Geothermal energy provider Fervo reported $11.9 billion in contracted revenue, nearly double the $7.2 billion from the prior quarter, according to its third-quarter earnings report released Monday. The surge reflects hyperscalers' hunger for reliable baseload power to fuel data centers—and the company's ability to lock in long-term deals.
A substantial portion stems from Fervo's work with Google on the Cape Google 400 project, a 396-megawatt power plant scheduled for grid connection by end of 2028. Fervo connected its initial 100-megawatt facility to the grid earlier this year, also at Cape Station in Utah. The company previously estimated the site's potential at 4 gigawatts; it now believes Cape Station could produce up to 8.4 gigawatts.
The contract structure reveals the business logic. Tim Latimer, Fervo's co-founder and CEO, explained on the earnings call: "We built Cape Station on our schedule. Google can develop a potential data center on their schedule, and neither party is waiting on the other." Google can draw power as needed or sell excess capacity. This phased approach gives geothermal an edge over nuclear or large natural gas plants, which require full commitment upfront.
Fervo's newer wells run deeper and hotter, each generating 15 megawatts. Multiple wells link into a GeoBlock—the company's unit of modularity. The first completed GeoBlock produced 33 megawatts; recent ones hit 50 megawatts, sized to a typical data center hall's power draw. Hyperscalers can add capacity incrementally as each GeoBlock finishes.
But the economics hinge on cost. The initial Cape Station phase cost $7,000 per kilowatt of capacity. The Google project shows progress: $5,500 per kilowatt. Fervo targets $3,000 per kilowatt long-term. At $5,500 per kilowatt with subsidies, Fervo's power prices roughly match a new natural gas plant with fuel at $6 per million BTUs—roughly double current natural gas prices.
Fervo reported a $43.7 million operating loss for the quarter. The company holds $1.9 billion in cash and $301.6 million in long-term debt, providing runway to scale. The test: whether cost curves bend fast enough to make the unit economics defensible without permanent subsidy support.
