FuelCell Energy's Data Center Pivot Produces Real Orders, but Profits Remain a Year Away

The fuel cell maker signs its first big data center capacity deal, expands backlog to $3.6 billion, and targets EBITDA positive by late 2027.

FCEL · Earnings Call · 2026-09-02

A Strategic Shift

FuelCell Energy's third-quarter earnings call was a clear inflection point. After years of positioning around distributed generation and international repowering projects, the company now speaks almost entirely in the language of AI infrastructure and power constraints. Jason Few's opening remarks captured the urgency:

The AI economy will not be constrained by silicon. It will be constrained by access to electricity. We believe distributed generation will become an essential layer of AI infrastructure, enabling the grid to expand while allowing AI factories to deploy on commercial timelines rather than utility timelines.
Jason Few, President and Chief Executive Officer · 2026-09-02

That framing is not just marketing. The company now has a real commercial commitment—a Fit Energy agreement covering up to 380 megawatts across four phases, backed by an upfront deposit on the first 30-megawatt phase. More importantly, management introduced a new category of awarded capacity backlog, which together with committed backlog totals $3.6 billion as of July 31. The company's own keyword trajectory confirms this is a new theme: terms like 'Fit Energy,' 'capacity reservation,' and 'awarded capacity' dominate the latest quarter, rather than the usual 'carbon capture' or 'Korea market' language.

Financial Reality Check

The pivot is not yet profitable. Third-quarter total revenue fell 29% year-over-year to $33 million, and the company recorded a gross loss of $24.5 million, driven by $17 million of charges tied to the Fit Energy phase zero. CFO Michael Bishop explained that these charges reflect the current cost structure: “We operated at an annualized production rate of approximately 37 megawatts during the quarter, which remains below the volume at which we expect our cost structure to align with market-based pricing for orders of this scale.” — Michael Bishop, Chief Financial Officer · 2026-09-02 Still, the company is materially derisked on the balance sheet. “We ended the quarter with the strongest cash position in our history,” — Michael Bishop, Chief Financial Officer · 2026-09-02 with $737 million in total cash, enabling a fully funded expansion of Torrington to 500 megawatts by 2028. Management now targets positive adjusted EBITDA in fiscal Q4 2027, a date that has slipped from earlier hints but remains anchored to the same throughput milestone. In prior calls, executives repeatedly cited the 100-megawatt production level as the breakeven trigger. As Michael Bishop stated in the June call: “once we achieved consistent production volumes at or above 100 megawatts on an annualized basis, we are targeting getting to adjusted EBITDA positive.” — Michael Bishop, Chief Financial Officer · 2026-06-08

The company's effective net cash position rose to $324 million, a 69% year-over-year gain, giving management room to execute before the next turn in the capital cycle. However, the path to profitability is not solely a cost curve exercise; it depends on converting a capacity reservation and the broader pipeline into definitive orders. As Jason Few acknowledged in the Q&A: “we are not just relying on 1 or 2 customers, we see significant opportunities here across our customer base” — Michael Bishop, Chief Financial Officer · 2026-09-02. That sentiment is backed by a claimed 10 gigawatts of active proposals, with data centers now ~97% of that pipeline.

Validation and Risk

The market backdrop is undeniably favorable: the global keyword trajectory is filled with data center power themes, and other recent reporters like HPE, DELL, and AVGO are all emphasizing AI infrastructure. FuelCell Energy is clearly riding that wave, but it is also differentiating itself by touting its supply chain resilience and its Exxon carbon capture demonstration. The latter, however, is not the near-term growth driver; it serves more as a validation of the carbonate platform's flexibility.

Investors should be cautious about the awarded capacity backlog—it is explicitly not a firm order. As CFO Michael Bishop put it: “I want to be clear that awarded capacity backlog is not contracted firm order backlog or a guarantee of future revenue.” — Michael Bishop, Chief Financial Officer · 2026-09-02 The real test will come over the next few quarters as the company ramps production from 37 megawatts to its stated 100-megawatt run rate in October 2026 and, more importantly, as it converts those reservations into revenue-generating contracts with customers like the unnamed co-location operator behind the 75-megawatt deal announced after quarter end. The stock, up 22% over 90 days but still down 60% from its June peak, reflects both optimism and skepticism about delivery.