SpaceX became a publicly traded company on June 12, 2026. Ticker: SPCX. IPO price: $135. First-day close: $160.95. Valuation: $1.97 trillion. It is, by a wide margin, the largest IPO in history — three times the size of Alibaba’s 2014 listing.
And it lost $4.9 billion last year.
That’s not a typo, and it’s not a reason to dismiss the company. But it is the central fact that any investor needs to sit with before clicking “buy.”
The Revenue Ladder
SpaceX reported $18.7 billion in total revenue for 2025 — a real and substantial number. To put it in context, here’s where SPCX sits among a set of well-known names, sorted by 2025 annual revenue, with net income included to show who’s actually making money:
| Ticker | Company | 2025 Revenue | Net Income |
|---|---|---|---|
| GOOGL | Alphabet | $350B | ~$100B |
| NVDA | NVIDIA | $130.5B | $72.9B (56% margin) |
| LLY | Eli Lilly | $45.2B | ~$10.6B |
| AMAT | Applied Materials | $27.2B | ~$7.2B |
| SPCX | SpaceX | $18.7B | −$4.9B net loss |
| GLW | Corning | $14.1B | ~$1.5B |
| CLS | Celestica | $12.4B | ~$560M |
| ANET | Arista Networks | $7.6B | ~$2.6B (34% margin) |
| MRVL | Marvell Technology | $5.8B | Improving rapidly |
| COHR | Coherent Corp | $5.8B | Limited profitability |
| ALAB | Astera Labs | $852M | $219M (25.7% GAAP margin) |
The company with the smallest revenue on that list — Astera Labs — is profitable. SpaceX, with $18.7 billion in revenue, is not.
The Valuation That Requires Explanation
Revenue alone doesn’t tell you what you’re paying. The multiple does. Here’s what each name costs per dollar of revenue:
| Ticker | Price-to-Sales | Profitability Context |
|---|---|---|
| SPCX | ~94x | −$4.9B net loss |
| NVDA | ~26x | $72.9B net income |
| ANET | ~14x | 34% net margins, clean balance sheet |
| ALAB | ~21x | Profitable at its scale |
| CLS | ~1x | Growing 28% YoY, profitable |
NVIDIA — the company most responsible for the AI infrastructure boom — trades at 26x revenue while producing the most profitable year in semiconductor history. SpaceX trades at 94x revenue while losing nearly $5 billion. The market is not pricing SpaceX on what it is today. It’s pricing what Starlink could become.
Two Companies Inside One Stock
The S-1 revealed something important that got buried in the IPO excitement: the xAI segment — Grok, Elon Musk’s AI product — lost $6.4 billion in 2025. That loss is consolidated into the entity you’re buying.
So when you buy SPCX, you’re not just buying Starlink and rockets. You’re also absorbing a $6.4B/year AI venture that is currently competing against OpenAI, Google DeepMind, and Anthropic with no clear differentiation — and losing that bet at a staggering rate.
Strip out the xAI losses and Starlink is a genuinely exceptional business: $11.4 billion in revenue, $4.4 billion in operating profit, 48% year-over-year growth. That’s a world-class internet infrastructure company. The problem is the price you pay to access it also includes everything else attached to it.
Starlink standalone (estimated): $11.4B revenue · $4.4B operating profit · 48% YoY growth
xAI segment: −$6.4B loss in 2025 — consolidated into what you’re buying
What the Valuation Requires You to Believe
At $160 per share, the market is pricing SPCX at roughly $1.97 trillion. For that to be a fair price — not a great price, just fair — you’d need to believe all of the following:
- Starlink continues scaling globally and becomes the default internet provider for underserved markets, enterprise mobility, and maritime/aviation segments
- xAI losses narrow meaningfully over the next 2–3 years and the product finds product-market fit
- Starship becomes commercially operational and creates an entirely new revenue category
- Musk stays focused on SpaceX through all of the above
- No major regulatory event disrupts the government contract pipeline (currently a significant revenue base)
All five of those things might happen. But you’re not being paid to take that risk at 94x revenue. You’re just being charged for the privilege of believing in them.
The Comparison That Should Stick With You
Celestica (CLS) manufactures AI data center hardware. It’s growing revenue 28% year-over-year. It’s profitable. It has $12.4 billion in revenue and trades at roughly 1x sales.
SpaceX has $18.7 billion in revenue, loses $4.9 billion, and trades at 94x sales.
The market has decided that SpaceX’s future is worth 94 times more per revenue dollar than Celestica’s present. That might be correct. But it’s a hell of a bet to make at the open.
The Smart Play
The long-term thesis on SpaceX is real. Starlink is a genuinely defensible asset. The rocketry monopoly — 82% of commercial launch market share — is real infrastructure. If the vision plays out across 10 years, the current price may look cheap in retrospect.
But IPO day is not the time to find out. The six-month lock-up on insider shares expires in January 2027. That’s when early investors and employees can sell — and that’s historically when IPO-era enthusiasm meets real price discovery. The stock found its first-day close at $160.95. It may run further on momentum. But the patient investor watches January 2027, when the lock-up expires and the real market makes its verdict.
A patient entry in the $90–110 range — if it ever gets there — would give you the same long-term exposure with a fraction of the valuation risk. That’s not guaranteed to happen. But it’s worth waiting to find out.
The company is extraordinary. The price is not.