Five years ago, the hard drive was considered a technology on its way out. This quarter tells an entirely different story.
Seagate (NASDAQ: STX) reported a gross margin that jumped from 37.4% to 52.3% - nearly fifteen percentage points in a single quarter - and earnings per share that more than doubled.
How Storage Became an Interesting Business Again
For years the thesis was simple: flash-based SSDs are faster, their prices are falling, and the mechanical hard drive will disappear. The industry consolidated to three main manufacturers, and the stocks traded like cyclical commodities.
And then AI arrived, and changed the question.
Artificial-intelligence models do not only consume compute - they consume data, and they also generate it. Every training run, every version, every output. And when you are dealing with petabytes of information that must be retained over time but does not have to be read in milliseconds - the price per terabyte is what matters, not speed.
The distinction that transformed the industry
Not all storage serves the same purpose. "Hot" storage - data read constantly - requires speed, and there SSDs clearly win. Mass-capacity storage - archives, training corpora, backups, historical copies - requires cheap capacity, not speed. And there the mechanical drive is still substantially cheaper per terabyte. And AI mainly generates the second kind. As models grow and data accumulates, demand for cheap capacity grows faster than demand for speed.
And what made it especially profitable: the industry consolidated to just three players after years of attrition. When demand jumped, there was no one to add supply quickly. The result is exactly what shows up in the margin line.
What Was Reported
This is the fourth quarter of fiscal 2026, ended July 3.
| Metric | Current quarter | Year ago |
|---|---|---|
| Revenue | $3,629 million | $2,444 million (+48.5%) |
| GAAP gross margin | 52.3% | 37.4% |
| Adjusted gross margin | 52.7% | 37.9% |
| GAAP operating margin | 43.0% | 23.2% |
| Adjusted operating margin | 44.6% | 26.2% |
| GAAP net income | $1,294 million | $488 million |
| Diluted GAAP EPS | $5.58 | $2.24 |
| Diluted adjusted EPS | $5.71 | - |
And it is worth pausing on the operating margin line. 43% at a hardware manufacturer is an exceptional figure. Companies making a physical product with factories and assembly lines typically run in the single digits to low double digits. 43% is a software company's margin.
Cash flow: $1.3 billion from operations and $1.1 billion of free cash flow in the quarter. The company retired $302 million of debt and returned $283 million to shareholders.
The Full Fiscal Year
| Metric | 2026 |
|---|---|
| Revenue | $12.2 billion (+34%) |
| GAAP gross margin | 45.6% |
| Diluted GAAP EPS | $13.90 |
| Diluted adjusted EPS | $15.58 |
| Operating cash flow | $3.7 billion |
| Free cash flow | $3.1 billion (record) |
| Debt retired | $1.4 billion |
| Returned to shareholders | $810 million |
Note the combination: the company retired $1.4 billion of debt and returned $810 million to shareholders - $2.2 billion in total, out of $3.1 billion of free cash flow. That is the profile of a company using a strong year to strengthen the balance sheet, not merely to distribute.
Guidance
| Metric | Next-quarter guidance |
|---|---|
| Revenue | $4.1 billion (±$100 million) |
| Adjusted EPS | $7.30 (±$0.20) |
And this is the guidance that sets the tone. Moving from $3.63 billion to $4.1 billion is a jump of about 13% in a single quarter. And adjusted EPS is expected to rise from $5.71 to $7.30 - about 28%.
And note the gap between the two: revenue is expected to rise 13% and earnings 28% - meaning management expects the margin to keep expanding.
The company noted the guidance assumes minimal impact from global tariff policies and from the conflict in the Middle East as of the release date.
What Management Said
The company's CEO, Dave Mosley, noted the quarter exceeded expectations on revenue and adjusted EPS, and that the year concluded with 34% growth, record profitability and record free cash flow.
And the reasons he gave map directly onto the thesis: robust cloud data center demand, and an expectation that momentum continues into 2027. He pointed to the Mozaic platform and HAMR technology as engines.
And what is HAMR? It stands for heat-assisted magnetic recording. In short: the technology uses a laser to heat a tiny spot on the disk during writing, which allows far more data to be packed into the same physical area. The result is higher capacity per drive - and therefore a lower cost per terabyte. That is precisely the metric that decides this market.
The Fourth Link of the Evening
This evening gave a rare picture of the AI infrastructure chain, layer by layer:
| Layer | Company | What was reported |
|---|---|---|
| Chip inspection | KLA | Guidance jumping 9% in the quarter |
| Optical connection | Corning | Enterprise networks +65%, but a warning on carriers |
| Power | Bloom Energy | Revenue +165%, guidance doubled |
| Storage | Seagate | Margin from 37% to 52% |
And three of the four layers reported acceleration. Only Corning signaled caution - and specifically in the part unrelated to data centers.
The Bull Thesis
Whoever reads it positively will point to the margin. A jump of nearly 15 percentage points in gross margin is not the result of cost cutting - it is the result of pricing. When a customer pays more per terabyte, it means supply is tight.
Beyond that: guidance implies continued expansion - revenue +13% against earnings +28%. And the market structure supports it: only three manufacturers, and enormous barriers to entry - nobody is building a new disk factory today.
And finally, the balance sheet. $3.1 billion of free cash flow in a year, $1.4 billion of debt retired, and $810 million returned. That is the conduct of a company that remembers it is in a cyclical industry.
The Bear Thesis
Whoever reads it critically will note first the cyclicality. Storage is an industry that has known unusually sharp booms and busts. A 52% margin is not normal - it is a cycle peak. And at a cycle peak, the question is not whether it comes down but when.
Second, customer concentration. Demand comes mainly from a small group of cloud operators. Orders from such customers arrive in waves, not at a steady pace - and when a wave ends, the slowdown is sharp.
Third, structural competition. Flash memory prices keep falling over time. Today the cost gap per terabyte justifies a mechanical drive for mass storage; in a few years, perhaps not.
And fourth, valuation after a run. A stock that has doubled its earnings enters every report with high expectations. Even a strong result may already be priced in.
The debate in one line
The bulls see a gross margin up 15 percentage points, a 43% operating margin, record free cash flow of $3.1 billion, guidance implying continued expansion and a market structure of only three players. The bears see a cyclical industry at a cycle peak, high concentration in cloud customers, long-term pressure from flash prices, and a valuation that already embeds part of the improvement. Both sides are reading the same report.
My Angle
A personal opinion of Ilan Abramov - not advice, not a recommendation
What catches me here is how quickly a thesis can invert.
For years the hard drive was the classic example of a technology in decline. A mechanical product, in a world moving to memory. And then something changed that has nothing to do with the technology itself: demand joined the other side of the equation.
And what I take from it: the differentiation did not come from the product - it came from market structure. Had there been ten drive makers, rising demand would have translated into added capacity and prices staying put. There are three. And so demand translated into margin.
That is a reminder worth holding: when an industry consolidates during a hard period, it emerges with pricing power it did not have before. The survivors benefit from the bad years of the competitors who disappeared.
And what I hold as caution: a 52% margin is not an equilibrium point. In a cyclical industry, a number like that signals we are close to a peak, not that the world changed permanently. Next quarter is guided to $7.30 per share - and if that materializes, it will be an exceptional year. Exceptional is precisely what does not persist.
And what I will follow: the gap between revenue growth and earnings growth in the guidance. 13% against 28%. As long as earnings grow faster, prices are still rising. The day they converge - that will be the sign supply has started to catch up, and that always happens before the numbers show it.
Summary
Seagate closed fiscal 2026 with an exceptional quarter: revenue of $3.629 billion (+48.5%), gross margin jumping to 52.3% from 37.4%, an operating margin of 43.0%, and diluted EPS of $5.58 against $2.24 a year ago. For the full year: $12.2 billion of revenue (+34%) and record free cash flow of $3.1 billion, from which $1.4 billion of debt was retired.
And next quarter's guidance: $4.1 billion of revenue and adjusted EPS of $7.30 - meaning revenue up 13% and earnings up 28%. Management expects the margin to keep expanding.
And beyond the company, this is the fourth link in this evening's AI infrastructure picture - after inspection equipment, optical connectivity and power. And what it adds is a lesson about market structure: the same demand arriving at an industry with ten players produces competition, and in an industry with three - produces margin.
Sources: Seagate Technology Holdings' official results announcement for the fourth quarter and fiscal year 2026 ended July 3, 2026, as filed with the U.S. Securities and Exchange Commission on Form 8-K, including revenue, GAAP and adjusted gross and operating margins, net income and earnings per share, cash flows, debt repayment and shareholder returns, management commentary and guidance for the first quarter of fiscal 2027. Data accurate as of the time of writing. The chart is shown in real time via TradingView. Nothing herein constitutes a forecast, recommendation or advice - see the full disclaimer at the bottom of the page.