Seagate (STX) · Deep analysis

Report previewSTX.USReport date Last updated
← Back to reports
Report contents

At a glance

  • Business model: Seagate sells nearline hard drives by capacity in a three-supplier market, where each platform must be qualified into a hyperscaler's rack and density, led by the HAMR Mozaic 4 ramp, sets the price reference for rivals. Qualification slows displacement, but buyers can bargain each density gain away at renewal.
  • Financial quality: FY2026 operating cash flow of $3.67 billion covered net income 1.15 times and left $3.10 billion of free cash flow, though conversion eased to 0.96 times in the September 2026 quarter. Debt fell to $2.35 billion with interest cover above 50 times, so the balance sheet no longer constrains the business.
01

Business model

Business model: density is the product, drives are the packaging

Seagate Technology Holdings plc builds hard disk drives and sells them by the unit, but what buyers actually pay for is storage capacity per dollar. The customer list is short: hyperscale cloud operators and storage OEMs that fill racks with high-capacity nearline drives, plus distributors serving video, surveillance, and small-business users. Revenue is booked when drives ship against purchase orders, so quarterly sales follow customer capex and inventory decisions, not subscriptions. This is a manufacturer in a three-supplier industry, well past expansion and into maturity, with the HAMR (heat-assisted magnetic recording) transition the main lever on cost per terabyte. The fiscal year ended July 3, 2026, per the 10-K, and the company reports in US dollars to a global customer base.

Exabytes grow faster than revenue, by design

Demand is counted in exabytes, and it rises with cloud storage and the data used to train and serve AI models. The complication is that density gains get handed to customers. Each generation fits more terabytes on a platter, and the industry has historically converted most of that improvement into a lower price per terabyte. Exabytes shipped can therefore grow at a healthy clip while revenue grows more slowly.

What lifted fiscal 2026 was the other side of that equation. Seagate reported revenue above $9 billion at a gross margin near 45%, against 35.2% in fiscal 2025 and 19.2% in fiscal 2023. A roughly ten-point margin move says pricing on scarce nearline capacity did much of the work, which makes gross margin, not unit volume, the variable that swings earnings. Seagate reports revenue and exabytes by market but not revenue by end customer, so the cloud share of demand has to be inferred from earnings commentary rather than read off a table.

Who holds the leverage

Hyperscale buyers purchase in multi-quarter volumes and qualify each drive platform before it enters a rack. Qualification is slow and expensive, and it cuts both ways. Once a drive is designed in, switching costs are real; at the next platform decision, the buyer has three qualified suppliers and can bargain the density gain away as a discount. Distributors and OEMs extend reach into smaller accounts but add little pricing power, and their inventory decisions amplify the cycle in both directions. The 10-K's customer-concentration disclosure is the cleanest read on how much negotiating leverage sits on the other side of the table, and it likely understates end-demand concentration, since one cloud operator can reach Seagate through several contract manufacturers.

Pricing power comes from the cost curve

No buyer pays a premium for the Seagate name. What Seagate sells is a cost per terabyte competitive with Western Digital and Toshiba, and whichever company ships the highest capacity per drive at volume sets the reference point for the others. HAMR is the current test. Management expects half of HAMR exabytes to come from the four-terabyte-per-disc Mozaic 4 platform by the end of calendar 2026, S&P Global reported. Reaching that would carry the density lead into 2027 and support pricing when supply loosens. A slower ramp would leave Seagate competing on price with a thinner capacity advantage, which is roughly where the fiscal 2023 margins came from.

The legacy end markets, consumer and video applications, remain a small and structurally shrinking part of the mix. They matter less for growth than for factory loading: they absorb capacity that would otherwise sit idle when nearline orders pause.

The fragility sits in the fixed-cost base

Drive assembly is capital intensive and capacity arrives in lumps. A line conversion takes quarters, and once in place it runs whether or not demand holds, so a volume decline spreads depreciation and factory overhead over fewer drives. That is the mechanism behind the fiscal 2023 margin trough, not a collapse in storage demand. Three suppliers sharing a market with slow-growing unit demand have a strong incentive to keep capacity tight, and the recovery since fiscal 2023 partly reflects that discipline. There is also a ceiling on pricing at the low-capacity end, where QLC NAND (flash memory storing four bits per cell) has become a cheaper substitute.

What would change this view

The core judgment is that Seagate's earnings are set by cost per terabyte relative to price per terabyte, and by the industry's willingness to add capacity slowly; the density roadmap is the one lever the company controls. Two consecutive quarters in which revenue per exabyte falls faster than average capacity per drive rises would undercut that, as would exabyte share loss paired with rising capex. Either would suggest fiscal 2026's margin reflects shortage more than a durable cost advantage.

Questions worth following

  • Can Seagate hold nearline pricing as hyperscale buyers absorb its mass-capacity output?

    Nearline drives anchor the profit pool, so the answer decides whether revenue growth comes from more exabytes or from price per terabyte. If prices erode while volumes rise, Seagate funds expensive capacity expansion with thinner margins and the cycle turns faster.

  • Will the HAMR transition keep Seagate at the leading capacity point with cloud customers?

    Areal density leadership determines who gets first call on the highest-capacity sockets, where pricing is best. A delayed ramp, poor yields, or narrow qualification would push mix toward lower-priced drives and hand the top node to the rival supplier.

  • Will the three drive makers keep capacity discipline as AI demand rises?

    Fiscal 2026's margin depends heavily on tight nearline supply. If Seagate, Western Digital and Toshiba add lines into the AI buildout, price per terabyte falls before utilization does, and the earnings base underlying the current profitability proves cyclical rather than structural.

02

Financial quality

Financial quality: cash is real, the margin is unproven

Seagate's earnings turned into cash in fiscal 2026, and the balance sheet those earnings once strained has been repaired. What remains open is whether the margin that produced them survives looser supply. Gross margin went from 35.2% in FY2025 to 45.6% in the year to June 30, 2026, and to 57.3% in the September 2026 quarter, when Seagate reported revenue of $4.23 billion (up 61% from a year earlier) and net income of $1.76 billion.

Returns: Exceptional, on a shrunken denominator. ROE is unusable here. Equity fell to negative $1.49 billion in FY2024 and was still negative in FY2025, so the 371.5% figure for FY2026 measures a rebuilt equity base, not a return on a stable one. ROA of 35.4% and after-tax operating profit of roughly $3.46 billion, set against about $4 billion of debt plus equity less cash, put returns far above any plausible cost of capital. The denominator was cut by three loss years and buybacks, so the ratio overstates what the asset base earns. The cleaner read is incremental: gross profit rose $2.36 billion in FY2026 on $3.10 billion of added revenue.

Metric (USD bn unless noted)FY2022FY2023FY2024FY2025FY2026Q1 FY2027Assessment
Revenue11.667.386.559.1012.204.23Adequate
Gross margin (%)29.719.223.435.245.657.3Exceptional
Net margin (%)14.1-7.25.116.126.141.6Exceptional
ROE (%)445.797.1-24.9-151.1371.562.7Not meaningful
ROA (%)18.7-6.44.418.635.417.4Strong
Operating cash flow1.660.940.921.083.671.69Exceptional
Free cash flow1.280.630.660.823.101.48Exceptional
OCF / net income (x)1.01n/m2.710.731.150.96Strong
Implied capex / revenue (%)3.34.24.02.94.75.0Adequate
Current ratio (x)1.131.121.081.381.591.82Adequate
Quick ratio (x)0.690.680.680.841.111.25Adequate
Total debt5.665.505.685.063.562.35Strong
Total liabilities / assets (%)98.8115.7119.3105.778.266.5Weak
EBIT / interest expense (x)7.7n/m2.45.714.250.5Strong
Asset turnover (x)1.320.900.861.151.360.42Adequate
Inventory turnover (x)5.924.414.224.404.411.10Adequate
Receivables turnover (x)8.416.159.8711.249.332.53Adequate

*Three months to September 30, 2026. Fiscal years end June 30. Turnover and return ratios in the final column are quarterly, not annualized. n/m = not meaningful on a negative base.

A margin built on pricing, with capex near 5% of revenue

Capex implied by operating and free cash flow was $0.57 billion in FY2026, or 4.7% of revenue, and about 5% in the September quarter. That is low for a disk manufacturer. Revenue rose 34% while gross profit rose 74%, so operating leverage came through price and mix; the asset base barely grew. Good for cash, less reassuring for durability, since nothing on the balance sheet locks the improvement in. The open question is whether HAMR media and nearline capacity are being paid for as they ramp or deferred to protect the current return.

The margin's source is the thing to test

Toshiba's plan to roughly double its AI data-center drive capacity by fiscal 2027, reported in early October 2026, is the first supply response from outside the top two producers, and Seagate shares fell about 10% on October 2, 2026 when it surfaced. Published commentary in late October argued the larger risk is capacity landing in 2028 instead. The capacity headline matters less than the contract: exabyte volumes and pricing under long-term agreements, and revenue per exabyte shipped, which Seagate can compute from disclosures it already makes. Flattening revenue per exabyte while exabytes grow would mark the margin as a shortage rent.

Cash conversion held, then slipped a notch

FY2026 operating cash flow of $3.67 billion covered net income 1.15 times, and about 85% of it survived capex. In the September 2026 quarter conversion fell to 0.96 times even with free cash flow of $1.48 billion. Working capital stayed orderly: receivables of $1.82 billion rose 52% against 61% revenue growth, holding collection periods near 36 days, and inventory of $1.70 billion rose 13%, leaving inventory days near 83. Cash still fell $0.44 billion in the quarter because $1.81 billion left through financing against $1.48 billion of free cash flow.

USD bn unless notedSep 2025Dec 2025Mar 2026Jun 2026Sep 2026AssessmentTrend
Revenue2.632.833.113.634.23ExceptionalRising
Net income0.550.590.751.291.76ExceptionalRising
Total assets8.448.718.899.9710.33AdequateRising
Total liabilities8.508.257.807.806.87WeakFalling
Total equity-0.060.461.092.173.45AdequateRebuilding
Operating cash flow0.530.721.111.311.69ExceptionalRising
Free cash flow0.430.620.951.121.48ExceptionalRising
Implied capex0.100.100.160.190.21AdequateRising slowly
Gross margin (%)39.441.646.552.357.3ExceptionalRising
Net margin (%)20.920.924.035.741.6ExceptionalRising
Current ratio (x)0.981.121.331.591.82AdequateImproving
Leverage: repaired, with room to spare

Total debt fell from $5.66 billion in FY2022 to $3.56 billion at June 30, 2026 and $2.35 billion at September 30. Against $1.26 billion of cash, net debt is roughly $1.1 billion. The current ratio moved from 0.98 to 1.82 and the quick ratio from 0.62 to 1.25 as equity climbed from negative $0.06 billion to $3.45 billion. Interest expense of $40 million against $2.02 billion of quarterly EBIT covers it more than 50 times. On total liabilities, though, the ratio is still 66.5%, so suppliers and other non-debt obligations remain a large share of the balance sheet.

What needs checking
  • Capex against depreciation. Free cash flow here is operating cash flow less capex, and Seagate does not split maintenance from expansion spending in that figure. A capex-to-depreciation ratio persistently below 1 while exabyte shipments climb would suggest current returns are being earned by sweating assets.
  • Distributions outrunning cash generation. Financing outflows exceeded free cash flow by about $0.33 billion in the September 2026 quarter. Debt repayment explains part of it, but two more quarters of the same shape would mean returns are funded from the cash pile.
  • Customer concentration in receivables. A 52% year-over-year rise in receivables is consistent with hyperscaler-led growth, yet it also means a handful of buyers now set payment terms. A jump in collection days past 45 would be the first sign of that leverage shifting.
What would change the assessment

Two consecutive quarters of operating cash flow below 0.9 times net income, gross margin back under 45%, or capex above 8% of revenue without matching exabyte growth. Any of those would move the reading from exceptional cash generation toward a peak-cycle verdict.

Questions worth following

  • Does Seagate's 60x P/E reflect durable earnings power or a margin peak the next downcycle erases?

    The market cap assumes FY2026 profitability persists. Gross margin jumped from 35.2% in FY2025 to roughly 45% in FY2026 while revenue rose sharply, so the multiple rests on cycle-peak margins rather than a long-run average the company has held before.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Go beyond the preview

Continue reading the full report

  1. Key risksWhat could break the investment case
  2. Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
  3. SummaryOur overall judgement and what to watch next

Free to sign up · 20 research points · unlocking this report uses 10

20 points cover 2 full reports. No card needed.

Research reflects its generation date. Preview the first two modules.

More on Seagate Technology PLC