Data center operators face a constraint that traditional economics can't address: power availability determines where facilities can be built, regardless of cost considerations. In Northern Virginia, lead times to power new facilities exceed three years. California projects requiring new substation work face interconnection processes stretching beyond five years.
US data center power demand sits around 30 GW in 2025. By 2030, projections show 90+ GW. That growth exceeds California's entire current power demand. Grid capacity determines what's possible—specifically, whether the grid can deliver power within project timelines.
Grid interconnection creates the binding constraint. More than 12,000 active projects are seeking interconnection, totaling 1,570 GW of generator capacity. The average duration from interconnection request to commercial operation has approached five years. In many jurisdictions, delays stretch into the decade-long range. Planning and construction of new transmission lines takes five to ten years, with major US utilities facing $436 billion in transmission and distribution spending through 2029.
Organizations evaluate locations using traditional metrics: land costs, tax incentives, labor markets, connectivity. Then power constraints extend construction timelines by 24 to 72 months in key markets. Where deployment windows matter—where business timelines demand capacity that grid timelines can't deliver—location preferences give way to power availability.
By the end of 2024, 62% of data centers were evaluating on-site power, with 19% already implementing behind-the-meter solutions. Organizations deploy mobile gas turbines delivering 100+ MW directly to facilities. One developer is planning a 362-acre campus incorporating hydrogen fuel cells, microgrids, and battery storage for primary power generation.
When the alternative is waiting five years for grid interconnection while competitors deploy capacity, paying premiums for on-site generation becomes rational.
These solutions cost more than grid power—sometimes significantly more. Power availability within deployment timelines drives the decision. What looks expensive on a per-kilowatt basis becomes necessary when time is the binding constraint.
Organizations evaluate Charlotte or Mumbai instead of their preferred Northern Virginia locations—those markets have power available within timelines that matter. About 70% of US data center capacity is projected to be located in four regions by 2028, driven by where power infrastructure exists rather than where traditional economics would suggest building.
The cascade extends beyond location decisions. A 24-month power delay pushes deployment into year three. Product launches that depend on that capacity get deferred. Customer commitments that assumed specific timelines require renegotiation. Engineering resources planned for optimization work get redirected to evaluating alternative locations and power solutions.
Time constraints override traditional economic factors. Physical infrastructure dictates location. The flexibility to evaluate options disappears when deployment windows close faster than grid infrastructure can be built.
Transmission infrastructure investment is accelerating, but planning and construction timelines remain measured in years. Organizations making deployment decisions today face power constraints that won't ease within their planning horizons. Getting power within the window that matters becomes the primary question—and what you're willing to spend in operational complexity and premium costs to make that happen determines what's possible.