Space Exploration Technologies Corp (SpaceX)
SPCX · NASDAQTelecommunications · Space, Satellite Internet and Compute Infrastructure
Since the Report + 29.9%
Close on Report DayAug 4
$114.92▼ 13.6%Report-Day Move
- Market Cap(Today)
- ≈ $1.96 T
In its first quarter as a public company SpaceX grew revenue 92% year over year to $7.81 billion and cleared every line of the Market Expectation, with AI revenue alone up 247%. The stock fell 13.6% the next day because the same quarter carried $18.4 billion of capital spending and management signaled a similar pace for the next two quarters. From here the question is whether that spending converts into the $100 billion annualized revenue run-rate and 2 gigawatts of compute targeted for December.
Revenue (Q2)
$7.81B
▲ 92% YoY
Adjusted EBITDA
$3.54B
▲ 191% YoY
Loss Per Share
-$0.09
Market Expectation -$0.26
Capital Expenditure
$18.37B
~39% Above Expectation
Starlink Subscribers
12.0 million
+1.7 Million in the Quarter
Cash & Securities
$100B
Backlog $47.5B
Quarterly Revenue ($ Billion)
4.07
4.69
7.81
- Q2 25
- Q1 26
- Q2 26
- Annual Growth
- ▲ 92%$4.07B → $7.81B
- Connectivity (Starlink) Share
- 55%$4.29B · ▲ 66% YoY
- Space (Launch) · Slowest
- $962M · ▲ 29%Segment EBITDA −$205M
Q3 2026 Company Guidance
- Q3-Q4 26 Quarterly Capital Expenditure18.4 billion
Similar to the Q2 levelFlat Trend
- December 2026 Annualized Revenue Run-Rate100 billion
Q2 annualized $31.3BAggressive Target ▲
- Year-End 2026 Compute Capacity2 GW
15-20 GW targeted by end-2027Rapid Buildout ▲
- 2030 Annual Revenue Target1 T$
Previously targeted for 2031Pulled Forward ▲
- Capital Expenditure (Q2)
- $18.37BAI $15.83B
- Contracted Backlog
- $47.5BAI Cloud $14.1B
- Weighted Average Shares
- 4.9 billionH1 26 · Diluted
“The combination of our global launch leadership, recurring Starlink revenue, and rapid growth in AI infrastructure continues to drive improved operating leverage.”
- Starship heat shield solved
- Starlink V3 capacity leap
- AI compute capacity buildout
Summary
SpaceX reported its first quarter as a public company after the close on August 4, roughly seven weeks after its June 12 IPO. Revenue came in at $7.81 billion against $4.1 billion a year earlier, a 92% increase. The Market Expectation ranged from $6.8 billion to $7.1 billion depending on the source; on any of those numbers the quarter was a beat. Adjusted EBITDA of $3.54 billion rose 191% year over year and comfortably cleared an implied expectation near $2.5 billion. Loss per share was $0.09 versus an expected loss of $0.26, and net loss narrowed to $541 million from $1.01 billion. All three segments grew: Connectivity $4.29 billion (up 66%), AI $2.56 billion (up 247%) and Space $962 million (up 29%). Starlink ended the quarter with 12.0 million subscribers, exactly double a year ago, after adding 1.7 million net subscribers in the quarter.
The stock still fell from $125.33 to $108.27 the next session, a drop of 13.6%; the initial after-hours reaction was about 8.6%. The trigger was not the revenue line but the spending line. Capital expenditures reached $18.37 billion in a single quarter against a widely held expectation near $13 billion, roughly 39% above it, and $15.83 billion of that went into AI and compute infrastructure. Management expects a similar pace over the next two quarters. The result is first-half free cash flow of roughly negative $25 billion. Every line of the income statement improved while the cash statement moved sharply the other way, and the market priced the second one. The calendar added to it: the first tranche of the lock-up released on August 6, making 911.5 million shares eligible to trade.
Our read on the quarter itself is hold. Revenue growth, margin improvement and subscriber gains are not in dispute, and a $47.5 billion backlog plus $100 billion in cash and marketable securities buys the company a long runway. But capital intensity sits at the center of this story and SpaceX is not yet generating cash. The launch business still loses money at the operating level, Starlink's monthly revenue per subscriber fell from $85 to $66 on geographic mix, and the company issued no formal guidance. None of that makes the quarter bad, but none of it earns a score above 80 either. The size of the share price drop does not change the grade: most of that selling was about how many shares became tradable, not about the company. More: What Is an IPO?
Full Review
ClaudeSatellite internet is now the revenue engine.
The Connectivity segment — Starlink — produced $4.29 billion in the quarter, 55% of total revenue on its own, up 66% year over year. Subscribers reached 12.0 million from 6.0 million a year earlier, with 1.7 million net additions in the quarter, the best subscriber quarter in the company's history. On the call, President and COO Gwynne Shotwell pointed to the enterprise side: enterprise and government revenue grew 108% year over year, and she noted the company is only 10% penetrated in aviation. Segment adjusted EBITDA was $2.60 billion, a 60.5% margin, making this the most profitable arm of the business. The one caveat is revenue per subscriber: blended ARPU is $66 a month against $85 a year ago. That is not discounting but mix, as growth increasingly comes from lower-priced markets. It is also why subscribers doubled while segment revenue rose only 66%. How that gap behaves over the next few quarters will define the segment's real growth rate.
The surprise was in spending, not in revenue.
Quarterly capital expenditure was $18.37 billion against a widely held expectation near $13 billion, about 39% higher. For scale: total revenue in the same quarter was $7.81 billion, so the company spent more than twice what it earned. First-half capex reached $28.48 billion versus $6.97 billion in the prior-year period. With $3.47 billion of cash from operations, first-half free cash flow lands near negative $25 billion. CFO Bret Johnsen said on the call that the pace should stay similar for the next two quarters. The company can fund it: the June 12 IPO raised roughly $85.7 billion, a June 26 bond issue added $25 billion, and the balance sheet closed the quarter with $100 billion in cash and marketable securities against $36.84 billion of long-term debt. The question is not solvency; it is how long that capital takes to come back. More: What Is Cash Flow?
The AI arm grows fastest and costs the most.
The AI and cloud segment grew 247% year over year to $2.56 billion in revenue and delivered $1.15 billion of adjusted EBITDA. It also absorbed $15.83 billion of the quarter's $18.37 billion of capital spending. SpaceX exited the quarter with 1.4 gigawatts of compute online, and Johnsen said the target is more than 2 gigawatts by year end and roughly 15 gigawatts of power and cooling capacity by the end of 2027. Contracted cloud services sales stand at $14.1 billion, and the roughly $60 billion Cursor acquisition is expected to close in the third quarter. The arithmetic is simple: the segment looks profitable, but it is consuming fourteen times its own EBITDA in capital. Management argues that payback on new compute investment has fallen below one year. That claim cannot yet be verified from the outside, and it is precisely what the market questioned on August 5.
Launch still loses money, Starship nears the threshold.
The Space segment grew 29% year over year to $962 million but posted adjusted EBITDA of negative $205 million. The company's founding business is now the smallest of the three arms and the only one losing money. Operationally the picture is brighter: 78 launches in the first six months delivered 1,041 metric tons to orbit. Starshield contracts total more than $6 billion, and total backlog is $47.5 billion. On the call, Elon Musk said of Starship Flight 13's heat shield performance: "I don't want to jinx it or anything, but I think I'd consider the heat shield problem solved at this point." Flight 14 is tentatively set for late August; pending regulatory approval it will attempt the first tower catch of the upper stage and deploy operational V3 Starlink satellites. Musk described the V3 satellite as roughly an order of magnitude more capable than V2. For Starlink's cost curve, that matters more than this quarter's segment revenue.
Part of the sell-off had nothing to do with the quarter.
The stock had rallied 9.4% into the print, closing at $125.33 on earnings day, then fell to $108.27 the next session. Other space names had an ordinary day; what was being sold was one company's spending plan, not the sector. A second pressure sat on top of it: the first lock-up tranche released on August 6, making 911.5 million shares eligible to trade — more than one and a half times the 639 million shares floated at the IPO. The market had been positioning for weeks, with short interest rising from about 40 million shares on June 23 to 219.3 million on July 29. The interesting part: on August 6, the day the lock-up opened, the stock closed up 6.1% at $114.92. The expected supply wave did not arrive on day one. The shares still trade well below the $135 IPO price and far below the June intraday high of $225.64. That does not change the grade on the quarter; it is a reminder of how thin the ground under the price still is. More: Market Cap, Float and Splits
Strengths
6- All three segments grew and total revenue rose 92% year over year to $7.81 billion, above every version of the Market Expectation.
- Adjusted EBITDA rose 191% to $3.54 billion and operating loss narrowed from $970 million to $143 million; the operating leverage is real.
- Starlink subscribers doubled in a year to 12.0 million, with 1.7 million net additions in the quarter — the company's best subscriber quarter yet.
- The balance sheet holds $100 billion in cash and marketable securities after the IPO and the June bond issue, funding a long investment cycle.
- A $47.5 billion backlog, $14.1 billion of contracted cloud sales and more than $6 billion of Starshield contracts improve revenue visibility.
- Starship's heat shield problem appears solved, and Flight 14 is set to deploy operational V3 Starlink satellites.
Risks
6- Quarterly capex of $18.37 billion ran about 39% above expectations and more than twice quarterly revenue, and management expects a similar pace for two more quarters.
- First-half free cash flow is roughly negative $25 billion; the income statement is improving while cash burn accelerates.
- The lock-up unwinds in tranches: 911.5 million shares on August 6, a roughly 28% tranche tied to the Q3 report, and the remaining balance on December 8. Supply pressure is a story that runs for months.
- Starlink's monthly revenue per subscriber fell from $85 to $66, which is why revenue grew only 66% while the subscriber base doubled.
- The Space segment posted $205 million of negative adjusted EBITDA; the company's founding business is still losing money.
- No formal guidance was issued; the $100 billion annualized run-rate and sub-one-year payback claims remain verbal commitments. The roughly $60 billion Cursor acquisition has also not yet closed.
What to Watch
6- Late August 2026: Starship Flight 14. Pending regulatory approval it will attempt the first tower catch of the upper stage and deploy operational V3 Starlink satellites.
- During Q3 2026: the roughly $60 billion Cursor acquisition is expected to close.
- September – October 2026: five time-based lock-up tranches release at 70, 90, 105, 120 and 135 days after the IPO, each roughly 7% of locked shares.
- ~November 2026: the Q3 report. That same report triggers a roughly 28% lock-up tranche — two events on one date.
- December 2026: the company targets more than 2 gigawatts of compute capacity and a $100 billion annualized revenue run-rate by year end. The remaining lock-up balance releases on December 8.
- June 12, 2027: Elon Musk's roughly 6.4 billion shares — 42% of the equity and 85% of the voting power — unlock, with no early-release condition.
Upcoming Earnings
To Understand This