Skip to content
LOAD FACTOR
Back to Insights
Market Analysis

The Buildable Queue: Signed IAs, Study Stage and Developers That Finish

Published October 25, 2026 • 7 min read • Load Factor team
Share this

The Buildable Queue: Signed IAs, Study Stage and Developers That Finish

By Load Factor Team

The United States ended 2025 with about 2,061 gigawatts of generation and storage actively seeking grid interconnection, according to Lawrence Berkeley National Laboratory. That number is quoted everywhere, and it misleads nearly everyone who uses it. It counts a project that entered the queue last month the same as a project holding a signed interconnection agreement, and a developer that has never built anything the same as one that has delivered ten projects. Ranked by raw gigawatts, the queue is a phone book. Ranked by buildability, it is a pipeline. This piece builds that second ranking.

The wrong ranking

The headline queue shrank in 2025, down 10 percent from the prior year, because more than 750 GW withdrew while only about 600 GW of new requests arrived. Solar fell 19 percent to 773 GW, storage fell 16 percent to 749 GW, wind fell 19 percent to 220 GW, and natural gas rose 86 percent to 253 GW. The queue is shedding speculative entries faster than it adds them. Neither reading tells you which projects will actually get built. Only position in the process does that.

A ladder, not a list

Every active project sits on a short ladder: at the top, projects holding a draft or executed interconnection agreement (IA) but not yet operating; below them, projects deep in system impact or facilities studies, with upgrade costs largely known; below those, projects in early studies, where cost and timing are still guesses; at the bottom, recent entries that have survived no study at all. Each rung down, the chance of reaching operation falls. A ranking that ignores the rung is not a forecast. It is a count of intentions.

What history says about the ladder

The historical record is blunt. Of all capacity that requested interconnection from 2000 to 2020, only 13 percent had reached commercial operation by the end of 2025, while 75 percent had been withdrawn and 10 percent was still active. The median project built in 2025 spent more than five years between its interconnection request and commercial operation, up from under two years for projects built in 2000 to 2007. Time in queue is itself a signal, cutting both ways: survival shows backing, but also proximity to expiring options and financing.

Completion also varies by fuel. In Berkeley Lab's earlier cohort analysis, covering requests from 2000 to 2018, gas projects completed at 31 percent, wind at 20 percent, solar at 13 percent, and batteries at 11 percent. A solar project and a gas project at the same study stage are not equally likely to finish.

Late-stage failure deserves its own warning. Berkeley Lab has flagged that withdrawals after the IA stage, while less common than early withdrawals, may be rising, and that they are costlier: they strand sunk costs and deposits, and they can force restudies for the projects behind them. An executed IA is the strongest single signal in the queue, but it is not a guarantee.

The tier that matters most: signed agreements

The most buildable slice of the queue is also the easiest to measure. Some 549 GW of capacity already holds a draft or executed IA but has not yet reached operation, more than a quarter of everything actively queued. Its fuel mix is instructive: solar dominates it, storage is second, and gas, despite its surge in new requests, is a small part of the signed tier.

Capacity holding a draft or executed interconnection agreement, by fuel Not yet operating. Other fuels make up the remaining 11 GW of the 549 GW total.
GWCapacity with draft or executed IA (GW) (GW)
Solar256
Storage161
Wind76
Natural gas45

Source: Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition (data through end of 2025)

Note what the chart does not show. The 549 GW signed tier is more than six times the 86 GW developers told the U.S. Energy Information Administration they plan to add in 2026. An IA is not a construction schedule. Much of the signed tier will take years to build, some of it will slip, and some will withdraw. Use the IA tier as the top of the ranking, not as a forecast of next year's additions.

The EIA cross-check

EIA's inventory offers a check. Developers planned a record 86 GW of utility-scale additions for 2026: 43.4 GW of solar, about 24 GW of battery storage, 11.8 GW of wind, and 6.3 GW of gas. That followed 53 GW added in 2025, the largest single-year installation since 2002. Two lessons follow. First, the signed IA tier is a multi-year stock while the 86 GW is a one-year flow, and confusing the two overstates near-term supply. Second, plans are developers' own dates, not outcomes; the queue's history of withdrawals and five-plus-year timelines says a large share of any year's plan will arrive late or not at all. Read the buildable ranking as a probability ordering, not a delivery calendar.

How the ISO ranking changes

This is where a raw gigawatt ranking fails hardest. An ISO can top the national queue by total capacity while holding a small share of signed, late-stage projects, and another ISO can look small while holding a dense core of projects ready to build.

The developer signal

Study stage is the strongest project-level signal; the developer is the strongest portfolio-level one. A parent that has carried past queue entries through to operation has shown it can survive studies, fund upgrades, and close financing. A parent whose history is a trail of withdrawals has shown the opposite. Both patterns should move a project up or down the ranking regardless of fuel or ISO.

How to use this ranking

The checks, in order: first, does the project hold an executed IA? Second, what study stage has it reached, and are upgrade costs assigned? Third, how long has it been in queue, and does that age signal staying power or expiring options? Fourth, has its parent built before? Fifth, what is the historical completion baseline for its fuel? A project that clears the first four checks is in the buildable core even if its fuel baseline is weak. A project that clears none of them is a headline gigawatt, not a pipeline megawatt.

The queue will keep shrinking as speculative entries withdraw and FERC Order 2023 reforms take hold, though Berkeley Lab cautions it is too early to measure their full impact. What will not change is the discipline: count what is signed, weight what is studied, discount what is new, and follow the developer, not just the megawatts. The buildable queue is smaller than the advertised queue. It is also the only one worth planning around.

Sources

  • Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition, Characteristics of Power Plants Seeking Transmission Interconnection As of the End of 2025 (data through end of 2025; published May 2026): https://emp.lbl.gov/publications/queued-2026-edition-characteristics
  • Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition slide deck (PDF): https://eta-publications.lbl.gov/sites/default/files/2026-06/queued_up_2026_edition.pdf
  • Lawrence Berkeley National Laboratory, Queued Up data portal and interactive visualizations: https://emp.lbl.gov/queues
  • Lawrence Berkeley National Laboratory, Queued Up: 2024 Edition (fuel completion rates for the 2000-2018 request cohort; late-stage withdrawal analysis): https://emp.lbl.gov/publications/queued-2024-edition-characteristics
  • U.S. Energy Information Administration, Preliminary Monthly Electric Generator Inventory (Form EIA-860M), 2026 planned capacity additions as reported February 2026 (86 GW total; solar 43.4 GW; battery storage about 24 GW; wind 11.8 GW; natural gas 6.3 GW; 53 GW added in 2025): https://www.publicpower.org/periodical/article/new-us-electric-generating-capacity-expected-reach-record-high-2026

Load Factor tracks 38,000+ interconnection queue projects, withdrawn included, across 16 North American markets, with the evidence behind every developer it names.

Next step

See your market in 30 minutes.

A walkthrough of the projects, developers and evidence in the markets you work in.

Get the weekly brief

Queue movement, newly identified developers and the week’s deals, every Monday. Read past issues

Book a 30-min walkthrough