisation of these systems was flagged as an emerging competitive advantage, with predictive load management and dynamic cooling control capable of materially reducing power consumption and improving the economics of behind-the-meter assets.
The regulatory and collaboration gap
The Power Innovation panel, which focused specifically on reducing energy costs in the UK data centre sector, identified government levies and non-commodity charges as areas where policy reform could deliver a 30% to 50% reduction in energy costs. A national plan aligning data centre development with regions that have available power capacity and integrating future generation from small modular reactors (SMRs) and wind was proposed.
The panel also raised the issue of regulatory uncertainty around grid fees for private wire and behind-the-meter assets, citing Germany as an example of a market where policy shifts have materially affected investment models. The call was for outcome-based regulation rather than technology mandates, and for greater real-time data sharing between data centre operators and grid operators to optimise connection processes.
Several panellists argued that the industry now needs to be as sophisticated about energy as it is about digital infrastructure, and that this means hiring energy expertise as a core organisational capability, not outsourcing it.
The power constraint is not going away. Even optimistic projections for grid expansion in Europe suggest the gap between demand and grid capacity will persist into the early 2030s. Operators who move fastest on behind-the-meter solutions, who secure equipment manufacturing slots early, and who build genuine energy expertise in-house will have a structural advantage over those waiting for the grid to catch up. The broader implication for digital infrastructure investment is significant: power is now the scarcest resource in the sector, and the ability to access and manage it has become a competitive moat.
Key Takeaways
- Power availability has structurally displaced fibre connectivity as the primary site selection criterion for new data centre development, with grid connection timelines of seven to fifteen years now the defining bottleneck in Europe.
- Transformer lead times of five to six years mean that operators who have not already secured manufacturing slots face a materially constrained development pipeline in the near term.
- Behind-the-meter and onsite generation have moved from transitional solutions to permanent components of power strategy, with BESS and AI-driven load management providing both grid support and commercial advantage.
- UK energy costs, currently several times higher than those in comparable European markets, represent both a policy failure and a structural opportunity if government intervention delivers on proposed reforms.
- The panel advocated for outcome-based regulation, real-time data sharing between operators and grid operators, and for the industry to treat energy expertise as a core in-house capability rather than a procurement function.
