FuelCell Energy heads into its next report with a widening gap between what management has been promising and what the financial statements have actually shown. The story investors have been buying into since the last call is a surge in demand from data centers and AI infrastructure buyers, but not a single one of those deals has yet turned into signed backlog. This report is the next opportunity to close that gap between narrative and numbers.

Wall Street currently expects a loss of 35 cents per share on revenue of $39.1 million, which would mark a 63% improvement in per-share loss year over year even as revenue contracts 16.3% from the prior-year period. That combination tells you something important. The market is not pricing this as a growth story in the traditional sense. It is pricing continued cost discipline and narrowing losses against a backdrop of lumpy, module-exchange-dependent revenue that can swing sharply from quarter to quarter. Sequentially, the expected revenue figure would represent an improvement from the $35.6 million reported last quarter, though still well below the $46.7 million posted in the same quarter a year ago, underscoring how uneven the top line has been.

The last call was genuinely eventful. Management reported that its submitted-proposal pipeline more than doubled sequentially to 4 gigawatts, with data centers now representing 89% of that pipeline and average deal size doubling to 130 megawatts. That confidence was significant enough that management raised its planned Torrington capacity expansion target to 500 megawatts annually, a meaningful increase from the 350 megawatts previously discussed, at an estimated cost of $200 million to $275 million. Cash also climbed to $440.9 million, giving the company real runway to fund that expansion. But none of this pipeline growth has yet produced a signed data-center contract, and backlog has fallen for four consecutive quarters, from $1.24 billion to $1.14 billion, with product backlog now a thin $36.1 million. That divergence between a swelling sales pipeline and a shrinking contracted backlog is the central tension this report needs to address.

Investors should also weigh the $42.6 million Groton impairment and the widened operating loss it caused last quarter, along with continued share dilution through ATM issuance. Neither is catastrophic on its own, but together they show that the balance sheet strength management touts is partly a function of selling more stock, not just operating leverage. The production run rate, still around 40 to 41 megawatts against a long-standing 100-megawatt annual target, remains the more mundane but arguably more important number to track, since it is the bridge between today's pipeline excitement and tomorrow's actual revenue.

Sentiment context adds nuance here. Bullish sentiment has swung to 32.3% from an 18.1% bearish reading last quarter, and the stock has climbed 7.4% since the last report, outpacing the S&P 500 by 4.7 percentage points. The shares trade well above their 200-day moving average of $12.96, though at $18.75 they sit meaningfully below the post-earnings high of $37.88, suggesting the market has cooled from its initial enthusiasm without abandoning the story. That leaves room for the report to move the stock significantly in either direction.

Ultimately, this report needs to answer one question clearly. Has any portion of that 4-gigawatt pipeline started converting into contracted, revenue-generating backlog, or is the AI-driven demand story still confined to proposals and press-friendly statistics. Until a data-center deal actually lands on the books, the bullish case remains a forecast rather than a fact.pdf