Grid interconnection, demystified

Ask most people what's slowing down clean energy and they'll guess the cost of panels, or batteries, or maybe permitting. The real answer is more boring and more consequential: getting permission to connect to the grid. As of the mid-2020s, more than two thousand gigawatts of generation and storage — several times the entire existing U.S. power fleet — were stuck in line waiting to interconnect. Over 95% of it is solar, storage, and wind.

This is the part of project development that never makes the brochure, yet it decides which projects get built and when. A developer can do everything else right and still be defeated by the queue. Understanding it isn't optional for a serious independent power producer — it's the difference between a project that energizes and one that quietly dies in a spreadsheet at the grid operator. So let's demystify it.

01The bottleneck nobody sees

The transmission grid is a shared, physical machine. Plug a new 200 MW plant into it and you change the flows everywhere nearby — voltages shift, lines that were comfortable may overload, protective equipment may need upgrading. Before any generator is allowed to connect, the grid operator has to study exactly what that plant will do to the system and what upgrades are required to keep everything stable. Multiply that by thousands of proposed projects, all interacting, and you have the interconnection queue.

By the time the backlog peaked, roughly 2.6 terawatts of proposed projects were waiting — and the composition tells the whole story of the energy transition. Solar and battery storage alone make up about 80% of the line.

02How interconnection actually works

A project joins the queue and then moves through a sequence of engineering studies, each more detailed than the last. The traditional path runs roughly:

• Feasibility study — a first look at whether the point of connection can physically host the project.
• System impact study — the deep analysis of how the project affects grid stability, and what network upgrades it triggers.
• Facilities study — the detailed engineering and price tag for the interconnection equipment and upgrades.

At the end, the developer receives an interconnection agreement specifying what they must build and what they must pay. The catch is in that last item. The cost of network upgrades is assigned to the projects that trigger them — and those costs can be enormous, sometimes large enough to kill an otherwise sound project. Worse, under the old rules, the costs assigned to your project depended on which other projects ahead of you stayed in or dropped out.

03Why the queue clogged

Two design flaws turned a manageable process into a multi-year traffic jam. The first was "first-come, first-served" studied one project at a time, in order. The second was that joining the queue was cheap and non-binding, so the line filled with speculative projects that had no real intention — or ability — to build.

These combined into a vicious cycle. Because each project was studied against all the others ahead of it, whenever a speculative project dropped out, the studies for everyone behind it had to be redone — a restudy cascade that reset the clock again and again. The queue became clogged with projects that would never be built, while real projects waited years for results that kept changing underneath them. By 2024, a record 112 GW of solar and storage withdrew from the queues in a single year as developers gave up or got cleaned out — painful, but a sign the system was finally shedding the speculative excess.

The queue wasn't slow because there were too many real projects. It was slow because the rules couldn't tell the real ones from the speculative ones — so everyone waited behind everyone.

04FERC Order 2023 and the cluster fix

In 2023 the Federal Energy Regulatory Commission overhauled the whole process. The core idea of Order 2023 is to stop studying projects one at a time and instead study them in clusters — groups analyzed together, with the cost of shared network upgrades allocated across the group. That alone breaks the restudy cascade, because a single dropout no longer forces everyone behind it to start over.

Just as important, the order replaced "first-come, first-served" with "first-ready, first-served." Projects now have to demonstrate genuine readiness — site control, larger financial deposits, real commercial intent — to hold their place, and they face penalties for withdrawing late. The effect is to push speculative projects out and let mature ones move. It's not an instant fix; grid operators will spend years working through the backlog. But the trajectory is right, and PJM, the largest U.S. grid operator, now expects to bring new study cycles down toward one-to-two-year timelines.

05The Texas difference

Not every grid runs on the same rules, and the contrast matters for where you choose to build. Texas's grid, ERCOT, is its own interconnection world — and it's famously faster. ERCOT uses a "connect and manage" philosophy: rather than requiring every network upgrade to be built and paid for before a generator can connect, it allows projects to interconnect sooner and manages any resulting congestion operationally. The result is one of the quickest paths from decision to operation anywhere in the country.

That single structural difference is a big part of why Texas has become the fastest-growing market for both solar and storage in the United States. When the queue elsewhere is measured in half-decades, a grid that connects projects in a fraction of that time is enormously attractive — and it's no accident that so much of the country's new battery capacity is landing in ERCOT.

06What gets a project through

Strip away the acronyms and the lesson for a developer is concrete. In a first-ready-first-served world, the projects that win are the ones that are genuinely ready: real site control, real financing, real intent — not a placeholder in a queue. Siting matters enormously, because connecting where the grid is already strong means smaller, cheaper network upgrades and a faster path through the studies. And pairing generation with storage can make a project a better grid citizen, easing some of the constraints that drive upgrade costs.

None of this is glamorous. But interconnection is where good projects prove they're real, and it rewards discipline over speculation. The developers who treat it as a core competence — not a box to check at the end — are the ones whose projects actually energize.