Company Overview
Super Micro Computer is the company most investors know for the wrong reasons. The San Jose-based server manufacturer spent 2024 and parts of 2025 under a cloud of accounting concerns, DOJ investigation headlines, and auditor turnover — a governance storm that sent the stock down sharply and kept institutional investors at arm’s length. Underneath that noise, the actual business of building the dense, energy-efficient, liquid-cooled server systems that AI data centers need was quietly becoming one of the most consequential in the infrastructure stack.
Last night, Super Micro reported Q4 fiscal 2026 results that forced every analyst with a model on this company to start over. Full-year revenue reached a record $39.1 billion, up 78% from $22 billion in fiscal 2025, as demand for AI infrastructure drove the company to its strongest year ever. Q4 gross margin nearly doubled to 17.5% from 9.5% a year ago, turning a slight revenue miss into an EPS beat that sent the stock up 13–15% this morning. FY2027 revenue guidance of $65–$72 billion crushed the $52.5 billion analyst estimate, and the company entered the new fiscal year with a record backlog after receiving more than $60 billion in new orders during Q4 alone. The stock is up 13% today. It is still approximately 30% below where it traded a year ago. The gap between the operational story and the market’s memory of the governance story is the setup.
Key Technical and Fundamental Drivers
77.5% EPS Beat + Gross Margin Doubling → The Most Underappreciated Detail in the Print
Non-GAAP EPS of $1.70 beat the $0.96 analyst consensus by 77.5%, while adjusted EBITDA of $1.67 billion surpassed the $760.6 million estimate — a beat of more than 100%. Gross margin nearly doubled to 17.5% from 9.5% a year ago, driven by what management called a favorable customer and product mix. The margin expansion is the detail that changes the long-term investment thesis: a company that was operating at commodity-like margins of 9–11% and is now sustaining 17%+ margins on a $39 billion revenue base is a fundamentally different business than the one most institutional models are pricing. CEO Charles Liang described the company as expanding beyond its historical role as a U.S.-based server manufacturer into a provider of total data-center building-block solutions combining compute, storage, direct liquid cooling, networking, management software, and lifecycle services.
$60 Billion in Q4 Orders Alone → Record Backlog Entering FY2027
The company received more than $60 billion in new orders during Q4 and entered fiscal 2027 with a record backlog, with CEO Liang saying demand for AI IT solutions is “even stronger than ever before.” NVIDIA gained 2.5% alongside Super Micro as the record $60 billion order book signals sustained conviction in GPU-powered AI server demand from more than 20 customers. A company receiving $60 billion in a single quarter’s new orders — more than 150% of its prior full-year revenue — is not running a commodity server business. It is managing a contracted backlog that gives it multi-year revenue visibility at a scale most industrial companies would envy.
FY2027 Guidance of $65–$72 Billion → 37% Above Analyst Estimates
Super Micro’s FY2027 revenue guidance of between $65 billion and $72 billion crushed the $52.5 billion analyst estimate, sending SMCI up 13% and Dell up 5%. At the midpoint of $68.5 billion, the guidance implies 75% growth from fiscal 2026’s already-record $39.1 billion — a growth rate that is extraordinary for a company at this revenue scale. The Q3 CY2026 guidance of $15 billion at midpoint also exceeded analyst expectations by 26.7%, signaling that the current quarter’s demand environment is accelerating rather than plateauing.
Direct Liquid Cooling Leadership → The Technical Moat Most Investors Miss
Super Micro’s competitive differentiation is not simply that it assembles servers. It is the company’s leadership in direct liquid cooling — the thermal management technology that allows AI training clusters running at 100+ kilowatts per rack to operate without thermal throttling. As Nvidia’s Blackwell and next-generation GPU platforms increase power density further, air cooling becomes increasingly inadequate and liquid cooling becomes a prerequisite rather than an option. Super Micro has been building and deploying direct liquid cooling systems for years, giving it a technical head start over competitors who are retrofitting designs that were originally built for air-cooled environments.
Governance Overhang vs. Operational Reality → The Market Is Still Pricing the Old Story
The company is not a named defendant in the DOJ investigation and says it is cooperating, while an ongoing equity raise of approximately $7 billion is being used to fund components for the roughly $39 billion AI server order backlog from more than 20 customers. Barclays, Citi, Goldman, and Bank of America all raised their SMCI price targets but kept cautious ratings, doubting whether the 17.5% gross margin can hold — the central institutional concern that keeps the stock below its prior highs despite the operational transformation. The risk is real: governance uncertainty and margin sustainability are both legitimate questions. But a company guiding to $68.5 billion in FY2027 revenue with $60 billion in Q4 orders is no longer the governance story the market priced a year ago. The question is which story catches up to the other first.
Market Takeaway
Super Micro’s Q4 print last night is the kind of result that resets a narrative. A company that spent the better part of two years defined by its governance problems just reported 78% annual revenue growth, gross margins that nearly doubled, a 77.5% EPS beat, $60 billion in a single quarter’s new orders, and FY2027 guidance that is 37% above what every analyst on the Street was modeling. The governance story hasn’t gone away — the DOJ investigation is ongoing, the equity raise is dilutive, and the cash burn of approximately $6.8 billion in negative operating cash flow for the year reflects the scale of component purchasing required to fulfill that $60 billion order book. Those risks are real.
But the operational story has clearly outpaced the governance concern. The company’s expansion into total data-center building-block solutions — combining compute, storage, liquid cooling, networking, software, and lifecycle services — is the strategic repositioning that CEO Liang described as moving beyond historical server manufacturing. Goldman Sachs projects AI demand will exceed compute center capacity for years to come, validating the demand environment that produced $60 billion in Q4 orders. For a stock still trading approximately 30% below its one-year-ago price despite an operational transformation of this magnitude, the question for investors is not whether the business is improving — the results answer that definitively. The question is whether the governance and cash flow concerns that drove the prior decline are resolved quickly enough for the market to begin pricing the FY2027 guidance range rather than the 2024 investigation headlines. Thursday’s session, coming alongside Applied Materials’ Q3 results after the close tonight, will provide the next data point on whether the AI infrastructure confirmation from CoreWeave, Super Micro, and the hyperscalers is sustaining institutional attention or running into profit-taking after one of the most consequential earnings weeks of the year.