Power, Not Land, Is Now the Binding Constraint on Data Centre Property
For much of the past decade, the competitive advantage in data centre real estate could be summarised in two words: land and capital. Developers raced to assemble land banks near fibre routes and subsea cables, and investors competed on the strength of their balance sheets. Those advantages have not disappeared, but they are no longer decisive. The binding constraint has moved to something less visible and far harder to acquire: electricity.
Across many markets, the ability of electricity networks to connect new load has not kept pace with demand for computing capacity. The result is a structural shift in how data centre projects are conceived, financed, valued and leased. Power procurement, once an engineering afterthought handled late in the development process, has become a core property skill — and increasingly the factor that separates a credible pipeline from a speculative one.
Grid Queues Have Become the Real Development Timeline
The traditional sequence ran from site acquisition to planning consent, then construction, then utility connection. In constrained markets that order has effectively been reversed. Developers now pursue interconnection agreements, capacity reservations and supply commitments years before planning applications are submitted, because a site without a viable route to power is not a deliverable project regardless of its other merits.
This changes what counts as a good site. Proximity to substations, existing transmission capacity and the condition of local distribution infrastructure now weigh alongside connectivity, latency and labour supply. Sites previously dismissed for their distance from established clusters are being reconsidered because they offer a clearer path to energisation. Diligence has broadened accordingly: understanding a network's headroom, the queue ahead of a project and the likely timeline for reinforcement works is now as important as understanding zoning or construction costs.
Power Is Entering the Lease and the Underwriting
Tenants, particularly large cloud and artificial intelligence operators, are no longer content to sign leases that promise space and leave energy supply to be resolved later. They want certainty on available capacity, on the resilience of supply and, increasingly, on the price of power over the life of the lease.
That is changing how leases are structured. Provisions addressing capacity allocation, supply interruption, energy pricing mechanisms and responsibility for securing or expanding connections are becoming more detailed. Where a landlord cannot demonstrate a credible route to long-term supply, its negotiating position weakens — and in some cases the transaction cannot proceed at all.
The effect flows into underwriting. An asset's value depends on its ability to deliver contracted capacity over time, which means power arrangements sit alongside rent rolls and lease covenants in the assessment of income durability. A building with secure, competitively priced supply supports a stronger covenant profile than one dependent on uncertain grid upgrades or short-term market pricing. Assets with constrained or expensive power face valuation pressure that conventional appraisal models may not fully capture. Energy arrangements and utility relationships are moving into the core of investment committee papers rather than sitting in technical appendices.
On-Site Generation Moves From Extra to Infrastructure
Where grid capacity is scarce or slow to arrive, behind-the-meter generation has shifted from a sustainability credential to a practical requirement. On-site gas generation, fuel cells, battery storage and, where conditions allow, renewable generation with storage are being deployed not primarily to reduce emissions but to guarantee that a facility can operate. Battery storage in particular provides backup and helps operators smooth demand peaks.
Heat reuse systems, which capture waste heat from server halls for district heating or adjacent facilities, are moving from demonstration projects into standard design considerations in some European markets, where planning authorities increasingly expect them. Efficiency carries a similar logic. Power usage effectiveness has long been a headline metric, but it now has direct commercial consequences: a facility that uses less power per unit of computing capacity can fit more revenue-generating load within a fixed connection.
Geography Is Shifting Towards Secondary and Edge Locations
Where power is available, development follows. Established clusters that built their reputations on connectivity and ecosystem depth are running short of capacity, and the queue for new connections can stretch beyond the horizons tenants and investors will accept. Demand is redistributing towards secondary markets and edge locations where power is more readily accessible, planning regimes are more accommodating and land costs are lower.
For secondary markets, this is an opportunity. Regions with surplus generation capacity, underused industrial land and supportive local authorities can attract investment that would previously have gravitated to the largest hubs. The shift is not uniform: latency-sensitive workloads will continue to cluster near major population centres, fibre intersections and financial exchanges, and those locations will retain premium positioning where power can be secured. But for a growing share of capacity, the calculus now favours places where electricity is obtainable.
Obsolescence Risk and the Widening Gap
The same forces are raising questions about older stock. Facilities built when power was abundant and cheap may lack the efficiency, resilience or capacity to compete. Where they sit in constrained networks, retrofitting them — upgrading cooling, adding storage, renegotiating supply — can require substantial capital expenditure, and in some cases the economics do not justify it.
This creates a divergence within the market. Well-located, efficiently designed assets with secure power arrangements are becoming more valuable, while less efficient facilities face growing obsolescence risk. The gap is not merely about age; it is about whether an asset can demonstrate a durable route to competitive energy supply.
The broader consequence is a separation of pipelines into those that can be delivered and those that cannot. Announcements of planned capacity are abundant, but credibility increasingly depends on evidence of secured power — an interconnection agreement, a capacity reservation, a generation strategy. Investors and tenants are learning to distinguish between the two.
For a sector watched closely within the Global Property Awards category, the convergence of real estate and energy infrastructure is becoming central to how projects and companies are assessed. The ability to secure, structure and manage power is no longer a technical detail appended to a property strategy. It is becoming the strategy itself — and the operators who master it will define the next phase of the data centre market.