The geothermal startup XGS Energy has now completed four months of tests to see whether its technology can maintain steady production of heat at temperatures above what’s needed to generate energy. Over 3,000 hours, the company monitored the drilling process and checked how heat flowed from its wells, the status of their temperature, and how precisely XGS’ mathematical predictions matched the outcome of the testing.
The results, which the company shared exclusively with Heatmap, were “almost too good,” XGS CEO Josh Prueher told me.
“Had we been within 10% of predictive performance, we would have been pretty happy with the outcome,” Prueher said. “Turns out we were within 2% under a variety of different parameters.”
“It worked like a charm,” he said.
To understand what makes XGS Energy stand out among the geothermal startups racing to commercialize next-generation technology, it helps to compare the company to its fellow Houston-based rival that’s currently leading the sector, Fervo Energy. Unlike Fervo, XGS doesn’t use fracking technology to drill horizontal wells in pursuit of hot, dry rocks from which to harvest energy.
Instead, XGS drills vertical wells and inserts a closed steel pipe with water and fills the gap between the metal and the rock with a patented slurry that conducts heat. Technology like Fervo’s requires pumping cold water over the fractured hot rocks to harvest heat. But with its method, XGS claims, it avoids losing any water.
The testing took place off the US-395 highway in a volcanic field in California’s Mojave desert, sandwiched between the eastern edge of the Sequoia National Park and western border of Death Valley National Park. The geothermal field XGS tapped is already actively producing energy for the Coso Operating Company, which runs a 270-megawatt geothermal power plant on the land. The results, the company said, showed the “unprecedented predictability and active control of field performance” of XGS’ technology “versus other geothermal systems, which are subject to complex and continuously changing subsurface reservoir conditions.”
At least one outside observer agreed. “This is impressive, and something to be proud of,” Advait Arun, an energy analyst and senior associate at the think tank Center for Energy Enterprise who co-authored a recent report on next-generation geothermal, told me.
While the 3,000 hours of testing still falls short of the year’s worth of data Fervo has produced at one of its sites, it’s the longest any other competitor in the space has successfully demonstrated its approach so far, Arun said.
“These guys would be second to Fervo in terms of their ability to prove a commercial-scale performance test,” he added.
XGS is now poised to build a 150-megawatt power plant for Meta’s New Mexico data centers. Even after that’s complete, however, Prueher said the surrounding area has nearly 3 gigawatts of untapped heat. In California, where the company is headquartered and carried out its demonstration project, there’s a growing need for clean power sources that don’t further tax the depleted water table.
“A lot of the historical sensitivities around developing in California — a state where, like many others, water usage for industrial development is kind of a no-no — because we don’t need water, we have some real advantages,” Prueher said.
At a moment when surging demand from data centers is supercharging dealmaking in the electricity sector, Prueher said XGS is looking beyond the boom from the artificial intelligence buildout.
“It’s not about data centers,” he said. “It really is just the fundamental power needs of California. With the restrictions around water usage, we line up really, really well for California.”
For now, the company remains focused on the U.S. But Prueher said XGS is well suited to export its technology to East Asia, as well, where countries along the Pacific Rim have vast geothermal potential and growing electricity demand but limited development. XGS already has ties to the Philippines and “may actually be subsurface” — i.e. digging wells — there by the end of 2026, Prueher told me.
The “big enchilada,” he said, would be establishing a foothold in Japan, where the onsen hotspring industry has long protested geothermal development they say could diminish the resource that makes the ancient bathhouse tradition possible. Prueher told me his technology mitigates concerns over fracking-induced earthquakes, as well.
For now, he said, his main market is in the fast-growing Southwest. The executive compared this moment to 2021, when he worked at a battery company. That February, Winter Storm Uri collapsed the Texas grid as natural gas pipes froze and demand for electricity to heat homes designed to stay cool in a typically arid climate skyrocketed. Back then, he said, batteries were “still a pretty new asset class.”
“People were still uncertain about how it would perform,” Prueher said. But his company was “able to keep our batteries up and operating 100% of the time, no one minute of downtime during that entire episode.”
“From a market perspective, the storm showed that, if you can bring this new type of technology into the market, it can really deliver remarkable value,” he added. “We made 10 years of revenue in six days.”
In a lot of ways, he went on, “this is the same thing.”
“We’ve proven a technology is reliable,” Prueher said. “It works at commercial scale over a period of time. We would regard this as a real pivot point in the industry.”
