What 2.2 GW Means for Land Rights

2.2 GW for one campus

When AWS announced its expanded Indiana data center campus – 23 buildings, 6.9 million square feet – the headline number was 2.2 gigawatts of power demand. That single campus needs roughly half as much electricity as all Indiana households consume combined. It is not a small build. It is also not unique.

The campus is one node in a national infrastructure problem that is about to need a very specific kind of software. The shape of that software, who builds it, and how it gets priced are all up for grabs this decade. Most of the public conversation about data center growth focuses on the silicon and the substations – but the work that actually paces these projects, the work that determines whether a 2030 power-up date holds or slides, is something else entirely. It is the parcel-by-parcel land rights work that no one talks about until they are stuck in it.

The demand picture, by the numbers

US data center electricity consumption was about 4 percent of national generation in 2023. By 2030, depending on which forecast you trust:

• S&P Global has data center grid-power demand rising 22 percent in 2025 alone, then nearly tripling by the end of the decade.
• The IEA expects global data center energy consumption to roughly double by 2030, hitting 945 TWh.
• More aggressive US-specific forecasts (Grid Strategies, others) put 2030 US data center demand at 134 GW, with electricity consumption increasing by up to 240 TWh – roughly +130 percent versus 2024.
• Pew Research’s read of the recent academic and government estimates puts data centers at 9 percent of US electricity generation by 2030, more than double the 2023 share.

The driver is AI hyperscale specifically. Training and inference workloads have a different power profile than traditional cloud, with much higher density per rack and much harder-to-flex demand curves. This is the wave – everything else is residual.

The growth is not evenly distributed. A handful of states are absorbing most of it:

• Virginia: data centers already consume more than 1 in 4 kilowatt-hours of state electricity (32 TWh out of 128 TWh in 2023). By 2030, that share could rise to between 41 and 59 percent. Northern Virginia alone has roughly 4,900 MW of operating data center capacity with another 1,000 MW under construction.
• Ohio: AEP Ohio paused new data center interconnections entirely. A July 2025 tariff now requires developers to make financial commitments before grid connections are approved.
• Indiana (and Arizona, Iowa, Nebraska, Nevada, Oregon, Wyoming): projected to each exceed 20 percent of state electricity consumption from data centers by 2030.

The pause in Ohio and the wait times in Virginia (four to seven years for new high-capacity grid connections in some areas) are the leading edge of a national problem.

The grid response

Utilities and regional transmission organizations have responded with the largest committed transmission expansion in over a decade. Transmission investment had been relatively dormant since 2013; that is no longer the case.

The five major US RTOs have collectively announced more than $54 billion in transmission expansion:

This is real money, not a wish list. Specific projects are already in motion. The Southwest Intertie Project North (SWIP-North) – 285 miles, ~2,000 MW from Twin Falls, Idaho to Ely, Nevada – broke ground in 2025. The Southline Transmission Project – 175 miles, 748 MW from Hidalgo County, New Mexico to Pima County, Arizona – started construction in Q1 2025. There are dozens more in various stages of planning, permitting, and right-of-way acquisition.

But $54 billion in transmission spend does not buy you transmission. It buys you a plan to attempt to build transmission. The actual rate-limiting step – the work that determines whether the new line is in service in 2028 or 2032 – is something else.

The bottleneck most people miss

A transmission line is not finished when the engineering is done and the money is approved. It is finished when every parcel along the corridor has an executed easement, a recorded right-of-way, and a clean path through construction.

That is land rights work. It looks like this:

1. Title research. Who owns the parcel? Who else has an interest in it – mineral rights, mortgage holders, easement holders, lien claimants? This typically requires pulling title commitments, reading recorded documents, and sometimes resolving ambiguities with county records.
2. Owner notification. A formal notice that a public-need transmission project will require an easement crossing their property.
3. Negotiation. Easement compensation, route adjustments to avoid structures, timing concerns for agricultural operations, restoration commitments. This is where the project either moves or stalls. Easy parcels close in weeks. Hard parcels take a year or more.
4. Execution and recording. Signatures, payment, recorded easement document, updated title.
5. Construction-phase damage claims. Crops damaged by access roads, soil disturbance, drainage changes – these get filed and adjudicated parcel by parcel during construction.
6. Closeout. A complete project record that survives the line’s 50-year operational life so the utility can respond to inspection requests, refinancing audits, neighboring projects, and litigation.

A 138 kV regional transmission line runs roughly five parcels per mile through typical rural corridor, more in suburban areas and fewer in agricultural belts. The math is unforgiving: $54 billion in committed transmission build-out implies thousands of miles of new corridor. Thousands of miles implies tens of thousands of parcels needing every step of that workflow.

Womble Bond Dickinson, the energy practice that watches this space closely, put it bluntly in their 2025 analysis: “siting can be very complicated and lead to litigation and eminent domain claims that take years to resolve.” Northern Virginia’s four-to-seven-year grid connection wait times are not a transmission engineering problem. They are a land rights problem at scale.

Why existing ROW software was not built for this

The incumbent enterprise ROW platforms – Quorum, Pandell, Irth (which absorbed geoAMPS), Trimble’s right-of-way module, and a handful of smaller players – were shaped by twenty years of serving large pipeline operators and integrated utilities. They are good at what they do. But they were not designed for this moment.

A few specific structural issues:

They are built on ArcGIS. Most of them are Esri-embedded by design. That means the customer pays for ArcGIS too: named-user licenses, server licenses, the whole stack. For a major pipeline operator with a multi-thousand-mile portfolio, that is fine – they already have an Esri shop. For a survey firm being asked to deliver ROW services on a single regional transmission project, the Esri bill is a non-starter.

They are sold as enterprise platforms. Six-figure deal sizes. Months of implementation. Custom configuration. This makes sense when you are signing a five-year contract with a pipeline major. It does not make sense when a survey firm needs to stand up a system for a single 27-mile, 150-parcel project that ships data to the utility in 18 months.

They are optimized for thousand-mile portfolios. Their data models, workflows, and reporting assume you are managing many projects in parallel across a national footprint. For a single regional project, that complexity is friction, not value.

The mid-market is structurally underserved. Most of the new transmission build-out comes through projects that look like a single regional line: 20 to 50 miles, 100 to 300 parcels, $10 to $50 million capital, 18 to 36 month timeline. That is the project shape the incumbents are too expensive to serve profitably – and the project shape that has nothing else.

What this moment actually needs

The right ROW platform for the transmission build-out has a specific shape:

• Open-GIS stack. PostGIS, MapLibre, QGIS, GeoJSON. No Esri tax. The customer’s GIS bill is not a function of how many surveyors are on the project team.
• Survey-firm native. Built to integrate with the survey workflow because that is who actually delivers most ROW services on transmission. Boundary survey deliverables flow in cleanly; executed easements flow out cleanly; cross-reference to a firm’s historical survey archive happens by API.
• Days to stand up. Multi-tenant, schema-per-tenant, one-command provisioning. Sign in, get a tenant, start importing the parcel layer the same day. Not a months-long implementation.
• Priced for mid-market. A 27-mile, 150-parcel transmission project should not require a six-figure software deal to manage end-to-end. The economics need to match the project economics.
• Server or Managed. Same image runs both ways. Customer-hosted on-premises when data sovereignty is the constraint; StrataLogic-hosted when hosting convenience is the constraint. Both fully featured.

What we built

That is the gap we built LandLedger for. The platform is built, tested, and production-ready: 64 API endpoints, tenant-isolated multi-tenancy via schema-per-tenant, integrated with the rest of the StrataLogic stack so boundary survey deliverables flow in from PointScout and historical archive context flows in from FieldIntel’s Theo assistant. We are in early conversations with our first design-partner projects now.

The market is going to need ROW software at a different shape than what the incumbents are offering, and it is going to need it soon. We built the platform we would want as a survey firm being asked to deliver land services on a regional transmission corridor in 2027. We think there is a real market for that. Now we get to find out.