Why Data Center Acquisition is Critical Now
Explore how data center acquisition can meet the rising demand for digital infrastructure and unlock new opportunities.
The numbers don't lie: global data center capacity is under siege. AI workloads, cloud migration, and the explosion of connected devices have pushed power demand at hyperscale facilities to levels that would have seemed absurd five years ago. Utilities are struggling to keep up. Construction timelines stretch to three years or more. Yet capital keeps flooding in — because whoever controls the infrastructure controls the stack.
That's the core tension driving data center acquisition right now. Building new is slow and expensive. Buying and repositioning what already exists is faster, often cheaper, and — done right — strategically superior.
The Demand Is Real, and It's Not Slowing Down
Data center power consumption in the United States is projected to more than double by 2030, according to multiple grid operator forecasts. The culprit isn't streaming video or e-commerce — it's compute-intensive AI training and inference workloads, which consume anywhere from 10 to 100 times the energy per rack compared to traditional enterprise servers.
The bottleneck isn't money or even land — it's power interconnection queues that now stretch years deep in major markets.
Northern Virginia, the world's densest data center market, has seen municipalities impose moratoriums on new construction simply because the grid can't absorb more load fast enough. Phoenix, Dallas, Chicago — similar stories. This constraint is precisely why existing, powered facilities have become some of the most coveted assets in commercial real estate. A building already connected to a substation and permitted for critical infrastructure use isn't just a building. It's a shortcut worth tens of millions of dollars in time and capital.
For investors and operators, this creates a clear signal: the arbitrage opportunity in data center acquisition is a function of infrastructure scarcity, not just technology demand.
What Acquisition Actually Gets You
Buying an existing data center — or a facility that can be repositioned as one — compresses the development timeline dramatically. Where ground-up construction typically runs 24 to 36 months from site selection to energization, an acquired asset can often be operational within 6 to 18 months, depending on the scope of modernization required.
But timeline compression is only one dimension of the value proposition. The more nuanced benefit is risk-adjusted predictability: you know the power situation, the cooling infrastructure, the fiber connectivity, and the structural load capacity before you write the check. These aren't variables you're modeling — they're facts you can diligence.
Technology integration is the other major lever. Many facilities built in the 2000s and early 2010s were designed around Power Usage Effectiveness (PUE) ratios of 1.8 or higher — meaning for every watt delivered to a server, 80 cents of additional power was burned on cooling and overhead. Modern hyperscale facilities operate at PUE ratios closer to 1.2. Acquiring an older facility and retrofitting it with liquid cooling, AI-driven power management, and updated UPS systems can transform a dinosaur into a competitive asset while preserving the hard-to-replicate infrastructure underneath.
Repositioning: More Than a Renovation
Asset repositioning in the data center context is a discipline unto itself. It's not just swapping out CRAC units and calling it a day. Effective repositioning requires a thesis — a clear view of what type of customer the facility will serve, what power density per rack that customer requires, and what connectivity ecosystem needs to surround it.
The market has fragmented significantly. Hyperscalers like AWS, Microsoft Azure, and Google Cloud have their own development pipelines and tend to absorb capacity through large wholesale lease structures. But the mid-market — enterprise customers, regional cloud providers, AI startups, and edge compute operators — is hungry for colocation space that hyperscalers aren't building. Repositioned facilities targeting this segment don't need to win on scale; they need to win on location, latency, and flexibility.
Practically, this means:
- Power density planning: Many legacy facilities are wired for 3-5 kW per rack. Edge AI and GPU-heavy workloads demand 20-40 kW or more. Repositioning for high-density compute requires significant electrical infrastructure investment, but that investment creates a defensible market position.
- Connectivity stacking: A repositioned facility becomes exponentially more valuable when multiple carriers terminate there. Negotiating carrier-neutral status or building a meet-me room can transform a single-tenant asset into a destination.
- Cooling architecture: Air-cooled facilities face a hard ceiling on density. Direct liquid cooling (DLC) and rear-door heat exchangers are increasingly standard for high-performance compute — and retrofitting for these systems is entirely feasible in most legacy mechanical plants.
What Successful Revitalization Actually Looks Like
The revitalization of underperforming or decommissioned data center assets has become a legitimate investment thesis, not just an operational convenience. Private equity firms, infrastructure funds, and REITs have all moved into this space with purpose-built strategies.
The pattern that works looks something like this: acquire a facility in a constrained market — often from a corporate seller doing a sale-leaseback or exiting owned infrastructure — at a basis well below replacement cost. Invest 12 to 24 months in capital improvements targeting specific customer segments. Stabilize occupancy through long-term leases. Refinance or exit at infrastructure REIT multiples, which have historically valued stabilized data center assets at 20-30x EBITDA.
The lesson embedded in every successful revitalization is the same: the physical shell is table stakes. The value creation happens in the operational layer — the power contracts, the cooling efficiency, the carrier relationships, and ultimately the tenant mix.
Companies that have stumbled in this space typically made one of two mistakes: they underestimated the capital intensity of bringing a legacy facility to modern density standards, or they acquired assets in markets where power capacity was genuinely exhausted and customer demand was thinner than the thesis assumed. Due diligence on utility relationships — including interconnection queue position and demand response program eligibility — is non-negotiable.
Where This Is Headed
The investment case for data center acquisition isn't going to weaken as AI adoption deepens — it's going to intensify. Every enterprise that deploys a large language model, every hospital building a patient data analytics platform, every automaker training autonomy systems: they all need somewhere to run compute. And the gap between where capacity exists and where new capacity can be built in time is only widening.
A few dynamics worth watching closely:
Secondary and tertiary markets are becoming primary targets. As Tier 1 markets like Northern Virginia and Silicon Valley hit saturation, acquirers are moving into markets like Salt Lake City, Columbus, San Antonio, and Boise — places with affordable power, available land adjacent to existing facilities, and growing fiber density. Assets that would have been overlooked three years ago are now attracting competitive bids.
The energy piece is becoming the acquisition thesis. Facilities with on-site renewable generation, power purchase agreements already in place, or located near new transmission infrastructure are commanding premiums that would have seemed irrational five years ago. Microsoft's $10 billion investment in data center infrastructure and its aggressive pursuit of nuclear and renewable power agreements signals exactly where the puck is going.
Corporate sale-leasebacks will continue to create supply. Enterprises that built their own data centers in the 2000s are increasingly recognizing that running a facility is not a core competency. As they exit owned infrastructure and sign long-term leases with operators, they're feeding a steady pipeline of acquisition candidates — often in desirable locations, with existing fiber and power, and a known anchor tenant already in place.
For anyone deploying capital into digital infrastructure, the strategic question is no longer whether data center acquisition belongs in the portfolio. It's which assets, in which markets, repositioned for which workloads — and how fast you can move before the next buyer does.
Explore the InfraSale Marketplace for strategic data center acquisitions.