Bloom Energy Q3 Revenue Doubles to $1.24 Billion, Full-Year Guidance Raised for Third Time This Year
nashnova research
Bloom Energy's Q3 revenue doubled year-over-year to $1.24 billion, net income swung from a loss to $250 million, and the company raised full-year guidance for the third time this year — AI data centers' power anxiety is funneling real money into on-site generation.
How strong is this quarter, exactly?
Revenue hit $1.24 billion, up 138% year-over-year, well above the $1.07 billion analyst consensus; core product revenue crossed $1 billion in a single quarter for the first time.
Net income reached $250.3 million versus a $23.1 million loss a year ago; adjusted EPS of $0.94 beat the Street's $0.69 estimate by 36%.
This means → not just a "beat" but a step-change: operating profit of $286 million exceeded the entire prior year's total in a single quarter.
Why did margins jump so sharply?
Operating margin widened from 8.9% a year ago to 23.1%.
In plain terms = revenue grew 138% while operating expenses rose only 26% — income doubled, spending didn't, and the gap dropped straight to the bottom line.
This reflects operating leverage — the effect where profits accelerate when revenue grows much faster than costs — kicking in hard, a turning-point signal that fuel-cell economics are shifting from cash-burning expansion to profitable scale.
Why has guidance been raised three times in one year?
The company lifted full-year revenue guidance from $3.9–4.2 billion to $4.4–4.6 billion; adjusted EPS guidance moved from $2.55–2.85 to $3.10–3.40.
The new midpoint of $3.25 sits well above the prior analyst consensus of $2.71.
This means → management has unusually high visibility on orders for the next two quarters; otherwise they would not raise guidance a third time on top of two prior increases.
AI power shortage — why are tech giants generating their own electricity?
CEO KR Sridhar quoted Nvidia's Jensen Huang: for a gigawatt-scale data center, every quarter of delay means $10 billion in lost revenue.
In plain terms = the traditional grid was designed to spread power evenly across millions of dispersed users. It simply cannot handle a data center consuming an entire small city's worth of electricity at a single point — so hyperscalers are forced to build power plants next to their server halls.
Delivery speed is the key selling point: the Oracle project began delivering power to its Utah site within two months of announcement; the Nebius project in New Jersey won planning approval three months after signing and had first power ready two months after that.
Cash flow and costs — what's hiding inside the good news?
Operating cash flow was a net outflow of $27 million this quarter; free cash flow was a net outflow of $181 million — the company earned profits but hasn't collected the cash yet.
In plain terms = large orders ship before payment arrives. The product is out the door; the receivable is still on the books. The bigger the company gets, the more working capital it ties up.
Capital expenditure reached $231 million year-to-date — four times last year's full total — going toward capacity expansion. Management explicitly warned that new-line ramp-up costs will pressure gross margins in 2027.
The stock fell 7% after hours — what is the market worried about?
Shares dropped roughly 7% to around $200 in after-hours trading; the stock had already more than doubled year-to-date.
This means → the results themselves aren't the problem, but the stock had already priced in the good news. Add management's own warning about 2027 margin pressure, and holders chose to lock in gains.
The next key test: whether gross margins recover on schedule once the capacity ramp completes in 2027 — that will determine if the current valuation is justified or stretched.
市场有风险，内容仅供研究参考，不构成投资建议。