What Data Center Operators Miss When Evaluating Sites in the Carolinas

By Brent Royall, SIOR, CCIM | 6 Summit Partners | April 2026

Every data center developer knows the checklist: power, fiber, water, flat land, favorable zoning. These are table stakes. If a site doesn’t have these fundamentals, it doesn’t make the first cut — and it shouldn’t.

But having spent the better part of the last several years advising a major hyperscaler on data center site selection and land assemblage across the Carolinas, I can tell you that the sites that fail rarely fail on the obvious criteria. They fail on the things that don’t show up in a GIS layer or a utility availability letter — speed, relationships, strategy, and knowing when to move.

North Carolina alone now hosts over 40 operating data centers with dozens more announced or under construction. Duke Energy’s data center contracts have grown from 3 GW to 4.5 GW in just the last year, and forecasts suggest power demand from data centers in the state could double to nearly 6 GW within the decade. AWS, Microsoft, Meta, Google — they’re all here, and they’re building at a pace that would have been unimaginable five years ago.

That kind of velocity creates opportunity. It also creates blind spots. Here are the ones I see most often — learned the hard way, in real deals, with real capital on the line.

1. Your First Call Should Be to the Power Company — Not a Land Broker

Most operators start their site search by looking at available land. That’s backwards.

Your number one conversation before entering a county or region should be with the business development team at the utility. In the Carolinas, that’s Duke Energy. They will guide you toward opportunity areas on their grid where capacity exists or can be delivered on a realistic timeline, and — just as importantly — they’ll steer you away from areas where securing the necessary power is a 7-to-10-year proposition regardless of how perfect the land looks.

I’ve watched operators fall in love with a site, tie up land, and then discover that the power timeline makes the project nonviable. That’s an expensive lesson. Start with power, then find the land that sits within the zones where power can actually be delivered.

And here’s a nuance that matters enormously in the current environment: in the era of constrained power, utilities are increasingly reluctant to lock in capacity commitments for merchant developers — because they don’t want to reserve megawatts for an end-user that may never materialize. If you’re an actual end-user, especially a publicly traded company with a track record of executing on commitments, you will often get pushed to the top of the queue ahead of developers who are building on spec. That’s a massive structural advantage, and it should inform how you enter the market and who you put in front of the utility.

2. Transaction Speed Kills — In Both Directions

Good sites with power will be found by the market. Period. If you’ve identified a site that checks the boxes, so has someone else — or they will within weeks.

I learned this firsthand working a major hyperscale search in one North Carolina county. We identified strong sites, but the end-user’s internal approval process couldn’t keep pace with how fast the market was moving. Sites kept getting scooped out from under us — not always by competing hyperscalers, but by merchant developers and industrial end-users who could move faster. In other markets we did run up against rival hyperscalers, but the lesson was the same regardless of who beat you to the punch: it was maddening, not because the sites were irreplaceable, but because each lost site meant restarting diligence, re-engaging the utility, and losing months.

The operators who win in this market have pre-authorized decision-making frameworks that let their site selection teams execute within defined parameters without going back to committee every time. If your approval process requires six weeks of internal review before you can make an offer on land, you’re going to lose to the developer who can move in six days.

This is also where having the right broker matters. An experienced buyer’s representative with a real pulse on local market activity can tell you when a site is about to hit the market, when a competing offer is forming, and when you need to move immediately to avoid a bidding war. You hired that broker for a reason — you should listen when they tell you it’s time to move.

3. Stop Benchmarking Land Comps Locally — Think Nationally

Here’s a frustration I’ve encountered repeatedly: corporate real estate teams that insist on a detailed local comparable sales analysis before they’ll approve a land acquisition. In a normal commercial real estate transaction, that’s sound practice. In data center site selection, it can be a counterproductive exercise.

The problem is simple: in many Carolina submarkets, there aren’t enough transactions involving land with the specific confluence of power availability, utility infrastructure, zoning potential, and acreage that a data center requires. You end up building a comps spreadsheet full of transactions that aren’t really comparable — farm sales, residential subdivisions, small industrial parcels — and trying to draw conclusions that don’t hold up.

The better framework is to evaluate your land basis relative to what you’re paying elsewhere in the country — or the world — to accomplish the same goal: build a data center campus. When you’re deploying billions in infrastructure, the difference between $25,000 and $40,000 per acre on a 200-acre site is a rounding error relative to your total project cost. The land basis is the least of your cost worries. What matters is power deliverability, speed to market, and entitlement certainty.

I’ve seen deals stall for weeks over land pricing negotiations that, in the grand scheme of a multi-billion-dollar campus, were immaterial. Often it makes sense to pay at or very close to what a seller is asking — particularly when the seller doesn’t fully understand the market value of their land for data center use and is pricing it based on traditional agricultural or industrial comps. Trying to negotiate them down another 10% might feel like good discipline, but if it costs you two months and a competitor swoops in, you’ve lost far more than you saved.

4. Confidentiality Is a Strategy, Not Just a Preference

Every hyperscaler wants to operate confidentially during site selection. But most don’t think about confidentiality as a strategic lever — they treat it as a blanket policy.

The reality is more nuanced. Early in a search, keeping your company name out of the conversation is critical. The moment a seller — especially one represented by a broker — learns that a major tech company is the buyer, their pricing expectations shift dramatically. What was a $3 million land deal suddenly becomes a $10 million land deal, and you’ve lost all negotiating leverage.

But confidentiality has a shelf life, and knowing when to reveal your identity is just as important as knowing when to conceal it. In a competitive situation where a seller has multiple offers, your brand name and balance sheet become your strongest card. A publicly traded hyperscaler that can demonstrate financial certainty and a track record of closing gives a seller confidence that the deal will actually happen — and that confidence is worth real dollars. You should know when to play that card and when to hold it.

The practical implication: your broker needs to be sophisticated enough to manage this dynamic. They need to control information flow, know when a seller is getting nervous about the anonymous buyer, and advise you on the right moment to step out from behind the curtain. That’s not a commoditized brokerage service. It requires judgment.

5. Find a Champion — Or Find a Different Market

Zoning and entitlements are where data center deals go to die. Community opposition is real and growing across the Carolinas — we’ve seen organized resistance in Stokes County, Rowan County, and Vance County. A developer recently withdrew a 250 MW campus application in Apex entirely after local pushback. These aren’t isolated incidents. They reflect a pattern.

The single most important thing an operator can do to de-risk the entitlement process is to find a local champion before they need one. An influential economic development director, a well-connected board chair on a city council, a county manager who understands what a data center investment means for the local tax base — these are the people who will help you navigate the political landscape, anticipate opposition, and develop a strategy to address NIMBY concerns before they metastasize into organized resistance.

The most successful site selection process I’ve been part of involved getting key local market players under NDA early — engineers, attorneys, economic development officials, local government contacts — so we could be mostly transparent about our goals and get genuine guidance on where the opportunities and landmines were. That level of openness, within a controlled circle, generated better outcomes than trying to operate in complete secrecy and stumbling into political tripwires we didn’t know existed.

When you find a market with a champion who is genuinely bought in, that’s the market you focus on hard. They’ll help accelerate your entitlement process, connect you with the right landowners, and help you build the community support that turns a contentious rezoning into a smooth approval.

6. Power Availability Is Not Power Deliverability

This one trips up even experienced operators. Duke Energy’s stated capacity in a service territory is not the same as what can be delivered to a specific site on a specific timeline. A utility availability letter might confirm that 100+ MW exists in the region, but the interconnection queue tells a different story. Transmission upgrades, substation construction, and easement acquisition can add 18 to 36 months to a timeline that a developer assumed was 12.

The developers who move fastest in the Carolinas are the ones who engage Duke Energy’s large load interconnection team before they tie up land — not after. They also evaluate whether a site can support phased energization, bringing a portion of the campus online while the full power build-out continues. That sequencing decision is a real estate negotiation as much as an engineering one, because it affects how you structure the land purchase, the construction phasing, and the lease-up timeline.

Don’t just ask “is there power?” Ask “when can I get power, how much will the infrastructure cost to deliver it, and who pays for the upgrades?” Those three questions will eliminate half the sites on your list — and that’s a good thing.

7. The Contrarian Play: Look Where Others Aren’t

North Carolina’s emerging data center corridor is attracting investment precisely because the infrastructure is concentrating there. But concentration creates competition — for land, for power, for permits, and for community goodwill.

The contrarian play — and one I’d encourage operators to seriously evaluate — is looking at sites slightly outside the established corridor. Lancaster County and York County in South Carolina, for example, offer proximity to Charlotte’s fiber and workforce infrastructure without the congestion and political fatigue that’s building in the more saturated North Carolina markets. Duke Energy serves both states. The land is cheaper. The communities are earlier in the cycle and more receptive to economic development conversations — if you approach them correctly.

The developers who secured sites in Catawba County and Lenoir five years ago locked in favorable terms. Today’s entrants are paying significantly more and waiting longer. Sometimes the best site isn’t the one everyone else is fighting over.

The Bottom Line

The Carolinas are one of the most active data center markets in the country right now, and for good reason. The fundamentals are strong: Duke Energy’s capacity, the fiber infrastructure, the business climate, and the geographic position between the Northeast and Southeast all create genuine competitive advantages.

But the easy sites are gone. The ones that remain require more sophistication in evaluation, more creativity in deal structure, more speed in execution, and more local knowledge than a spreadsheet can provide. The developers who win from here will be the ones who see around the corners that the checklist doesn’t cover.

Having the right advisor — one who understands the real estate, the infrastructure, the political landscape, and the pace at which this market moves — isn’t a nice-to-have. It’s the difference between landing a site and watching someone else build on it.

Brent Royall, SIOR, CCIM is a co-founder of 6 Summit Partners, a veteran-owned commercial real estate advisory firm based in the Charlotte, NC metro area. A West Point graduate and former U.S. Army officer, Brent has spent nearly 30 years advising clients on site selection, tenant representation, and strategic real estate consulting — including direct hyperscale data center site selection in the Carolinas. 6 Summit Partners specializes in helping organizations navigate complex real estate decisions with an operator’s perspective.