The batch-review process is ERCOT’s structural response to that volume; Abbott’s directive adds a political layer that did not exist six months ago
Decision Focus
On June 11, 2026, ERCOT’s board voted to shift data center connection reviews to a batch-study model requiring developers to pay $50,000 per megawatt proposed and prove land control before their project enters the queue. That same week, Governor Greg Abbott directed the Texas Public Utilities Commission and ERCOT to take immediate steps to protect residential ratepayers from data center expansion costs. Together, these moves redefine the operating environment for any energy head with Texas assets in their interconnection pipeline.
90-Second Brief
As the week closes, eRCOT received 519 large-load connection requests over the last two years, against 24 the year before, with 90% from data centers targeting operation by 2030. The cumulative proposed load reached 438,595 megawatts, roughly a third of total U.S. Generation capacity. The batch-review process is ERCOT’s structural response to that volume; Abbott’s directive adds a political layer that did not exist six months ago.
What Is Really Happening?
The pipeline is neither evenly distributed nor uniformly credible. At least 248 data centers are now planned across Texas — 86 in North Texas, 56 in Central Texas, 45 in West Texas — yet power estimates have been disclosed for fewer than half. The projects that have disclosed capacity are individually massive: one Abilene facility targets 1,200 megawatts, and Fermi America’s campus outside Amarillo could require 11,000 megawatts — more than the combined household load of Houston, Dallas, Austin, and San Antonio. These are not incremental additions. They represent step-change demand events that existing transmission infrastructure was not designed to absorb on this timeline.
Texas arrived here through the same structural features that made it attractive: a deregulated market, permissive land use outside city limits, available land and fiber, and a political environment that welcomed industrial investment. That combination produced a first-mover flood that has overwhelmed the review capacity of both utilities and ERCOT itself. The batch process acknowledges that the prior bilateral workflow cannot handle current volume without degrading reliability analysis.
The political shift is harder to model. Abbott’s directive explicitly names residential ratepayer protection, water conservation, and community impact as co-equal priorities alongside economic development — a meaningful reframing from a governor who had previously positioned Texas as the epicenter of AI infrastructure investment.
Why It Matters for Global Heads of Data Center Energy
The immediate operating implication is queue sequencing. Projects that have not yet submitted a formal interconnection request — or that assumed Texas would process requests under the prior bilateral model — are now subject to batch timing. ERCOT has indicated the first group will study approximately 100,000 megawatts of proposed load, with review expected by April. Projects outside that first batch face at least one additional full review cycle before commercial operation becomes possible.
The $50,000-per-megawatt deposit is both a financial filter and a signal about where ERCOT is placing the cost of speculative applications. A 500 MW project now requires a $25 million deposit before it qualifies for study — shifting capital allocation decisions earlier in the development cycle and raising the real cost of maintaining optionality across multiple Texas sites simultaneously.
Cost allocation is equally material. ERCOT’s review will determine what transmission infrastructure must be built to serve approved loads and who pays for it. With Abbott’s directive explicitly shielding residential ratepayers, political pressure favors greater cost assignment to large industrial customers. How that resolves through the PUC will directly affect total delivered power costs for Texas data center operations and should be factored into PPA and tariff modeling now, not at financial close.
Water exposure is a parallel constraint that is beginning to limit siting choices. University of Texas at Austin research estimates data centers could account for 3% to 9% of total Texas water use by 2040, up from less than 1% today. Community opposition in San Marcos — where a project was voted down over river and aquifer concerns — demonstrates that water permitting is no longer a formality in Texas markets. Projects in Central Texas and the Hill Country face the highest exposure; West Texas projects sit in even more water-constrained terrain.
Forward View
Three fronts warrant active tracking. First, the PUC’s rulemaking on infrastructure cost allocation will set the template for how transmission build-out is charged across Texas — an outcome that materially changes the economics of projects currently in queue. Second, Abbott’s directive will likely produce specific regulatory guidance before year-end; the open question is whether it formalizes water-use minimums, noise standards, or grid contribution requirements that currently sit outside Texas’ permissive regulatory framework. Third, roughly half of the 248 planned facilities sit outside city limits, in a legal and political environment in flux — county commissioners lack zoning authority, but legislative pressure to grant them additional tools is building ahead of the next regular session in January.
What Is Still Uncertain
The confirmed pipeline volume does not establish how many projects will reach financial close, begin construction, or achieve full capacity. ERCOT acknowledges that actual build-out is unknowable at current proposal density. The infrastructure cost split between data center operators and residential ratepayers remains unresolved — regulators are still evaluating this and no formula has been established. Abbott’s directive has not yet been translated into enforceable regulatory text, and the compliance implications of a policy statement differ materially from those of a formal rule. Whether the batch process will produce faster or slower study timelines than the prior bilateral model is also untested.
One Question for Your Team
Given the new $50,000-per-megawatt deposit requirement and ERCOT’s first-batch threshold, which of your Texas interconnection applications are positioned to qualify in the first review cycle — and what is the fallback site or timeline if they are not?
Sources
- Timesrecordnews — Data center boom brings new challenges. What’s planned near you? (Link)
