Seagate: HAMR kicks off "profiteering mode"—is the golden age of hard drives just beginning?

① **Doubling the growth target is the core signal:** Management raised the annual revenue growth target from "low to mid double digits" to "at least 20%," backed by locked nearline production capacity until CY2027, full-year FY2027 BTO contracts with confirmed quantity and price, and the three major CSPs' RPO doubling to $1.1 trillion. ② **Gross margin at a historic high of 47%, incremental gross margin over 70%:** Non-GAAP gross margin jumped from 36.2% YoY to 47.0%, incremental gross margin reached 71.6%, well above analysts’ target of 50%. The key driver is the increase in HAMR areal density—achieving 30%+ capacity growth with the same number of disks/heads and negligible material cost changes. ③ **Data center revenue up 55% YoY to $2.5 billion, accounting for 80% of total revenue:** 175EB data center shipment volume up 47% YoY. Per-EB data center revenue up about 5% YoY. The dual engine of volume and price increase is now established. ④ **FQ4 guidance accelerates: Revenue $3.45 billion (+41%), operating margin breaks 40% for the first time:** Non-GAAP EPS guidance of $5.00, up 22% QoQ. Management expects quarterly revenue and margin to continue sequential growth until FY2027. ⑤ **FCF nearly $1 billion, a ten-year high; Fitch upgrades to investment grade:** Free cash flow $953 million, margin 31%. $641 million debt repaid this quarter, net leverage down to 0.7x. Remaining $400 million convertible bonds planned to be dealt with over next 1-2 quarters, after which focus will shift to share buybacks. ⑥ **Unit economics reveal the real profit driver:** Per-EB cost down 14.4% YoY while per-EB revenue rose only 3.7%—the cost scissors contributed to most of the margin expansion. This means even with moderate price increases, HAMR-driven cost curves can continue to raise margins. The most important signal in this financial report is not how good the numbers are for a particular quarter—especially since last year was still at an industry low—but that management has rarely raised the overall growth targets, backed by BTO contracts locked until CY2027. Combined with WDC FQ3 guidance also reaching 47-48% gross margin, the margin recovery in the HDD industry is systemic, not a short-term phenomenon of a single company. The more noteworthy forward variable is the pace of the HAMR technology roadmap. Mozaic 4 achieves 30%+ capacity growth with the same number of disks and heads and nearly unchanged material costs. Once this economic model continues in Mozaic 5 (50TB, end of CY2027), per-EB cost will reach a level where SSDs can't compete in the cold storage tier. WDC has yet to mass produce HAMR, marking Seagate’s biggest differentiator right now, but this window is time-limited—if WDC achieves HAMR mass production before CY2028, pricing power may be redistributed. **Detailed analysis of financial report content** Seagate uses a quarterly report with 44% YoY revenue growth, 47% Non-GAAP gross margin, and nearly $1 billion in free cash flow to formally announce its entry into a phase of “structural growth.” Management raised the annual revenue growth target from “low to mid double digits” to “at least 20%”—this is a framework shift, not a simple numerical adjustment. Anchors supporting this judgment are very specific: Nearline HDD capacity is “almost entirely allocated until CY2027,” FY2027 full-year BTO contracts locked for product mix, shipment, and price. The three global CSPs’ RPO (remaining performance obligations) nearly doubled to $1.1 trillion, the strongest indicator of persistent demand. **Data Center: Dual engines of volume and price increase** Data center revenue about $2.5 billion, up 55% YoY, accounting for 80% of total revenue. 175EB data center shipments grew 47% YoY, nearline shipments accounted for 88% of total EB. In unit economics, data center per-EB revenue increased about 5.4% YoY—consistent with CFO’s “mid-single digits”—the rising proportion of HAMR-driven 40TB+ products was the main driver. Data center revenue growth rate (+55%) is significantly faster than shipment growth (+47%), indicating accelerated improvement in price/structure. Mozaic 4 has shipped to 75% of leading CSPs, with certification for the remaining two expected to be completed in FQ4. As HAMR EB crossover is achieved at end of CY2026, the share of high-capacity products will continue to boost unit revenue. Comparatively, WDC FQ2 data center revenue growth was about 25%, Seagate’s gain from explosive cloud storage demand is clearly greater. **Margin: Areal density economics behind 71.6% incremental gross margin** Non-GAAP gross margin increased from 36.2% last year to 47.0%, up 10.8 percentage points YoY. More telling is the incremental gross margin—in this quarter’s $952 million new revenue, $682 million was converted into Non-GAAP gross profit, reaching 71.6%, far exceeding last year’s analyst target of 50%. Looking at per-EB cost, this quarter’s per-EB cost was about $8.37M, down 14.4% YoY; per-EB revenue only rose 3.7% to $15.64M. The real engine of margin expansion is the cost side, not the price side. HAMR technology’s economics are straightforward: Mozaic 4 achieves 30%+ capacity increase without increasing disk/head count, with minimal material cost changes; areal density improvement directly translates to per-EB cost decline. When asked about sustaining 70%+ incremental margin, CFO said “we see no reason it would change,” with confidence from HAMR’s next-gen Mozaic 5 (50TB) iteration still ahead. Notably, WDC FQ3 guidance now reaches 47-48% gross margin, comparable to Seagate’s levels. This shows profit margin recovery in the HDD industry is systemic, with supply-side discipline (duopoly structure) and demand-side pull (AI storage) affecting both companies similarly. **Operating margin 37.5%, GAAP legal settlement dragged by 3.4 percentage points** Non-GAAP operating margin 37.5%, up 14pp YoY, up 5.6pp QoQ. GAAP operating margin at 32.1%, with the biggest difference being $105 million in legal settlement fees this quarter (about 3.4% of revenue). Stripping out this one-time item, true operating margin is about 35.5%, still a historic high. Non-GAAP operating expenses $296 million, only 9.5% of revenue, already hitting management’s long-term target. Management explicitly said OpEx will “remain roughly flat in absolute dollar terms,” unleashing expense leverage as revenue grows. **Net profit: $69 million loss on early debt repayment** GAAP net profit $748 million includes $69 million loss on early debt repayment (retired $641 million debt this quarter, including over $600 million in 2028 convertible bonds). Non-GAAP net profit $934 million, Non-GAAP EPS $4.10, up 116% YoY. **Cash flow and capital allocation: Deleveraging nearly done, buyback about to start** Free cash flow $953 million, up 57% QoQ, a ten-year high. FCF margin 31%. Debt retired year-to-date about $1.1 billion. Ending cash $1.146 billion, total liquidity $2.4 billion (including unused revolver). Net leverage down to 0.7x, Fitch has upgraded Seagate’s credit rating to investment grade. Remaining ~$400 million convertible bonds are planned to be settled in FQ4 or next quarter. Management made clear in the earnings call: after deleveraging, “most will be used for buybacks.” The company has already begun buybacks in the market. With about $3.8 billion in annualized FCF generation, and CapEx only 4-6% of revenue, Seagate can release over $3 billion annually for shareholder returns, a significant EPS growth lever. **Edge/IoT: Steady but not yet benefiting from HAMR migration** Edge/IoT revenue $612 million, about 20% of total revenue, up 2% QoQ. Supply tightness and high net costs offset seasonal declines in client/consumer business. Management said the current 40TB+ HAMR products are too in-demand in the cloud, so there’s not enough capacity to migrate down to lower capacities—but after Mozaic 5 mass production, using 5TB/platter tech to make 20TB products will have strong cost economics. HAMR penetration in the Edge market is a CY2028 onward story, and not a short-term catalyst. **Management signal: Entering the 'visibility lock-in period' from 'execution validation'** The earnings call’s core signal was clear: visibility. BTO contract lock-in until FY2027 means high certainty for the next four quarters’ revenue and margin. Six analysts repeatedly asked about pricing sustainability (Morgan Stanley, JP Morgan, Bernstein, Goldman Sachs, etc.), management responded consistently—“We have not changed pricing strategy, margin improvements have been consecutively sustained for 12 quarters, same trend will extend to FY2027.” This consistent phrasing is itself a signal. A year ago, management’s “low to mid double digit” growth target was met with skepticism. Now, not only are results far beyond target, but the target is proactively raised to “at least 20%,” anchored with locked-in contracts. This shift from conservative to proactive usually marks that a company’s confidence in its own growth engine has crossed the critical threshold. UBS’s Tim Arcuri asked whether unit shipment volume grew; management admitted “no”—all growth came from areal density innovation. This is essentially the optimal economic model: no extra components, no expansion of production lines, just technical innovation yielding more EB output. **Forward-looking Strategic Analysis** **HAMR technology roadmap: Locking in at least a 12-18 months leadership window** Mozaic 4 (4+ TB/platter, 44TB/drive) has started commercial shipments, expected to dominate HAMR EB shipments by late CY2026. HAMR EB crossover achieved at the same time. By end of FY2027, about 70% of nearline EB will be HAMR-based. Mozaic 5 (50TB) R&D is progressing well, aiming for certification/shipment by end of CY2027. WDC currently relies on ePMR+UltraSMR tech, capped at 28TB. Mass production timeline for HAMR is not yet disclosed. If HAMR progresses as planned, Seagate’s per-EB cost advantage will keep expanding for the next 2-3 years. But after CY2028, WDC’s HAMR catch-up pace will be key to reevaluating the competitive landscape. **AI inference inflection point: From sporadic queries to sustained autonomous workflows** Management, for the first time, systematically explained Agentic AI’s storage demand impact—from sporadic queries to sustained autonomous workflows, referencing large historical datasets for inference and generating new unstructured data. Physical AI (autonomous driving/robots) can generate up to 4TB/hour per vehicle, with storage needs for 5-10 years. Management believes these trends support mid-20% nearline EB CAGR. If Agentic AI’s data density exceeds expectations, growth could be higher; if AI spending cycles slow, EB growth could return to 15-20% range. **Capital Sustainability** Current FCF annualized at about $3.8 billion, CapEx guidance 4-6% of revenue (annual $0.5–0.75 billion), net free cash flow over $3 billion. Once the remaining $400 million convertible bonds are resolved, capital allocation focus switches to buybacks. With a market cap around $23 billion, annual buyback capacity is >13% of value, providing significant EPS accretion. After deep cyclical lows in 2022-2023, the HDD industry is entering an unprecedented profit zone, driven by AI infrastructure demand and supply discipline in a duopoly. Seagate’s locked-in contracts, validated tech roadmap, and rebuilt balance sheet offer the most convincing annotation so far for the phrase “structural growth.” Risk warning and disclaimer The market is risky, and investment requires caution. This article does not constitute personal investment advice and does not account for individual users’ specific investment objectives, financial situation, or needs. Users should consider whether any opinions, viewpoints, or conclusions herein suit their situation. Investing based on this is at your own risk.