➤ Key Highlights
PUCT approved Batch Zero on June 24 — a one-time, systemwide review of new large loads of 75 MW or more.
It replaces the legacy utility-by-utility interconnection process with centralized ERCOT evaluation.
ERCOT’s large-load queue hit 226 GW (November 2025), up from 63 GW a year earlier.
About 73% of the queue is data centers; the 2026 summer peak forecast is only ~92 GW.
Roughly 23 GW of generation was added in 2024–25, with ~9 GW more in early 2026 — far short of requests.
➤ Signal
The bottleneck has moved from capital to demand to interconnection — and the queue is now roughly 2.5x the entire grid’s peak.
Texas didn’t add power on June 24. It added a filter. When interconnection requests run 2.5x the size of the grid itself, the queue stops being a forecast and becomes noise — duplicate filings, speculative land plays, and option-value placeholders all sitting in the same line. Batch Zero is the state’s admission that the old first-come process can’t tell a real 1 GW campus from a paper one. By forcing every load of 75 MW or more through one centralized gate, ERCOT is trying to price the scarcest asset in American development right now: a firm place in line. For developers, the signal is blunt — a power letter is no longer a power commitment.
Speculative data-center land underwritten on “queue position” just lost value. Bankable projects are the ones with firm interconnection, secured generation, or co-location with existing power — everything else is now a regulatory coin-flip. Expect a widening basis between sites with real power and sites with only a request number.
➤ Takeaway
In Texas, the grid — not the capital stack — now decides which data centers get built.
Source: PUCT / ERCOT — June 24, 2026
