Why United Power Belongs on Colorado's Data Center Map

United Power deserves a closer look from anyone searching for Colorado data center land. Its public records show a sharp increase in developer interest, a formal process for reviewing large loads, and detailed planning for a much larger power system.

The clearest evidence is in the cooperative's 2025 annual report. United Power received five large-load inquiries in 2023, 24 in 2024, and 55 in 2025. The report also describes prospective requirements exceeding 3 GW and a historical system peak of approximately 680 MW in July 2025. The figures appear on printed page 9 of United Power's annual report.

That is useful information for an early land search. It gives us a local measure of interest and a basis for deciding where to invest more research time.

My interest is in the next layer of detail: how a specific property moves through United Power's process, which transmission arrangements it depends on, and when a proposed first phase could receive service.

Developer interest increased sharply

The inquiry figures produce a simple comparison:

Growth is calculated as the change in inquiry count divided by the prior year's count.

The 2025 total was eleven times the 2023 total. These counts measure inquiries received by the utility. Project completion requires additional evidence about site control, studies, agreements, approvals, and construction.

For origination, the increase justifies more attention to the territory. It also suggests a useful way to structure future research. Track inquiries, projects that enter studies, projects that sign agreements, and projects that begin operating as separate stages.

Over time, those stages could support a local conversion analysis. The important work would be matching the same project across records and preserving its history. Counting each mention as a new project would make the resulting market picture difficult to trust.

United Power reported 55 new large-load inquiries in 2025, up 129.2% from 24 in 2024. The count was eleven times the five inquiries reported in 2023. For Colorado site selection, this is evidence of rising developer interest in one utility territory. The next research step is to follow individual projects through studies, service agreements, and local approvals.

That last step is where a useful local dataset starts to become harder to reproduce. It requires reading across records that were created for different purposes.

The 600 MW number has a public planning context

A September 9 report on Xcel's large-load proceeding says United Power witness Jonathan Aust disclosed a pending 600 MW interconnection request and raised concerns about its processing. The report identifies the testimony as part of proceeding 26AL-0137E. Rocky Mountain Voice's account is the source for that pending-request claim.

Separately, United Power's 2026 Integrated Resource Plan models a 600 MW large-load addition. Its assumption is unusually specific: continuous operation, introduced in 50 MW blocks every eight months beginning in January 2027. Section 7.1.7 of the resource plan describes the scenario.

United Power's published 600 MW planning scenario translates into twelve 50 MW additions. Applying its eight-month spacing from January 2027 puts 300 MW in May 2030 and 600 MW in May 2034. The first and final additions are 88 months apart. These are calculated milestones from the utility's modeling assumption.

Those dates come from the published modeling assumption. The pending interconnection request and the resource-plan scenario have separate source records; their matching size alone leaves their relationship unresolved.

For a land developer, the schedule is still informative. It offers a concrete example of what phased growth can look like in a utility planning model. A campus concept can involve several years of construction, equipment installation, and electrical expansion after its first phase begins operating.

That affects how I would evaluate the property. The initial building area matters, along with the land reserved for later electrical facilities, access routes, and future construction. The first phase and the ultimate campus need to work together on the same plan.

A phased campus needs a phased land strategy

A property marketed for a large ultimate load should have a clear first phase. The initial service request needs a specific location, demand level, and target date. Later phases need enough definition to show how the site would grow.

Using the resource plan's sequence as an illustration, the first 50 MW block begins more than seven years before the final block. A site review built around that sequence would need to consider how construction can continue alongside an operating facility.

The layout questions are practical. Where would later equipment deliveries enter? Which corridors must stay open for additional utilities? What space is reserved for electrical equipment? How would a later building connect to shared infrastructure?

The land-control structure also deserves attention. If the first phase depends on a neighboring parcel for access or an electrical route, that dependency belongs in the initial diligence package. If a later phase depends on acquiring more land, the development schedule should identify that milestone.

A useful map can show these relationships directly. Give each phase its own boundary, requested load, target date, and supporting document. A reviewer should be able to select a phase and see the specific assumptions behind it.

That would make a 600 MW campus concept much easier to evaluate. It would also help the project team revise the plan when a service date, tenant requirement, or equipment layout changes.

United Power's application tells developers what to prepare

The cooperative's large-load inquiry asks for a location, initial and final MW, delivery dates, project status, site-control information, a site plan, and a load ramp. Requests over 50 MW receive review by both Engineering and Power Supply. United Power publishes the process and questionnaire.

Those fields are a useful specification for a land research package.

Start with a single project location that matches the proposed electrical service area. Identify the parcel boundaries and the development entity. Explain how the developer controls the property and what supporting land rights still need to be secured.

Then provide a load schedule that connects the physical phases to the requested service. A building scheduled for completion in one year should have a corresponding electrical requirement. If the load arrives gradually, show the increments and their timing.

The project status should also be concrete. A preliminary inquiry, an executed option, an approved development plan, and an active construction project represent different levels of commitment. Describe the evidence supporting the current stage.

This is a good place for discipline in the site database. Each field should have an owner and a source. The utility team should be able to trace the proposed load back to a project plan, and the land team should be able to trace the site-control claim back to the relevant document.

A package built that way makes the first utility conversation more productive. It gives the reviewers something specific to evaluate and gives the developer a clear record of what was submitted.

Follow the connection across utility boundaries

United Power's distribution system is interconnected with Xcel's transmission network. The Colorado Public Utilities Commission's June order recognizes United Power's concern about serving large customers that depend on that network. Paragraph 22 of the commission's order describes the relationship.

For a candidate parcel, I would begin with the retail service territory and then identify the transmission arrangements that require study. That produces a more complete picture of the project's power path.

The next questions concern responsibility and sequence. Which utility receives the application? Who studies the relevant facilities? Which agreement follows the study? What must happen before construction can begin on the required infrastructure?

A project record should name the responsible party at each step. Otherwise, a promising site can accumulate several encouraging conversations while the most important dependency remains unclear.

The same approach helps compare properties on different sides of a service boundary. Give both sites the same proposed load and delivery schedule. Then compare the documented review paths, required facilities, commercial terms, and unresolved dependencies.

A shorter mapped route may deserve a closer look. A documented sequence of studies and agreements deserves its own weight in the decision. Keeping both visible makes the comparison more useful to someone who has to commit time and money.

Read supply announcements alongside the planning assumptions

United Power announced the Fortress solar and storage agreement in October 2025. The announcement describes 200 MW of solar near Brush and an expected 2027 start. The cooperative's Fortress announcement provides the project context.

Its 2026 resource plan assumes a January 1, 2028, in-service date for Fortress. The large-load scenario with carbon-reduction requirements raises modeled resource cost from $51.56 to $55.95 per MWh. Customer bills require additional service-cost and rate inputs. Sections 7.1.1 and 7.2.7 document the assumptions and results.

For a site model, an announced completion date and an assumed modeling date should occupy separate fields. They serve different purposes. Preserving both lets a reviewer see the schedule used in the analysis and decide which assumption needs an update.

The same principle applies to the role of a supply project. Record its technology, planned location, expected service date, and the agreement described in the source. A parcel-level delivery claim requires the corresponding service documentation.

This is where reading the underlying reports pays off. A press release helps identify the project. A resource plan shows how a utility has represented that project within a broader set of planning assumptions. Together, they provide a better basis for follow-up questions.

Build a local record that improves with each update

The public material already supports three useful research products: an annual inquiry series, a reconstruction of the published large-load ramp, and a source-by-source record of supply assumptions.

The next layer should connect those records to individual land-use cases. Match developer names, project locations, parcel identifiers, service dates, and requested loads only when the documents support the connection. Preserve uncertain matches for review.

Each update should answer a specific question about progress. Did a project secure site control? Did it enter a study? Did a local government approve its first phase? Did a service date change? Did the scope increase?

That work creates a useful history of how projects move through this territory. It can also help identify the evidence that tends to appear before a project becomes a serious development candidate.

United Power's rising inquiry count gives me a reason to watch. Its application process and resource plan tell me what to watch next. For the land search, I would focus on properties that can support a defined first phase and a credible path through the utility's review process.