By Nick David, Editorial Lead, SatNews
The S-1 reveals a company that is three businesses (Connectivity, Space, and AI), only one of which is profitable today. With $80 billion on the table and an implied $1.75 trillion valuation, here is what every satellite operator needs to read before SPCX prices on June 11.
The S-1 At a Glance · By the Numbers
~$1.75T
Implied valuation
$18.7B
FY2025 revenue
+49.8%
Starlink revenue YoY
<7%
Profitable share of valuation
Executive Brief
SpaceX filed Form S-1 on May 20, registering an offering of up to $80 billion in Class A common stock on Nasdaq and Nasdaq Texas under the ticker SPCX. Pricing is expected June 11, with first trading on June 12. Underlying the headline raise is an implied valuation of roughly $1.75 trillion, which would place SpaceX among the most valuable public companies on Earth at debut. The filing also makes public, for the first time, segment-level economics that reframe the competitive math for every satellite operator: Connectivity revenue of $11.387 billion in 2025, +49.8% year over year, with a 63% Segment Adjusted EBITDA margin. The AI segment lost $6.355 billion in 2025. Musk will hold Founder, CEO, CTO, and Chairman titles, with SpaceX listing as a controlled company under Nasdaq rules.
On Wednesday, May 20, SpaceX filed Form S-1 with the SEC, registering the long-anticipated initial public offering of Space Exploration Technologies Corp. The company will list on Nasdaq and Nasdaq Texas under the ticker SPCX. Pricing is expected June 11; trading begins June 12. The offering targets up to $80 billion in proceeds, which would make it the largest IPO in U.S. history by a substantial margin. Analyst reporting around the filing implies a valuation of roughly $1.75 trillion. If that holds at pricing, it places SpaceX in the company of the five largest public companies on Earth.
That headline is the easy story. The S-1 itself is more interesting. For the first time, SpaceX has had to put numbers next to claims it has been making for years, and the document reveals a company that is three businesses, not one. Only one of those three is profitable today.
The Numbers
The S-1 puts segment-level figures on the record for the first time. Consolidated 2025 revenue: $18.674 billion. Adjusted EBITDA: $6.584 billion. Loss from operations: $2.589 billion. The first quarter of 2026: $4.694 billion in revenue, $1.127 billion in Adjusted EBITDA, an operating loss of $1.943 billion. Capital expenditures across the company totaled approximately $20.7 billion in 2025. Q1 2026 capex alone was approximately $10.1 billion. If anything close to that quarterly pace holds, SpaceX will spend more on capex in 2026 than it generated in revenue in 2025.
The 2025 figures are recast to include xAI (acquired February 2, 2026) and X (absorbed by xAI in March 2025) as common-control transactions. That accounting choice means the historical SpaceX-only revenue line is no longer what the prospectus reports. The headline number is xAI-inclusive; the standalone-SpaceX comparable is roughly $15.5 billion.
The segment breakdown is where the document earns its keep.
Three Businesses, One Profit · FY 2025 Segments
CONNECTIVITY
STARLINK · MOBILE · STARSHIELD
$11.4B revenue
+49.8% YoY · 63% Seg. Adj. EBITDA margin · $4.4B operating income · $4.2B capex
The engine. 61% of total revenue.
SPACE
LAUNCH · DRAGON · STARSHIELD HARDWARE
$4.1B revenue
($657M) operating loss · $653M Seg. Adj. EBITDA · $3.8B capex · $3.0B consumed by Starship R&D
The platform investment.
AI
xAI / X CONSOLIDATED
$3.2B revenue
($6.4B) operating loss · ($1.2B) Seg. Adj. EBITDA · $12.7B capex · $45B Anthropic compute contract through May 2029
The new question mark.
The Architecture
Reading the segment math makes the company’s structure plain. Connectivity is the engine. It produced 61% of 2025 revenue and effectively all of the consolidated EBITDA. Its 49.8% growth rate dwarfs every public satellite operator on the market. Starlink reports approximately 10.3 million subscribers across 164 markets and operates approximately 75% of all maneuverable satellites in orbit, performing more than 1,000 automated collision-avoidance maneuvers per day.
Space, the launch business that made the company famous, is essentially a platform investment now. It lost money on $4.1 billion in revenue in 2025 because Starship R&D alone consumed $3 billion of its operating cost base. Falcon 9 is profitable. Starship is not. The filing acknowledges this directly and names Starship as the single most important execution risk in the entire prospectus. It is the first risk factor listed.
AI is the new question mark. SpaceX is now structurally a major AI compute provider. It has signed Anthropic to a $1.25 billion-per-month compute agreement that runs through May 2029. That is approximately $45 billion in contracted revenue over three years, a single customer engagement nearly four times the size of Starlink’s entire 2025 revenue. It holds an option to acquire Cursor (Anysphere) at an implied $60 billion valuation, with $10 billion in termination and deferred-service fees payable if it walks. Its segment EBITDA in 2025 was negative $1.237 billion. Its capex was nearly $13 billion.
That is not a satellite company line item. That is a separate company being financed by the satellite company.
What This Means for Every Satellite Operator
Every satellite operator now has a problem the wire services won’t lead with. When SPCX prices on June 11, Starlink’s economics become public benchmarks against which every comparable business will be graded, daily, by analysts who don’t care about your spectrum position or your decades of GEO heritage.
An $11.4 billion broadband business growing 49.8% with a 63% EBITDA margin reframes what “good” looks like in this industry. Viasat, itself a useful comparator after the Inmarsat integration, guided FY2026 to low single-digit revenue growth and flattish Adjusted EBITDA. SES, Eutelsat, and Iridium all operate at growth rates an order of magnitude below Starlink’s. As we noted in our April 7 analysis, that disclosure asymmetry was the structural change the IPO would force. The S-1 confirms it.
Public-market scrutiny is now permanent.
For launch customers, the read is more direct. The S-1 makes explicit what SpaceX has been telegraphing since the Starship V3 briefing on May 14: Falcon’s centrality will be retired in favor of Starship-based deployment of V3 satellites at one terabit per second per spacecraft, beginning in the second half of 2026. That is a twenty-fold increase in downlink capacity per launch. With Blue Origin grounded by the FAA since April 30 and no near-term Western alternative for heavy lift, dependence on SpaceX is now disclosed as a financial fact rather than an industry intuition. AST SpaceMobile’s mid-June pivot to Falcon 9 for its next three BlueBirds is what that dependence looks like in practice.
EchoStar’s $42 billion strategic shift, its spectrum sale to SpaceX which received FCC approval on May 12, reads less like a restructuring in retrospect and more like a recognition. And the January waiver granted to SpaceX before the FCC formally rewrote its spectrum-sharing rules now appears in the prospectus as a competitive moat the regulator quietly priced into the offering.
The closest counterweight is Amazon. Project Kuiper has roughly 1,500 satellites in orbit, an enterprise beta running since April, and a publicly stated $20 billion annual revenue target by 2030. Amazon’s consumer service launches later this year. By the time SPCX reports its first full quarter as a public company, the LEO broadband market will have its first real duopoly disclosure.
What the Filing Asks the Market to Accept
The S-1’s structural disclosures are where the optimism encounters resistance. SpaceX will list with a dual-class share structure: Class A common at one vote per share, Class B at ten. Elon Musk will hold a majority of Class B, which alone elects a majority of the board. He will simultaneously hold the titles of Founder, Chief Executive Officer, Chief Technical Officer, and Chairman. SpaceX will be a controlled company under Nasdaq rules and intends to use the exemption: no majority-independent board, no independent compensation committee, no independent nominating committee. Only the audit committee must be fully independent.
The filing’s most candid passage acknowledges that many of SpaceX’s strategic initiatives, including orbital AI compute at scale, AI chip manufacturing, a lunar economy, human augmentation systems, and Moon and Mars transport, “involve significant technical complexity, unproven technologies, or technologies that do not exist or may require significant advancement, and such initiatives may not achieve commercial viability.”
Read against a roughly $1.75 trillion implied valuation, the framing one analyst at PitchBook has offered around the filing is difficult to argue with: the businesses generating actual profit today represent under 7% of the number being sold to investors. That is the gap institutional buyers will price.
Devil’s Advocate
The skeptical read can be overdone. Connectivity’s 49.8% growth is not a single-year anomaly. It is the third consecutive year of compounding expansion against an addressable market that the S-1 sizes at $28.5 trillion (excluding China and Russia). Starshield has not yet been broken out as a separate revenue line, which means the $11.4 billion Connectivity figure understates the U.S. government exposure that almost certainly carries even higher margins. The 27-underwriter syndicate, led by Goldman Sachs with Morgan Stanley, BofA, Citi, and JPMorgan as co-bookrunners, is the broadest in IPO history for a reason: institutional demand for the deal is real, and it will not require Starship to succeed to clear the book. If Starship works, the upside case justifies the valuation. The bull case is not a stretch. It is a bet on execution against a market structure that has, so far, rewarded SpaceX’s execution.
Key Takeaway
The S-1 reveals SpaceX as three businesses, not one. Only Connectivity is profitable today. With less than 7% of the $1.75 trillion implied valuation backed by current profit, June 11 pricing becomes a referendum on Starship execution, AI-segment capital intensity, and a controlled-company governance structure at trillion-dollar scale. For every satellite operator, the disclosure asymmetry that existed before the IPO is now permanent: Starlink’s economics become the public benchmark against which every comparable business is graded.
What to Watch
Pricing on June 11 will resolve some of it. The roadshow will narrow the range. The first day of SPCX trading on June 12 will be a referendum on whether the public market is willing to underwrite Starship execution risk, AI-segment capital intensity, and a controlled-company governance structure at trillion-dollar scale. The S-1 has made the case the company is asking the market to accept. Whether the market accepts it at $1.75 trillion, or at a meaningfully different number, is now a question for an order book, not a balance sheet.
What every satellite operator should read before SPCX prices on June 11: pages 130 through 226.
About the Author
A storyteller at heart, Nick David covers space policy, satellite markets, defense, and the technologies reshaping how humanity operates beyond Earth. With a background in creative direction, brand strategy, and editorial storytelling, he brings a modern lens to complex subjects and a relentless curiosity about what comes next.
