Seagate Stock Fundamentals 2026: STX Earnings, Cash Flow and Valuation

Toshiba has committed roughly JPY 60bn — about USD 380m — to double its hard-disk-drive production capacity for AI data centres by fiscal 2027, and the two incumbents wore the news: Seagate fell about 10% on 2 October and Western Digital fell alongside it. It is Toshiba's first major hard-drive investment in roughly five years, centred on its Philippines plant. Toshiba currently holds just over 10% of the hard-drive market measured by capacity and has said it wants 30%.

The reason that is the central fact in any look at Seagate stock fundamentals is a single number from its own accounts. Seagate's group gross margin went from 18.3% in fiscal 2023 to 45.6% on a GAAP basis in the year to June 2026 — 46.1% non-GAAP. That expansion was not a cost programme. It was scarcity: an industry running without spare nearline capacity while cloud operators and AI builders absorbed every high-capacity drive available. Toshiba has now committed capital to easing exactly that scarcity, and the market spent a session working out what a 45.6% gross margin is worth without it.

The industry condition that changed

Hard-disk supply has spent three years as a seller's market. Nearline drives — broadly the 18TB-to-26TB-and-beyond class used in data lakes, backup and object storage — stopped being a commodity and became a build-to-order product, allocated to hyperscalers under long-term agreements with volume commitments in exabytes and contracted pricing. That shift is the mechanism behind the margin expansion at both US leaders. When output is sold before it is made, the manufacturer sets terms.

Toshiba is the number three producer behind Seagate and Western Digital, and a doubling of its AI-data-centre drive capacity inside fiscal 2027 is the first credible supply response from outside the duopoly. The investment is modest in absolute terms — USD 380m is a rounding error against what the memory industry is spending — but it is directed at the precise bottleneck that has been generating the incumbents' pricing power, and it comes with a stated ambition to triple market share.

The sell-side reaction was notably cooler than the share-price reaction. Analysts at Morgan Stanley, Rosenblatt and Citi largely treated the drop as overdone, keeping positive ratings on both names on the argument that Toshiba's plan is unlikely to close the industry's supply gap. That is a defensible position, and the reason is in the order book rather than in the capacity maths.

What it means for Seagate specifically

Seagate is the most directly exposed listed name, because high-capacity nearline is the core of its book rather than an adjacency. Its fiscal fourth quarter, reported in late July 2026, showed revenue of USD 3.6bn, GAAP gross margin of 52.3% — 52.7% non-GAAP — free cash flow of USD 1.1bn and operating cash flow of USD 1.3bn. Data-centre revenue rose 57% year on year to USD 2.9bn in that quarter, and HAMR products reached approximately 40% of nearline exabyte shipments. For the full fiscal year, revenue was USD 12.2bn with free cash flow of USD 3.1bn.

Note the direction of travel inside those figures. The quarterly GAAP gross margin of 52.3% sits nearly seven points above the full-year 45.6%, which means margin was still expanding sharply through the final quarter of the fiscal year. Whatever Toshiba does by fiscal 2027, it has not yet touched Seagate's realised pricing.

The more important disclosure for this particular risk is contractual. Seagate has said the vast majority of its nearline exabytes are allocated through calendar 2028 under long-term supply agreements. Western Digital's position is comparable: its chief executive described the company as essentially sold out for calendar 2026 on firm purchase orders from its top seven customers, with long-term agreements already in place covering two customers for 2027 and one for 2028.

That is the crux. New capacity arriving in fiscal 2027 lands in a market where the incumbents' output is already committed, at agreed volumes and agreed prices, into 2028. Toshiba's drives do not displace contracted exabytes; they compete for the increment beyond them. The question Toshiba's investment genuinely raises is not what happens to Seagate's revenue next year — that is largely written — but what the renewal round looks like when those agreements come up, and with how much more supply in the room.

The Openbook read

Profitability is the factor carrying the entire weight here, and it remains the strongest of the five on current evidence while becoming the most assumption-dependent. A GAAP gross margin that has gone from 18.3% to 45.6% in three years, with the latest quarter at 52.3%, is an exceptional outcome for a hardware manufacturer. The honest framing is that this is a cyclical margin achieved through a structural shortage, and the score should be read as high and contingent rather than high and durable. Toshiba's announcement does not change the current number. It changes how much of it you should capitalise.

Growth is intact and visible. Data-centre revenue up 57% year on year, HAMR at roughly 40% of nearline exabytes, and guidance for fiscal first-quarter 2027 revenue of USD 4.1bn plus or minus USD 100m against USD 3.6bn delivered in the quarter just reported. That guided figure implies continued sequential growth, with non-GAAP diluted EPS guided to USD 7.30 plus or minus USD 0.20. The growth score is being driven by AI storage demand rather than by unit volumes, and the HAMR transition gives Seagate a cost-per-terabyte lever that is independent of the pricing environment.

Solvency is the quiet improvement in the story and is often underweighted when attention is on margin. USD 3.1bn of free cash flow in a single fiscal year against USD 12.2bn of revenue is a conversion rate that repairs a balance sheet quickly. A company generating cash at that rate has the capacity to absorb a weaker pricing cycle without the financing stress that would normally accompany a commodity downturn in this industry. Whatever the supply picture does, this is not a solvency story.

Momentum has broken. A roughly 10% single-session fall is a genuine trend interruption, and it came without any change to Seagate's own trading, guidance or order book — it was a third party's capital-expenditure decision. Momentum scores do not distinguish between a company-specific disappointment and a sector-wide re-rating of a margin assumption, so the factor registers the full move. The sell-side's view that the reaction was overdone is a statement about valuation, not about momentum, and the two can both be right.

Reward/Risk is where the scores interact in an interesting way. The reward side is anchored by the contracts: revenue and, to a large degree, pricing are visible into 2028, and the near-term guidance points to further sequential growth. The risk side is not about next year at all — it is about the terminal margin. If 45.6% GAAP gross margin proves to be the new normal, one set of conclusions follows; if it proves to be a cycle peak that reverts as supply normalises, a very different set does. Toshiba's commitment is the first dated, funded piece of evidence on which of those is right. That is why a USD 380m investment by the number three producer moved so much combined market value in a single session.

Read the five factors together and the picture is a company with excellent current profitability, real growth, a strong cash position and freshly damaged momentum, where the entire reward/risk question reduces to the durability of one margin.

Which other listed names sit in the same flow

Western Digital is the closest comparison and fell on the same news, with reported declines ranging from about 7% to about 10% across sources. Its exposure is of the same kind rather than merely similar: nearline HDD, allocated under long-term agreements, with margin built on the same shortage. Anything that is true about the terminal-margin question at Seagate is true at Western Digital, with the difference that Seagate's HAMR transition is further advanced.

Toshiba itself sits on the other side of the trade as the share gainer, though for a diversified Japanese industrial the drive business is one line among many rather than the whole investment case.

The read-across to the memory names is weaker than it looks and should be handled carefully. DRAM and NAND are in their own, separate up-cycle with their own capacity economics; HDD and flash compete at the margins of the storage hierarchy rather than head to head in nearline bulk capacity. A supply response in hard drives is not a supply response in memory. Investors screening the storage complex on our screener are better served separating the two than treating "AI storage" as one exposure.

What to watch next

• Seagate's fiscal first-quarter 2027 results. The quarter ended on 2 October 2026 — the same day as the selloff — so the next release reports a period that closed before the news landed. The reference points are the guided USD 4.1bn of revenue and USD 7.30 of non-GAAP diluted EPS, and more importantly whether the gross margin held near the 52.3% GAAP level of the prior quarter.
• Commentary on long-term agreement renewals. With the vast majority of nearline exabytes allocated through calendar 2028, the renewal cycle rather than current trading is where Toshiba's capacity will first show up in pricing. Any management language on 2029 terms is the signal.
• Toshiba's execution against fiscal 2027. Capacity announcements and capacity delivery are different things, particularly in a single-plant expansion. Evidence that the Philippines ramp is on or behind schedule is the direct input.
• Western Digital's next results. A second, independent read on nearline pricing and allocation from the only comparable operator.

The industry condition has changed, but the contracts have not. That gap — between a supply response announced now and an order book committed through 2028 — is the space in which this argument will be settled, and it is a matter of quarters rather than days.