Qualcomm's Biggest 5G Order Ever Landed in the Same Week Its Margin Guide Fell to 20%

The two announcements landed within hours of each other. SpaceX chose Qualcomm's Dragonwing QRU100 radio units, QDU100 distributed units and commercial-grade Layer 1 software for Massive MIMO radios with 64 transmit and 64 receive paths, with nationwide rollout starting in 2027 across tens of thousands of sites. Financial terms were not disclosed. That evening, Qualcomm reported fiscal Q4 revenue of $10.44B, down 7%, handsets down 26% to $5.14B, and a first-quarter guide whose midpoint sat below consensus.

The sell-off was not a verdict on the SpaceX order. Qualcomm fell 7% in extended trading on Nov. 4 and was down more than 11% in premarket trading on Nov. 5, when the stock traded at $155.31 against the prior close of $173.19 - a 40% discount to its 52-week high of $259.92. Half of that move was market-wide: the Nasdaq Composite fell 5.2% and the VIX jumped 11 points to 27.65 the same session. The company-specific damage was roughly the other half, and it was concentrated in margins, not in revenue.

Qualcomm guided QCT operating margin down to 20%-22% for the December quarter, from 24% in Q4 and 29% a year earlier, against a long-term target of 30% that management says is unchanged. Three of the points are mechanical: consolidating the Data Center business into the QCT segment starting in fiscal 2027. The rest is mix. Network radios, baseband and data-center hardware do not carry the margin of a handset application processor or a licensing contract, so the businesses that are supposed to replace the handset line are the ones dragging the segment margin down.

Management still expects non-handset revenue to grow more than 60% year over year in fiscal 2027, with Data Center revenue of $650M in the December quarter on the way to a $5B fiscal 2027 target and $15B or more of scale revenue by fiscal 2029. The revenue story can be true and the margin story can still disappoint at the same time, because Data Center hardware and cellular infrastructure both sit below the 24% QCT margin that handsets and licensing have been subsidizing.

Qualcomm's handset collapse has an upstream cause that is running the other way. CEO Cristiano Amon said the company has a supply problem, not a demand one, and management said higher memory-component prices eat into profits while the supply chain describes a markedly worsening environment. Handset makers passed the cost through in price, and units went elsewhere. Qualcomm buys the phone industry's most inflationary input and sells into a market that had just cut 26% of its revenue.

That asymmetry is why the order matters strategically rather than financially. Qualcomm has spent a decade selling into the world's most price-sensitive hardware market; cellular infrastructure sells to carriers on multi-year build cycles, and a network operator's cost of capital does not have to move with DRAM spot prices. A company whose franchise was defined by the smartphone cycle is being paid to sell into the carrier capital cycle instead.

SpaceX can fund a national network, but it is funding something else first. Third-quarter revenue was $14.58B, up 161% from $5.6B, with net income of $4.40B; the Connectivity segment that contains Starlink was $5.25B, or 36% of company revenue and 85% of operating profit. Capital spending told the rest of the story: SpaceX spends 83 cents of every capital dollar on AI compute, not on the terrestrial network that carries Qualcomm's radios.

Two details argue for patience on timing rather than for the deal itself. Starlink consumer ARPU fell to $69 a month from $83 even as subscribers rose to 13.6M, so the consumer business is maturing even as the company pivots to build a national carrier. And the customer's own chief executives frame the terrestrial build as a multi-year, multi-hundred-billion-dollar program: Verizon's chief executive has said SpaceX would need four to six years and $300B-$500B to fully compete. Qualcomm has told investors revenue begins to contribute as it exits fiscal 2027, with fiscal 2028 the more significant period. The order is therefore a fiscal 2028 revenue event dressed as a November 2026 headline.

The build distributes money in a way that does not require anyone to win the consumer fight. SpaceX has been reported to be in talks for access to at least 80,000 US cell sites, bought up to 14 MHz of paired nationwide 800 MHz spectrum, and is now hiring the radio layer that Qualcomm is supplying. The incumbents have answered in the two places they can: AT&T, Verizon and T-Mobile US have pressed the FCC against SpaceX's spectrum and waiver requests and formed a satellite joint venture, which turns a competitive threat into a second round of network spending.

The near-term and long-term questions have different answer keys, and only one of them is on the calendar this winter. Over days to quarters, the story is a memory bill of materials and a margin floor. Over one to three years, it is a funding decision by a company that is currently spending its capital on something else.

The two Nov. 4 events are usually read as contradictory, but they describe the same company. Qualcomm has won the biggest 5G order in its history, a demand pool that is neither handsets nor data-center accelerators, and in the same breath it told investors that the mix replacing handsets is worth 4 to 6 fewer margin points and that its own handset line fell 26% on costs it cannot pass on. The strategic logic is sound: infrastructure sells on carrier capital cycles, not on phone replacement cycles, and it opens a market where Qualcomm had previously only a handful of reference deployments in Japan and Vietnam.

What the announcement cannot do is pay for the transition. Qualcomm generated $11.15B of operating cash flow in fiscal 2026, spent $2.49B on capital expenditure, and returned $10.69B through buybacks and dividends - more than the $8.65B of free cash flow it produced, which is why cash fell to $4.64B from $7.84B. Meanwhile the customer's trailing free cash flow was negative $39.27B and funded with $102.97B of equity issuance, with the terrestrial network fourth in line behind AI compute. So the honest read is that Qualcomm trades near-term margin and cash for a fiscal 2028 revenue option, priced today by a market that has already marked the shares 40% below their 52-week high.