Fluence revised its 2026 fiscal year guidance down for a second time last week, citing ongoing supply chain issues that the company expects will pull its annual revenue down by around $1 billion from original estimates — and that analysts worry could hamper its burgeoning work with data centers.
The majority of that revenue decrease came down to delays at the energy storage provider’s new manufacturing facility in Houston, which Fluence announced last summer. Part of a manufacturing partnership with Bergstrom designed to produce 15 gigawatt-hours of capacity per year once fully ramped up, the factory pushes out HVAC and chiller equipment for Fluence’s Gridstack Pro battery containers.
The project is part of Fluence’s broader efforts to onshore production of “every major product and component” of a grid-scale energy storage system, the company said. But the company “underestimated the complexity of the ramp-up of the Houston facility,” CEO Julian Nebreda told investors last week.
The primary issue, Nebreda explained, is that a “customized automated welding process” continues to operate significantly below its targeted level. Additionally, the assembly process at the plant has been below targets. For now, the company has switched to manual welding, which is significantly slower, requires more quality inspections, and means hiring additional skilled labor. Fluence had been planning on producing an average of 11 units per day during August and September, as the factory ramped up. Actual production, however, averaged less than a single unit per day in August; in the early weeks of September, it’s still averaging just three.
In addition to the massive cost of the production delays, Fluence is also facing approximately $56 million in late delivery penalties.
On top of its problems with the manufacturing process itself, Fluence has run into the same issue facing large-scale projects around the country: grid connection delays that forced the company to turn to onsite generators to facilitate the ramp-up. According to Nebreda, that has meant a significant slowdown in getting all of the systems online.
As of last week, the facility is still not connected to the grid, though Fluence has brought in additional generator capacity that “essentially addresses the issue,” Nebreda said. The company expects a grid connection sometime between October and November, he added.
A flurry of changes
The first warning signs that all was not going entirely according to plan came in the company’s third quarter earnings call in early August. At that time, Nebreda predicted a delay of “a few months” thanks to construction delays and issues relating to the automation equipment; at that point, he expected the facility to be grid connected in a matter of weeks. The official Q3 estimates revised annual revenue down from $3.4 billion to $3 billion.
But challenges in Houston persisted. Fluence terminated its chief product officer on September 11, “effective immediately,” days ahead of the release of its revised guidance. According to an 8-K filing, the company hired a new chief operating officer on September 15: Bernerd Da Santos, former head of clean energy at AES. (AES and Siemens jointly formed Fluence in 2018.)
On September 16, the company again revised its estimated revenue, down to $2.4 billion, and told investors it is now anticipating $200 million in losses.
The news has analysts worried about potential impacts to Fluence’s customer relationships, particularly given the high levels of competition among battery suppliers, and the timeline demands of the AI boom.
As Jeffries analysts put it in a note to investors, they see runway for resolving the manufacturing issues, “and a market backdrop with robust demand should enable ongoing recovery.” However, given that both orders and new backlog additions could still be delayed, customer confidence in Fluence’s ability to meet servicing and delivery timelines could become an“acute” issue.
The question now, the analysts added, is how quickly Fluence’s competitors make their products available, potentially dampening the company’s leading position.
Data center impacts
Data centers — which are increasingly turning to batteries as part of their onsite power mix — are a small but growing part of Fluence’s customer base; the company had a 16 GWh pipeline in Q3. That’s a 35% jump from Q2, spurred by a single $300-million order for a behind-the-meter data center project, and another $550 million awarded via tender with a hyperscaler for batteries across several campuses.
Those deals add to a record-high backlog of orders, totaling $6.4 billion, with $2.2 billion scheduled for delivery in 2027.
The delays “could have an effect” on Fluence’s data center ambitions, Nebreda said, to the extent that the company has to talk to customers about delaying 2026 deliveries into 2027 — or even has to decline certain projects.
In the meantime, in order to reduce the impact the factory delay may have on data center projects themselves, Nebreda said Fluence will conduct commissioning in parallel — overlapping the construction, controls checks, energization, and performance testing processes to recover time.
But ultimately, Nebreda acknowledged, the delay “creates problems” for data center customers, which move much faster than the rest of Fluence’s pipeline: going from lead to contract in less than three months, compared to the roughly 18 months it takes for other customers. And that turnaround means that Fluence is in a tight spot.
As Jeffries put it: “We anticipate a deterioration of utility and data center confidence to challenge the future business prospects. The pitch of batteries today is speedy, relatively cheap capacity: If investors do not have confidence, they will pursue alternatives.”
