Computer Peripherals · Outside the Retail 100

Western Digital

WDC

Our rating: buymedium conviction
12-month target: $459.60 +35.7%
Expected vs. S&P 500: +17.4 ptsSee the math
Next earnings: Jan 28In 84 days

Western Digital outsold Seagate last quarter without shipping HAMR drives in volume, so its 2027 switch to HAMR adds capacity growth on top of an existing lead.

Summary

Western Digital is now a pure hard drive maker, and nearly all its sales go to cloud data centers storing ever more data. Demand is running ahead of supply, and customers have committed much of its output in advance; price per terabyte rose 28% from a year earlier and gross margin reached 58%. Yet the stock is about 47% below its June peak as investors question how long the shortage lasts. It depends on prices holding through 2027 as HAMR and Toshiba add supply.

$347.22

+$189.20 (+119.73%)Past year

What Western Digital does

Western Digital and Seagate make nearly 90% of the world's hard drive storage; Toshiba, the only other maker, has just over 10%. Since spinning off Sandisk in 2025, Western Digital sells only hard drives.^3,8,2,1,4

What it sells

Cloud drives89% of revenue, FY2026
High-capacity drives, now up to 40TB, for cloud data centers. Western Digital outsold Seagate last quarter, though Seagate leads on capacity per drive.
Client drives6% of revenue, FY2026
Drives for computer makers and distributors, competing with Seagate and Toshiba. Small, but revenue more than doubled from a year earlier last quarter.
Consumer drives5% of revenue, FY2026
WD-branded external hard drives sold through retailers. A brand business that barely moves the stock.

Who buys it

Mostly the big cloud companies, which the 10-K does not name. The top 10 customers bought 73% of fiscal 2026 sales, up from 55% two years earlier, and three each bought 13% to 16%. A few buyers' budgets drive the business.

Where you've seen it

The WD-branded external hard drives sold at electronics stores and online retailers.

How the drives get made

Western Digital builds its own heads and disks, and ships more terabytes than Seagate at a lower cost each

A cloud company buys hard drives by the terabyte, mostly under multi-year agreements. A high-capacity drive is a sealed box of 11 spinning disks read by tiny magnetic heads. Western Digital designs and makes substantially all of its own heads and disks, then assembles and tests drives in its own plants, mostly in Thailand, Malaysia and the Philippines.

That shows up in volume. Last quarter Western Digital shipped 248 exabytes (an exabyte is a million terabytes), up 22% from a year earlier. Seagate shipped 220 exabytes, up 21%.

By our math, Western Digital took in about $17.70 per terabyte to Seagate's $19.20, yet both kept about 58% gross margin. So Western Digital's cost per terabyte is lower too. These figures blend all drives, and the two companies sell different mixes, so treat the gap as rough.

Management argues makers that don't build their own heads and disks will find it harder to grow output. It uses some outside heads, but says every leading high-capacity drive runs on its own.^{3,1,4,19,18,5,20}

Revenue per terabyte: About $17.70 vs $19.20
Cost per terabyte: About $7.40 vs $8.10
Top 10 customers: 73% of sales

Maker	How it records	Biggest drive now	Last quarter
Western Digital	ePMR, a refined version of the standard method, plus shingled tracks; HAMR from 2027	40TB, 11 disks	248 exabytes, $4.38B revenue
Seagate	HAMR, which heats the disk with a laser to pack bits tighter	44TB, 10 disks	220 exabytes, $4.23B revenue
Toshiba	FC-MAMR, a microwave-assisted method; no HAMR in production	30 to 34TB samples, 11 disks	Not disclosed; about 11% of the market

Where we landWestern Digital's lead is in volume and cost, not recording technology. That holds while its older method keeps pace. Its cost per terabyte needs HAMR to keep falling after 2027, which is why the HAMR ramp is the risk that matters.

How the business makes money

The backstory

Until February 2025 Western Digital sold both hard drives and flash. It spun flash off as Sandisk, then swapped its retained Sandisk shares for debt and its own stock, cutting debt from $4.7B to $1.1B in fiscal 2026. Meanwhile cloud and AI companies are storing far more data, and disk remains far cheaper per terabyte than flash. CEO Irving Tan's plan is to grow output through denser drives, not new factories, and hand most cash back.

Western Digital sells storage by the terabyte. When customers need more terabytes than the industry can make, price per terabyte rises while each denser drive lowers cost per terabyte, and the gap becomes margin.^1,2,3,7,16

1 · DemandData piles upCloud companies keep more data for longer. Western Digital says enterprise SSDs cost about six times more per terabyte than its drives.
2 · Supply limitDensity, not plantsIndustry supply grows slowly. Western Digital and Seagate both add terabytes by packing more data per disk, not by building factories.
3 · Pricing powerLong contractsCustomers commit to volumes years ahead. Some agreements reset prices periodically, and part of output stays uncontracted.
4 · Cost edgeBigger drivesEach denser drive lowers cost per terabyte. The 40TB platform should reach about half of nearline shipments by fiscal Q3.
5 · PayoutBuybacksLittle capex means most cash is free. It spent $1.0B on buybacks last quarter, about three quarters of free cash flow.

Advantages

40TB without HAMR, at Seagate's margins

Major advantage

Western Digital gets more out of older technology. Seagate bet on HAMR, which heats the disk with a laser to pack data tighter, and leads on capacity per drive. Western Digital stretched its existing ePMR method to 40TB and earns the same margin. The edge is temporary: from 2027 both sell HAMR, and it goes to whoever ramps better.

1.Old method, stretched. Its 40TB drive uses ePMR, the recording method customers already run, plus shingling, which overlaps tracks to fit more data. Seagate's 44TB drive needs new HAMR heads and disks.
2.HAMR late, ePMR early. Western Digital's 2025 plan put first HAMR drives in 2026; they slipped to 2027. Its 40TB ePMR drive arrived on schedule and shares HAMR's design, so customers can mix them.
3.Same money, older tech. Last quarter revenue was $4.38B against Seagate's $4.23B, and gross margins were 58.1% and 57.8%. Management says 40TB is being adopted faster than past transitions.
4.Our view: temporary. Seagate already got about 40% of nearline exabytes from HAMR by mid-2026. If Western Digital's HAMR qualifications slip past mid-2027, Seagate's denser drives take share.^1,4,5,6,7,2

Sandisk stake cleared the debt

Strong advantage

Western Digital kept part of Sandisk at the spin-off and swapped all of it for debt and its own shares by July 2026. It now holds about $0.5B more cash than debt, while Seagate still had $1.6B of gross debt after its latest repayments. More of Western Digital's free cash can go to buybacks.^1,3,4

Customers planning to 2031

Supporting

Real, but Seagate makes the same claim, so it supports the first card rather than standing alone. Management says its top seven customers have given visibility into 2030 and 2031, and that HAMR drives are priced per terabyte like ePMR, not discounted.^2,4

What's changing

Converting to 40TB and qualifying HAMR

The 40TB ePMR drive is in volume production, expected at about 30% of nearline shipments this quarter and 50% next. First-generation HAMR should finish qualification at two hyperscalers in early 2027, with two more starting by mid-2027. Capex will run modestly above the 4% to 6% of revenue target for two years.^2,9

Competitors

Who else an investor would consider, and how they compare.

Company	Market value	What it does	Compared with Western Digital
Western Digitalthis stock	$129.97B	Western Digital and Seagate make nearly 90% of the world's hard drive storage; Toshiba, the only other maker, has just over 10%. Since spinning off Sandisk in 2025, Western Digital sells only hard drives.
Seagate TechnologySTX	—	Hard drives only, led by laser-assisted HAMR Mozaic drives of up to 44TB.	Seagate leads on capacity per drive; Western Digital matches its sales with older technology.^4,5
Toshibaprivate	—	Japan's hard drive maker, spending about $380M to double data center drive output.	Toshiba is adding plants to gain share; Western Digital adds terabytes through denser drives.⁸
SandiskSNDK	$181.49B	Western Digital's former flash unit: memory chips and SSDs, increasingly for data centers.	Sandisk sells fast, costly flash; Western Digital sells cheaper disk for stored data.¹⁷

Bull vs. bear: two bets on the same stock

1
A share is a small slice of the company. Its value comes from the profits the company will make in the future.
2
Nobody knows those profits yet. Buyers (bulls) pay today's price betting they'll beat what most people expect. Sellers (bears) bet they'll fall short.
3
When results come in, the price moves toward whichever side was right, and that side makes money.

The question they disagree on

Will Western Digital keep its gross margin above 55% through 2027, as its own HAMR drives and Toshiba's new lines add supply?^{1,2,3,5,7,8,11}

Bull case: why buyers buy

If supply keeps growing only as fast as drives get denser, prices hold, margins stay near 60%, and a stock at about 10 times fiscal 2028 estimates rerates.

Supply grows only with density. Management says extra capex buys heads, media and automation, not more drive-making capacity.
Customers are planning years out. Management says its top seven customers have given demand visibility into 2030 and 2031.
Estimates rise as the stock falls. Fiscal 2028 consensus EPS (earnings per share) rose from $29.34 to $40.40 in 90 days.
HAMR is still ahead of it. First HAMR drives finish qualification at two hyperscalers in early 2027, adding capacity per drive beyond today's 40TB.

Bear case: why sellers sell

If AI spending pauses or Toshiba's new capacity lands, disk prices turn, and today's margins prove a cyclical peak, as they did before fiscal 2024's losses.

Hard drives have crashed before. In fiscal 2024 this business lost money on $6.3B of revenue.
A third maker is adding capacity. Toshiba plans to double data center drive output by March 2028 and wants about 30% share.
Seagate leads on capacity per drive. It has shipped 44TB HAMR drives since March 2026; Western Digital's HAMR slipped from 2026 to 2027.
Few customers carry the business. Ten customers bought 73% of sales, so a pause in AI data center spending hits fast.

Where Delphi landsbuyWe think the bull case is more likely than today's price assumes.See why

The HAMR catch-up

Western Digital's HAMR arrives about a year after Seagate's, and so far the catch-up costs capex, not margin

HAMR is the longer road to bigger drives. Western Digital's own roadmap stretches its current method to 60TB by 2028, while HAMR runs to 100TB by 2029. Its 2025 plan put first HAMR drives in 2026. Now two hyperscalers should finish qualifying them in early 2027, with volume shipments in the first half and two more customers qualifying by mid-2027.

Seagate is well ahead. It expects HAMR to be about half its nearline (data center) terabytes by the end of 2026.

The catch-up shows up in spending. Western Digital expects capex (spending on equipment) to run modestly above its 4% to 6% of revenue target for two years, partly because HAMR heads and disks take more process steps. Seagate expects to stay inside the same range this fiscal year.

What isn't disclosed: HAMR yields, cost per terabyte, or a target share of shipments. Management says HAMR sells at the same price per terabyte as older drives, so any problem would hit cost, not price.^{18,2,5,6,7,19}

First HAMR qualifications: Early 2027
40TB older-method drive: About 50% of bits
Seagate HAMR share: About 50%

Where we landThe delay matters less than it looks: the 40TB drive carries growth through mid-2027. It becomes the risk that matters if qualifications slip past mid-2027 or capex stays high, because then Seagate's cost per terabyte falls faster.

Where the market is wrong

Toshiba's doubling adds about 4 to 5 points a year to supply; the leaders' own growth matters more

Toshiba plans to nearly double data center drive output by its fiscal 2027, which ends March 2028, spending about $380M in the Philippines. It has about 11% of the market and wants 30%.

Our estimate: doubling an 11% share over about two years adds roughly 4 to 5 points a year to industry supply growth. Western Digital and Seagate are each growing terabytes shipped about 21% to 22%, and both say they grow through denser drives, not more drive-making capacity. Together that is roughly 23% to 24% supply growth.

Western Digital puts demand growth at 25% or more a year, its own claim. On those numbers supply roughly catches up, which ends price rises without forcing cuts. The wild card is Western Digital itself: its CFO says new drives will speed up its terabyte growth over the next six to eight quarters.

Contracts cushion the turn. Western Digital's agreements set prices in advance or by formula, with base and upside volumes. Coverage thins further out, so a turn would show first in renewals and spot prices.^{8,20,1,19,18,3}

Toshiba share: About 11%
Toshiba's added supply: About 4 to 5 points
Duopoly terabyte growth: About 21% to 22%

Where we landWe think the market overweights Toshiba. The number to watch is the duopoly's combined terabyte growth: if it runs well above 25% while demand holds there, prices likely turn by 2028 whatever Toshiba does.

What to watch

The numbers that will settle the debate.

Non-GAAP gross margin▲ improving

58.1% in fiscal Q1 2027, up from 43.9% a year earlier; 60–61% guided for Q2¹

Shows whether prices still outrun costs. Below 55% would signal the cycle turning.

Price per terabyte vs a year earlier▲ improving

Up 28% in fiscal Q1 2027, from high-teens growth the quarter before²

The cleanest read on the shortage. Slowing growth comes before falling margins.

HAMR qualification■ unclear

Two hyperscalers, completion expected calendar Q1 2027; volume shipments first half of 2027²

Western Digital's next capacity step. A slip widens Seagate's lead per drive.

Next earnings report

Jan 28

In 84 days

Expected earnings per share: $5.82
Expected revenue: —

What to listen for: Whether gross margin reaches the 60–61% guided, and whether the first HAMR qualifications are still on track for early 2027.^1,2,11

What would prove this wrong

✕Price per terabyte stops rising from a year earlier.
✕HAMR qualification slips past mid-2027.
✕Gross margin guidance falls below 55%.
✕A top cloud customer cuts storage orders or delays agreements.

Our ratingBuyMedium convictionas of Nov 2 close, $338.59 · what ratings mean

At about 10 times fiscal 2028 consensus, the price already assumes a downturn; the model gains about 24% if margins hold and loses about 16% if prices turn.

12-month target: $459.60 (+35.7%)
S&P 500, same method, with dividends: +18.5%
Expected vs the market: +17.4 points

The math behind the recommendation

Three 12-month price scenarios, weighted by probability, compared with the S&P 500 under the same method.

Now $338.59

Target $459.60

bull case25% likely

$659.81 +95%

Prices keep rising through 2027, HAMR ramps on time, and fiscal 2028 earnings land near the top of forecasts as cycle fears fade.

$52.75 earnings per share × 12.5 = $659.81

base case50% likely

$437.78 +29%

Margins hold near 60%, HAMR ramps in 2027, and earnings match consensus of about $40 a share for fiscal 2028.

$43.74 earnings per share × 10 = $437.78

bear case25% likely

$303.04 -10%

Toshiba's capacity and slower AI spending flatten prices in 2027, and fiscal 2028 earnings fall toward the low end of forecasts.

$36.65 earnings per share × 8.3 = $303.04

Why these odds: Analysts raised fiscal 2028 estimates 38% in 90 days and guidance beat by about 20%. Against that, Toshiba's added capacity and AI spending worries make a downturn a real risk, so bull and bear odds are equal.

How the prices are built: Wall Street expects earnings per share of $43.74 for the year starting a year from now (analysts range from $29.56 to $61.76). Today the stock trades at 11.8 times next year's earnings per share; its five-year middle is 12.4. Because earnings per share is forecast to grow fast, valuations are cut 16%, since fast growth rarely keeps its price.^{10,11,12,13,14}

What tests the call: FQ2 2027 earnings: gross margin against the 60–61% guide, and HAMR qualification status (Jan 28)

Buy means expected to beat the S&P 500 by 10+ points over 12 months; Sell means expected to trail it by 10+. Inputs gathered Oct 29; the recommendation updates with the price every day. Model-generated, not personalized investment advice. How ratings work · Track record

What the price assumes

At $338.59, investors pay less for each dollar of Western Digital's earnings than for the average stock's

Western Digital trades at 12 times the earnings analysts expect over the next 12 months. That ratio is the multiple: what investors pay today for each dollar of a year's earnings. That is below its five-year middle of 12.

Analysts expect earnings to grow about 53% the year after, against about 16% for the S&P 500. The S&P 500 trades at 19 times. At the market's multiple, today's price would only make sense if Western Digital earned about 37% less than forecast over the next year.

Western Digital multiple: 12×
S&P 500 multiple: 19×
Base case: +29%
Bear case: −10%

The skewOur bear case loses about 36¢ for every $1 our base case makes. That ratio, how much you stand to lose against how much you stand to make, is the skew.

What a price cut costs

Prices would have to fall about 12% to push margin below 55%, after rising 28% in a year

Gross margin is the share of each sales dollar left after making the product. Western Digital guides to 60% to 61% this quarter on about $5.1B of revenue.

Price changes fall almost straight to profit, because a drive costs the same to build whatever it sells for. On $5.1B of quarterly sales, each 1% of price is about $51M of gross profit. After the 17% guided tax rate and 380M shares, that is about 11 cents of earnings per share a quarter.

Holding costs fixed, prices would have to drop about 12% to take margin from 60.5% to 55%, the line in our crux question. That would give back less than half of the past year's 28% rise. Costs also keep falling: management expects cost per terabyte to decline, if by less than 10% a year for now.^1,18,3

One point of price: About $51M a quarter
Per share: About $0.11 a quarter
Price cut to reach 55%: About 12%

Where we landThe 55% line is safe for the next few quarters, since prices are still rising. The test comes as agreements renew for 2028. Watch price per terabyte: growth will slow before margin falls, and if it stops rising from a year earlier, our case breaks.

Last earnings

FQ1 2027 results, reported Oct 22

Revenue and earnings beat, and Western Digital guided next quarter to $5.1B of revenue, well above the $4.57B analysts expected.^1,2,9,10,8

Reported Oct 22	Expected	Actual	Difference
Earnings per share	$4.06	$4.64	+14.3%
Revenue	—	—	—

Beat expectations in 7 of the last 7 quarters

Stock reaction: Down 3.0% next day, closing at $343.67

Why the stock moved

Positive: guidance for earnings per share was about 20% above consensus, and the dividend rose 17%.
Weighing: capex will run above the 4% to 6% of revenue target for two years.
Weighing: management flagged rising DRAM and flash costs, and Evercore questioned how durable price increases are.
Context: the stock fell 17% from Sept. 30 to the day before results, amid AI selling and Toshiba's plan.

What it means for the stock: Strengthened the earnings case: pricing accelerated and guidance jumped. It did not settle the cycle question, which is why the stock fell.