tapebrief

WDC · Q2 2026 Earnings

Bullish

Western Digital

Reported January 29, 2026

30-second summary

Western Digital beat its own Q2 guide on every line (revenue $3.02B vs. $2.8–3.0B, EPS $2.13 vs. $1.73–2.03, gross margin 46.1% vs. 44–45%), and the Q3 guide is the real signal: revenue $3.1–3.3B with management framing ~40% YoY growth at the midpoint — a step-change from this quarter's 25% YoY print. Gross margin guide steps to 47–48%, customer commitments now extend to 2028, and capital returns exceeded 100% of FCF. The supply-constrained-through-2027 thesis is no longer being argued — it is being priced.

Headline numbers

EPS

Q2 FY2026

$2.13

Revenue

Q2 FY2026

$3.02B

+25.0% YoY

Gross margin

Q2 FY2026

46.1%

Free cash flow

Q2 FY2026

$0.65B

Operating margin

Q2 FY2026

33.8%

Key financials

Q2 FY2026
MetricQ2 FY2026YoYQ1 FY2026QoQ
Revenue$3.02B+25.0%$2.82B+7.1%
EPS$2.13$1.78+19.7%
Gross margin46.1%43.9%+220bps
Operating margin33.8%30.4%+340bps
Free cash flow$0.65B$0.60B+9.0%

Guidance

Guidance is issued for both next quarter and the full year. Both may appear below.

Actuals vs prior guidance

MetricPeriodPrior guideActualΔResult
RevenueQ2 FY2026$2.8B - $3.0B$3.017B+$0.017B above high end of guideBeat
EPS (non-GAAP)Q2 FY2026$1.73 - $2.03$2.13+$0.10 above high end of guideBeat
Non-GAAP Gross MarginQ2 FY202644% - 45%46.1%+1.1 - 2.1 pts above guide rangeBeat
Non-GAAP Operating ExpensesQ2 FY2026$365M - $375Mwithin rangein-lineMet

New guidance

MetricPeriodGuideYoY
Revenue YoY GrowthQ3 FY2026approximately 40% year-over-year at mid-point~40% YoY
RevenueQ3 FY2026$3.1B - $3.3B
EPS (non-GAAP)Q3 FY2026$2.15 - $2.45
Non-GAAP Gross MarginQ3 FY202647% - 48%
Non-GAAP Operating ExpensesQ3 FY2026$380M - $390M
Interest and Other ExpenseQ3 FY2026approximately $50M
Tax RateQ3 FY2026approximately 16%
Diluted Weighted Average SharesQ3 FY2026approximately 385M

Management tone

Q4 FY25: structural demand reset → Q1 FY26: timelines pulled forward → Q2 FY26: customer lock-in formalized through 2028.

Customer commitment horizon has extended again — now to 2028. Q4 FY25 was "all top-five hyperscalers under firm PO/LTA for FY26." Q1 FY26 added "one hyperscaler signed for all of calendar 2027." This quarter management said: "We have firm purchase orders with our top seven customers through calendar year 2026...robust commercial agreements with three of our top five customers, two through calendar year 2027, and one through calendar year 2028." In three quarters the visibility window has doubled. This is no longer demand framing — it is a posture that the supply-demand imbalance is being underwritten by customers, not management talking their book.

The margin algorithm keeps moving up. Q4 FY25 anchor was 38%; Q1 FY26 print was 43.9% with 44–45% guide; this quarter prints 46.1% with 47–48% guide. Per the Wells Fargo Q&A exchange: "75% is achievable" on incremental gross margin. Management is now talking about an incremental flow-through that would push corporate gross margin well above 50% over time. The 38% reference point is effectively retired.

AI-driven demand has shifted from training to inference framing. Last quarter the demand story was a broad AI tailwind. This quarter management was specific: "It is data that is needed to fuel the entire AI process, from training to inference, to enable stronger models and sharper inference results." The BNP exchange reinforced this — exabyte growth at low-20s pace is now attributed to inference workloads generating storage need, not training. Reframing storage as inference-essential (rather than training-discretionary) is the strongest version of the demand argument the company has offered, because inference scales with end-usage rather than with one-off model builds.

HAMR qualification is now active, not pending. Q4 FY25 had HAMR ramp targeted for H1 2027. Q1 FY26 pulled qualification forward to H1 2026. This quarter management confirmed qualification has started this month with one customer, with next-generation ePMR qualification also underway at a different hyperscaler — and 3.5M+ flagship ePMR units already shipped in the quarter with low-90s% yields. The technology roadmap has moved from "ahead of milestones" language to active customer-by-customer execution evidence.

Capital return discipline has shifted from defensive to aggressive. Last quarter the framing was building optionality with a $2B authorization and $0.10 dividend. This quarter $1.4B was returned in a single period — more than 2x the quarter's FCF — alongside a strategic investment in CoLab. The posture is no longer "preserve flexibility"; it is "deploy now while the cycle and the cash flow both support it."

Recurring themes management leaned on this quarter:

AI-driven data storage demand acceleration across training, inference, and physical AILong-term customer lock-in through multi-year commercial agreements securing exabyte commitmentsDensity roadmap execution (HAMR, ePMR, UltraSMR) with active hyperscale qualificationsMargin expansion through product mix upshift to higher-capacity drives and manufacturing efficiencyCloud revenue dominance (89% of total) with 28% YoY growthCapital return discipline with $1.4B deployed while maintaining financial flexibility

Risks management surfaced:

Forward-looking statements subject to various risks and uncertaintiesHAMR and next-generation ePMR qualification completion timing with multiple hyperscale customersCustomer concentration dependency on top seven customers with firm POs through 2026Supply chain execution risk for achieving stated exabyte delivery targetsCompetitive dynamics in higher-capacity drive adoption and market share

Q&A highlights

Aaron Rakers · Wells Fargo

How durable is the 75% incremental gross margin flow-through, and how should we think about the cost curve on a per terabyte basis over the next several quarters?

Management is comfortable with incremental gross margins higher than 50%, noting 75% is achievable. Gross margin improvement driven by two factors: pricing (stable, up 2-3% ASP per terabyte last quarter) and cost reduction (down 10% per terabyte year-over-year). Cost benefits come from upshifting customers to higher capacity drives, manufacturing execution, and supply chain improvements. Expect continued gross margin expansion over next couple of quarters.

46.1% gross margin, up 220 bps QoQ and 770 bps YoYGuidance 47-48% (47.5% midpoint), up 740 bps YoY75% incremental gross marginASP per terabyte up 2-3% last quarter

Eric Woodring · Morgan Stanley

Given HDD market tightness and NAND inflation, is the company using patience in PO timing to extract better economics for 2027, and what are the plans for SanDisk share monetization?

Company is sold out for calendar 2026 with firm POs from top 7 customers and established LTAs with two customers for 2027 and one for 2028. LTAs include volume and price terms reflecting structural value shift to customers, particularly on total cost of ownership benefits from inference workloads. Regarding SanDisk: 7.5M shares remain; company intends to monetize before one-year anniversary (Feb 2027), likely via debt-for-equity swap, with proceeds used to reduce debt.

Sold out for calendar 2026LTAs with 2 customers for 2027, 1 customer for 20287.5M SanDisk shares to be monetized by Feb 2027Debt-for-equity swap planned for SanDisk monetization

Wamsi Mohan · Bank of America

How is UltraSMR mix trending in the near-line portfolio and what role does this mix shift play in driving gross margins forward?

UltraSMR crossed 50% mix in near-line portfolio last quarter and is expected to increase. Top three customers fully adopted UltraSMR; another 2-3 are in adoption process. UltraSMR offers 20% capacity uplift over CMR and 10% over industry standard SMR, benefiting customer demand fulfillment and margins. As software-based solution, UltraSMR is accretive to gross margins. New JBOD launch expands UltraSMR reach beyond current targets.

UltraSMR reached 50% mix of near-line portfolio last quarterTop 3 customers fully on UltraSMR, 2-3 more adopting20% capacity uplift over CMR10% capacity uplift over industry standard SMR

Carl Ackerman · BNP Paribas

Given agentic AI demand expected to drive double-digit growth in conventional servers in 2026, can the company exceed its long-term low-20s exabyte growth cadence?

Management expects exabyte growth to continue at low 20s pace driven by inference workloads rather than training. Inference generates more data requiring storage; hyperscalers manage economics across storage tiers (SSD, HDD, tape). Hard drives well-positioned for inference storage due to superior economics. Customer conversations indicate inference will drive significant data storage requirements positively for HDDs.

Recent exabyte growth in low 20sInference workload shift creates data generationHard drives suited for inference storage economicsHyperscalers managing multi-tier storage economics

Mike Cadiz · Citigroup

Can you provide color on yields and reliability of EPMR products, and implications for cost per bit declines, plus progress from Rochester test site and customer transitions?

EPMR yields in low 90s percentage range with strong customer feedback on reliability/quality evidenced by 3.5M+ flagship units shipped last quarter. Cost declines driven by yield improvements and UltraSMR mix increases within new products. Rochester test and integration lab critical for smooth, quick qualification of new products. Hammer qualification pulled forward to H1 2026; started this month with one customer. Next-generation EPMR qualification also underway. Innovation Day (Feb 3) will showcase updated roadmaps.

EPMR yields in low 90s percentage range3.5M+ EPMR units delivered last quarterHammer qualification started H1 2026 (pulled forward)Next-generation EPMR qualification started

Answers to last quarter's watch list

Whether Q2 gross margin breaches 45%. Yes — 46.1% printed, +110bps above the high end of the 44–45% guide. The pattern of clearing the top of the band by ~100–200bps now holds for three consecutive quarters.
Resolved positively
2027 LTA count. From one hyperscaler signed for 2027 last quarter to "two through calendar year 2027, and one through calendar year 2028" this quarter. The supply-constrained-through-2027 narrative is now underwritten by multiple customers, with one extending into 2028.
Resolved positively
HAMR customer count by end of calendar 2026. HAMR qualification started this month with one customer; next-gen ePMR qualification with a different hyperscale customer also underway. Management did not commit to the "up to three" year-end count from last quarter, but the start point has been hit on schedule.
Continue monitoring
First formal FY26 full-year revenue range. Not provided. Three consecutive quarters without a quantified FY range — the no-FY-guide posture is now the established disclosure pattern, not an interim stance.
Resolved negatively
Capital return execution. $1.4B returned in the quarter via buybacks and dividends — more than 100% of FCF. The $2B authorization is being executed aggressively rather than ratably. No disclosed change to the $0.10 quarterly dividend.
Resolved positively
Pricing commentary. Management put a number on it: ASP per terabyte up 2–3% sequentially last quarter, with management attributing roughly half of margin lift to pricing and half to cost. Pricing is now an explicit lever in the narrative, not an implicit one.
Resolved positively

What to watch into next quarter

Whether Q3 revenue lands above the $3.3B high end. The pattern of beating the top of the range by ~$15–20M holds across the last three prints; the $3.3B threshold becomes a soft floor for what counts as a beat.

Whether gross margin reaches 48%. Q3 guide midpoint is 47.5%; Q2 cleared its high end by 110bps. The Wells Fargo exchange flagged 75% incremental flow-through, which would imply continued upward bias.

Innovation Day disclosures (Feb 3). Watch specifically for any quantified exabyte CAGR reframing, an explicit HAMR ramp customer count, and whether the 38% margin algorithm is formally replaced with a higher anchor.

SanDisk share monetization timing. Management committed to monetizing 7.5M shares before February 2027, with debt-for-equity swap as the likely structure. Execution within the next two quarters would meaningfully reduce gross leverage.

Whether a fourth customer signs an LTA extending into 2027 or 2028. Three of top five are now committed beyond 2026; a fourth would push the share of revenue locked in beyond ~80%.

Capital return cadence. $1.4B in a single quarter is well above run-rate; watch whether Q3 continues at >100% of FCF or normalizes, which would signal a different read on the cycle.

Sources

  1. Western Digital Q2 FY2026 press release, filed 2026-01-29 — https://www.sec.gov/Archives/edgar/data/106040/000162828026004131/a4ex991-pressreleaseq226.htm
  2. Western Digital Q2 FY2026 earnings call prepared remarks and Q&A (as extracted)
  3. Western Digital Q1 FY2026 and Q4 FY2025 prior tapebrief coverage (for trend comparisons)

Get the next brief, free.

We publish analyst-grade earnings briefs the same day or morning after every call — headline numbers, segment KPIs, Q&A highlights, and tone analysis. Free during beta.

This is not investment advice.