Earnings Preview | With AI Trading Concerns Looming, Will Seagate Technology (STX.US)'s "Beat All" Results Still Fall Short?

According to Zhitong Finance APP, Seagate Technology (STX.US) will release its Q1 FY2027 financial results after the U.S. stock market closes on Tuesday (morning of October 28, Beijing time).

The most commonly asked question about this earnings report is no longer "Can it beat expectations again?". For the past five quarters, Seagate has consecutively exceeded both revenue and non-GAAP EPS expectations. This quarter's guidance had already been released on July 28: revenue of $4.1 billion, plus or minus $100 million, and non-GAAP EPS of $7.30, plus or minus $0.20. The consensus forecast by FactSet is revenue of $4.14 billion, a year-over-year increase of 57.5%. More importantly, management stated that the vast majority of nearline storage EB-level capacity has already been locked in through 2028, and shipments to data centers now account for about 90% of the company’s EB-level shipments.

When a company’s capacity and pricing are locked in by long-term agreements for the next two and a half years, there’s little suspense about its revenue. What remains truly unresolved is something else: in a market beginning to doubt whether there is enough capital to sustain the AI boom, this earnings report is being seen as a test of whether the storage chain can still hold—meaning it’s not just a turning point for Seagate’s performance, but a litmus test for the entire storage sector.

Revenue Side: Demand is "a number on a contract," not "a forecasted imagination"

The basis for judging that "revenue is not a suspense" comes from a very tangible set of figures. The FY26Q4 results (ended July 3, 2026) announced on July 28 showed: revenue of $3.629 billion, up 48.5% yoy; GAAP gross margin of 52.3%, non-GAAP gross margin of 52.7% (previous year: 37.4% and 37.9%, respectively); non-GAAP diluted EPS of $5.71, higher than the consensus of $5.09; free cash flow of $1.1 billion. For the whole FY2026, revenue reached $12.195 billion, up 34.1% yoy; full-year free cash flow hit a record $3.1 billion; non-GAAP EPS was $15.58, almost doubling from $8.10 in the previous year.

Looking ahead: the consensus expectation for this quarter is a yoy growth rate of +57.5%, even higher than last quarter’s 48.5%. In other words, even as the market now considers beating expectations the norm, sell-side models are still being revised upward—the Q1 non-GAAP EPS consensus has risen from $5.88 to $7.40 over the past 90 days.

Why is this happening? Because the essence of Seagate’s revenue structure is to translate "AI compute demand" into already-signed contracts. Management said very directly in last quarter's earnings call: backed by long-term supply agreements already in place, most nearline EB-level capacity is locked through the end of 2028, cloud business EB-level shipments have increased quarter-on-quarter for three consecutive years with no signs of slowing. Morgan Stanley analyst Erik Woodring, after cross-checking with HDD industry insiders over the weekend, added: in the past one or two months HDD demand is still accelerating, with new buyers coming from neocloud service providers and "physical AI" companies; agency-model AI has higher data retention requirements than traditional chatbots.

This creates a very important distinction: capital expenditures for training may have cyclical fluctuations, but "data retention" is a rigid expense—each model run generates data that must be stored. This is also why Seagate’s revenue is not like a typical cyclical commodity.

The True Variable Isn’t in Seagate’s Financials, But In Customers’ Capital Expenditures

From another perspective, the reason Seagate’s revenue side is so certain is precisely because its risk has been shifted off its own balance sheet—the ability of its customers to keep paying depends on their own financing capacity. This is the true crux of this financial report.

Just last week, a clear fracture appeared in this financing chain. On October 22, Meta tanked about 7% to close at $648.85, its market value shrinking by approximately $124 billion in a single day. The trigger was raising the lower end of its 2026 capex range by $5 billion, setting it to $135–145 billion, while contract and lease commitments increased by $125 billion to $825 billion. The company candidly admitted in its filings that "fundraising conditions are becoming increasingly difficult". A more sensitive signal came from the credit market: Meta's five-year CDS spread widened by 7.6 basis points to 114.79 bps, a historic high; AI cloud service provider CoreWeave saw some bond yields top 15%, with its stock dropping over 50% from the year's high.

The drag on the broad market was immediate: on October 22, the Nasdaq fell 2.60%, the biggest one-day drop since early June; the Philadelphia Semiconductor Index plummeted 6.7%, and Nvidia dropped 4%. The Nasdaq’s official research column Chartstopper gave a very accurate summary on October 23—although big-cap earnings exceeding expectations mean the Nasdaq 100 may see Q3 earnings growth of 45%, "AI spending fears outweigh revenue growth."

What does this mean for Seagate? It means that even if the company delivers fully on guidance, what the market is really pricing isn’t "is this company good", but "after 2028, will there still be anyone willing to sign contracts".

On the supply side, the answer is technical barriers. Morgan Stanley points out that, while repurchasing, it's notable that Toshiba lacks high-capacity nearline capability and Heat-Assisted Magnetic Recording (HAMR) technology; Citi adds a harder constraint: Toshiba doesn’t produce recording media or heads, and its upstream suppliers also need capacity expansion to turn the so-called "doubling of capacity" into real shipped drives. For Seagate’s own execution, it depends on ramping up the 44TB Mozaic 4 HAMR platform at the two largest cloud service providers; the company plans for Mozaic 4 to account for 50% of HAMR EB-level shipments by the end of calendar year 2026. Since such product transitions inevitably involve yield fluctuations, whether they can deliver close to 50% operating profit margin guidance is the real test of this "moat".

Peer Resonance: Western Digital’s Recent “Perfect Earnings, Awful Stock Performance”

If more evidence is needed that "great performance is already not enough," peers have already played out the script. On October 22 after market, Western Digital (WDC.US) released FY27Q1 results: revenue of $4.381 billion, up 55% yoy and 17% qoq; GAAP gross margin 57.8%, non-GAAP gross margin 58.1%, up 1420 bps yoy; non-GAAP operating margin 49.1%; non-GAAP diluted EPS $4.64, up 161% yoy; operating cash flow $1.604 billion, free cash flow $1.377 billion; quarterly dividend raised 17% to $0.175 per share. Even more aggressive was the guidance: CFO Kris Sennesael stated that FY27Q2 midpoint forecasts revenue of about $5.1 billion (up 66–72% yoy), non-GAAP gross margin 60.5%, non-GAAP EPS $5.80—gross margin guidance still rising. Yet, the stock dropped 3.7%.

Micron (MU.US) was even more extreme. FY26Q4 (reported after the close on Sept 30) revenue of $54.229 billion versus $11.315 billion a year ago, a 379% yoy increase; GAAP gross margin 86.8% compared to 44.7%; FY27Q1 guidance revenue of $61.5 billion, plus or minus $1.5 billion. Management disclosed that 26 Strategic Customer Agreements (SCA) have been signed, remaining performance obligations (RPO) about $150 billion, client funding commitments up to $32 billion, and agreements now extended to 2031; CEO Sanjay Mehrotra stated that supply-demand for storage will be even tighter in 2027-2028 than in 2026, and explicitly identified physical AI as the next major growth market.

Though their numbers and cycles differ, all three point to the same thing: simultaneous increases in volume and price, rising gross margins, visibility extended to 2028 or even 2031—fundamentals have never been better, but prices have never been harder to move. This is exactly what a "litmus test" means: data can no longer convince the market, the only thing left is whether "visibility" can be extended another notch.

Oil Prices, U.S. Treasuries and AI Panic: Triple Headwinds on Valuations

This earnings report lands in a very crowded macroeconomic window. On the oil front, as Yemeni Houthi forces attacked Saudi energy facilities, Brent crude settled at $116.72 per barrel last Friday (Oct 23), the highest close since April 29; on Monday (Oct 26), it retreated to $114.51 intraday, with WTI at $101. High oil prices are directly increasing inflationary persistence, providing a solid reason for central banks to hold rates steady.

On the U.S. Treasuries side, Oct 22 saw safe-haven flows out of AI stocks and into treasuries; the 10-year U.S. Treasury yield fell 19.3 bps to 4.953%—the largest single-day drop since March 2023—with the 30-year yield down 15 bps to 5.287%. Barclays' head of U.S. inflation strategy Jon Hill was blunt: the day's gains in Treasuries "were not in response to any fundamental data, but rather related to potential problems in AI trades". By this Friday, the 10-year yield is around 5.0%.

These two forces combined are very unfavorable to high-valuation, long-duration assets like Seagate: its earnings certainty is high, but so is its discount rate, and its valuation already assumes perfection. As of the close on Oct 23, STX traded at $630.49 (down 1.81% on the day), up 129.69% year-to-date, but down nearly 32% over the last month—the 52-week range is $217.67 to $1,145.00, with a trailing P/E of 45.36. When the discount rate stands near 5% and the CDS market is repricing the AI funding chain, every year that "visibility past 2028" is extended, real value is added; conversely, any hint that "visibility has peaked" will be amplified by the market.

Miller Tabak chief market strategist Matthew Maley’s summary is a good footnote for this environment: the core reason for recent sell-offs is the market's "serious doubts about huge returns on investment, as it becomes increasingly clear that this will take much longer."

Wall Street Remains Bullish Overall, But AI Trade Worries Loom

Morgan Stanley (Erik Woodring): Maintains Overweight rating, STX target price $1,187, WDC target $676, with STX as the top pick. He said Toshiba’s October 2 news "created opportunity rather than changed the argument", and cited channel data showing customers without contracts paying $0.03 to $0.05 per GB for nearline allocation, with some deals even higher.

Bank of America Securities (Wamsi Mohan): On Oct 21 maintained a Buy, raising target from $1,150 to $1,200.

Rosenblatt: Maintains Buy, but categorizes Toshiba’s expansion as a real mid-term risk for 2029–2031 contracts; Citi counters the "flood of capacity" narrative with supply chain constraints; Goldman Sachs rates STX a Buy with a target of $960; Evercore ISI (Amit Daryanani) includes both companies among the "six most worthwhile stocks to hold for the second half".

Meanwhile, the options market isn’t buying it: according to 247wallst data, the put/call ratio for the entire chain is 1.06, leaning cautious; Seagate’s own comparable fund flow also signals cautious sentiment.

Taking all this together, the conclusion is actually clear: since revenue and margins are nearly written into contracts, the real question to be answered after the October 27 report is whether management can extend the phrase "locked in through 2028" forward to 2029 or even further. If yes, storage remains the most resilient layer in the AI infrastructure, and the current pullback is merely a valuation reset; if not, the market may reclassify it from an "AI structural asset" back to a cyclical HDD business—even if this quarter’s numbers are stellar.

The current scenario aligns more with Rosenblatt’s "walking a tightrope" description: acknowledging real mid-term risk, yet still maintaining a Buy. The issue has never been about this quarter’s numbers, but rather whether the AI spending spree can continue to pave the way for the next decade.