The SpaceX IPO, taken apart: how to read a stock-market listing

On 12 June 2026, SpaceX is set to take its first steps on the public market, listing on the Nasdaq under the ticker SPCX. The deal targets roughly $74 billion raised, for an implied valuation near $1,770 billion — one of the largest listings in history.¹

This article won’t tell you whether to buy: that’s neither my role nor what this blog is for. It does something more durable — it hands you the reading grid for a stock-market listing, the reflexes that apply to any IPO. And the SpaceX file is an almost perfect case study to illustrate them: a primary offering, profitability that looks better than it is, a vertiginous valuation, locked-down control, and a wave of restricted shares. We’ll walk through eight reflexes, with figures from the Prospectus.

Reflex 1 — Who gets the money? Primary or secondary

The very first question when facing an IPO: does the money go to the company or to exiting shareholders?

SpaceX is running a Primary offering: the offer covers only newly issued shares (up to 555.5 million Class A shares, plus an 83.3 million over-allotment option). No existing shareholder sells a single share. The net proceeds — around $74.4B at the expected $135 price — therefore go to SpaceX to fund its growth.

That’s the opposite of a secondary offering, where early backers use the IPO to cash out. A good sign here: those who know the company best are still funding it rather than leaving. One caveat, though — part of the proceeds will refinance existing debt (reflex 5).

One note for European retail investors: a tranche of 55.5 million shares is reserved for European retail (price cap of $162, commitment given before pricing). Platforms such as Trade Republic or Revolut offer access to it. Taking part in an IPO is no small thing, though — all the more reason to run through the reflexes that follow before committing.

Reflex 2 — What does the company actually do?

Before any number, understand the business. SpaceX isn’t one company but three businesses on opposite trajectories:

• Space — launch services (Falcon 9, Falcon Heavy) and Starship development. 2,213 tonnes to orbit in 2025, 165 Falcon launches.
• Connectivity (Starlink) — a network of about 9,600 satellites. Subscribers rose from 8.9 million at end-2025 to 10.3 million at 31 March 2026, across 164 countries. But ARPU (average monthly revenue per subscriber) is falling, from $81 to $66 quarter on quarter — the effect of expanding into cheaper markets.
• AI — the COLOSSUS compute platform, 1.0 gigawatt installed at 31 March 2026, plus the EchoStar spectrum acquisition for $19.6B (expected to close end-2027).

Keep this structure in mind: almost everything that follows stems from the fact that these three businesses don’t play in the same league.

Reflex 3 — Strong growth isn’t profitability

Revenue is impressive: roughly +33% per year ($10.4B → $14.0B → $18.7B from 2023 to 2025). And yet 2025 swings to a net loss of $4.9B, after a single profitable year (2024). Q1 2026 already shows −$4.3B.

The reflex: where does the loss come from? Here, from R&D, which jumps +150% to $8.6B in 2025 (Starship and AI), and more than doubles in Q1 2026. That’s deliberate growth spending expensed through the income statement — not a collapse of the core business. An “offensive” loss doesn’t read like a “defensive” one.

Reflex 4 — The adjusted-EBITDA trap

Here’s the single most important point in the whole analysis. Companies love to highlight their adjusted EBITDA, which flatters better than net income. The right reflex: drop down to operating income by segment, the real arbiter.

Read at adjusted EBITDA, AI looks “nearly break-even” (−$1.2B). At operating income, it actually burned $6.4B in 2025. The $5.1B gap is mostly depreciation ($3.6B) and stock-based compensation ($1.1B): very real costs, the flip side of massive investment. Adjusted EBITDA isn’t a lie — but it erases precisely what’s most expensive in a capital-intensive business.

And the corollary: Starlink funds everything else. Connectivity throws off $4.4B of operating income, which subsidizes Space (back in the red because of Starship) and AI (deeply loss-making). The group’s profitability rests on a single segment.

Reflex 5 — Follow the cash, not just the earnings

Accounting earnings don’t tell you whether the company is burning cash. SpaceX’s Capex (capital expenditure) is exploding, especially in AI: from $0.5B in 2023 to $12.7B in 2025, then $7.7B in Q1 2026 alone. Total 2025 capex: $20.7B.

The result: Free cash flow is heavily negative. Operating cash ($6.8B in 2025) doesn’t cover investment (−$19.6B); the gap is filled by debt and equity (+$26.4B of financing in 2025). Also worth watching: a $20B bridge loan (March 2026) used to repay xAI/X debt, to be refinanced within six months of the IPO. So part of the money raised settles the past as much as it funds the future.

Reflex 6 — What does it really cost?

At the expected $135 price and about 13.1 billion shares after the offer, the implied Market capitalization is around $1,770B — roughly 95 times 2025 revenue. At that level, you’re not paying for earnings: you’re paying for near-flawless execution over ten years.

The companion reflex: measure Dilution. Pro forma net tangible book value comes to $3.32 per share, while the IPO price is $135. In other words, whoever buys at $135 pays 93.4% above the immediate net asset value. That’s common for a growth stock — but better to see it plainly than to absorb it blindly.

Reflex 7 — Who’s in control, and the lock-up overhang

Two governance questions often decide a minority shareholder’s fate: who’s in charge, and when can insiders sell?

On control, SpaceX rests on a Dual-class share structure (Class B shares with 10 votes). Elon Musk alone holds ~84% of the voting rights, and even after the IPO the public will weigh only 11.5% of the votes. Buying the stock, here, gives you no say over strategic decisions.

On liquidity, the Lock-up is the calendar to know. Since no existing shareholder sells at the IPO, their shares only become sellable when their lock-up expires. There are three distinct regimes:

• Founder (Elon Musk) — ~6.1 billion shares locked for 366 days, with no early release: full liquidity only around 12 June 2027.
• Extended lock-up (large holders other than the founder) — ~1.7 billion shares, released in tranches from Q1 2027.
• “180-day” block (everything else) — ~4.7 billion shares, the earliest, releasing from summer 2026.

It’s this 180-day block — the first that can weigh on the market — that the table below details, tranche by tranche:

From the first tranche, nearly 0.9 billion shares can hit the market. Such a wave of supply, if demand doesn’t keep up, can weigh on the price — regardless of the company’s quality.

Mind the reading trap: that 100% in early December covers only the 180-day block, not all of SpaceX. Most of the capital — the founder’s ~6.1B shares and the ~1.7B of extended lock-up — stays restricted and only releases through 2027. For the record:

• Extended lock-up: 20% after Q4 2026 earnings, +10% on 18 March 2027, +20% after Q1 2027 earnings, +10% on 17 May 2027, +20% on 12 June 2027, remainder after Q2 2027 earnings.
• Founder: the entire stake at once, around 12 June 2027.

In other words, the potential selling pressure doesn’t stop in December 2026: it extends, in waves, throughout 2027.

Reflex 8 — Read the risk factors without dozing off

A prospectus’s “risk factors” section is long and dry, but it’s where the company states, in black and white, what can go wrong. For SpaceX, the essentials:

• Reliance on Starlink — a single segment carries profitability; any slowdown (competition, falling ARPU) weakens the whole.
• Starship execution — the entire strategy (next-gen satellites, in-orbit compute) depends on Starship’s ramp-up and reliability. The topic is live: the FAA grounded flights after the Flight 12 test mishap (22 May 2026), with an investigation required before the next flight.² Conversely, the FAA approved two Starfall reentry test flights in late May 2026³ — a positive signal, but limited to testing, not commercial operations.
• Regulation — launch and reentry licenses (FAA), spectrum (FCC); a shifting AI framework.
• No dividend — the company plans to reinvest everything.
• Volatile lines — erratic “other expenses” and $1.6B of digital assets (crypto) on the balance sheet, sources of earnings variability.

The grid, as a checklist

You can reuse these eight reflexes for any IPO, not just SpaceX:

1. Primary or secondary? Where the raised money goes.
2. What’s the business? And its segments, if they diverge.
3. Growth vs profitability. Where does the loss come from?
4. Adjusted EBITDA vs operating income by segment. What does “adjusted” hide?
5. The cash. Capex, free cash flow, debt to refinance.
6. The valuation. Revenue multiple, dilution vs net assets.
7. Control and lock-up. Who votes, when insiders sell.
8. The risk factors. What the company admits itself.

None of these steps says “buy” or “don’t buy” — they tell you what you’re actually buying. The rest, as always, depends on your horizon, your risk tolerance and your written method. If you don’t have one yet, that’s where to start .