[Eldodi]

Crazy this company will IPO for >1B with such bad financials! That said, Starlink seems to be a real cash machine, not as good as ads but enough to support AI bets.

2025:

- Revenue: $18.7B, up from $14.0B in 2024

- Operating loss: -$2.6B

- Net loss: -$4.9B

- Adjusted EBITDA: $6.6B

- Operating cash flow: $6.8B

- Capex: $20.7B

Segment breakdown:

- Starlink / Connectivity: $11.4B revenue, $4.4B operating income, $7.2B adj. EBITDA

- Space / launch: $4.1B revenue, -$657M operating loss

- AI / xAI / X: $3.2B revenue, -$6.4B operating loss

Starlink metrics:

- Subscribers: 8.9M at end-2025, 10.3M by Mar 31 2026

- ARPU: $99/month in 2023, $81 in 2025, $66 in Q1 2026

Balance sheet as of Mar 31 2026:

- Cash: $15.9B

- Marketable securities: $7.8B

- Total assets: $102.1B

- Total liabilities: $60.5B

- Debt / finance leases: about $30.3B

[runako]

The numbers overall are worse than I expected. I can't believe Serious People are talking about putting this in the market at a trilly.

> Starlink seems to be a real cash machine

It has been said more than once that Starlink financials cannot be analyzed apart from SpaceX financials. Very easy to move the launch costs from one entity to the other depending on whether it is more beneficial to show more revenue for SpaceX or more profit for Starlink.

[amluto]

The use of EBITDA for Starlink is also interesting. For something like terrestrial fiber, I can imagine thinking that there’s a lot of depreciation on the books, but that most of the equipment keeps working after the depreciation period or is cheaper to replace than it was to buy, and that the right of ways and attachments don’t really depreciate. But Starlink satellites are actually gone at the end of their useful life.

I have not dug into the filing to see how this really breaks down.

[Robotbeat]

Starlink satellites can likely have their life extended by quite a bit, like past constellation satellites did. But it doesn’t really matter as SpaceX is still upgrading capability like crazy. Starship will mean a factor of 10-100x Starlink capacity expansion.

[sethops1]

They most certainly cannot - LEO satellites literally fall out of the sky after about 5 years; they talk about this in the S-1.

[Robotbeat]

It’s a function of altitude. They keep pushing the altitude lower. This is not the only option, although it has a lot of advantages.

[jcattle]

I mean, the SpaceX bet is that what you mention for terrestrial fiber

> or is cheaper to replace than it was to buy

will also hold true for cost of mass to orbit. There's a lot riding on making that prediction come true for SpaceX, hence all the CapEx going into Starship.

[Analemma_]

I can't believe that my index funds are going to be looted to pay for this turd.

[thephyber]

We can thank Nasdaq for lowering the standards to fast track SpaceX into an index with only having 5% float. Soon after it lists on the major indexes, we are gonna have some turbulence.

[mnicky]

Time to invest in some value-stocks index funds then :)

[mooreds]

I thought the same darn thing. Too bad there's not a ex-Space-X index fund.

[weregiraffe]

Short SpaceX to compensate

[huevosabio]

The space launch operating loss is like 10% of the Starlink operating income.

So Starlink is a cash cow!

[throwaway85825]

The launch market is just very small and always has been.

[aeternum]

Looks like it's gonna be closer to 2 trilly

[3eb7988a1663]

What is a realistic scenario for how this plays out? After index funds mechanically buy SpaceX, the insider lock out expires and they all sell - how low does SpaceX go? Will reality hit and a $2T valuation instantly drops to a more "reasonable" value for an unprofitable company? Will it stay in the clouds?

Tesla seems in a world of hurt unless robots start making space centers from moon rocks, yet that is also defying gravity.

[maerF0x0]

Why are index funds going to buy it? Market cap is not the only inclusion metric for S&P500, for example.

[zippothrowaway]

They just changed the rules for NASDAQ and they are changing the rules for S&P500.

The game is rigged. Our pensions and 401Ks are being forced to buy this slop. They can't resist taking all of the money off the table.

[maerF0x0]

Please source this, people need to read it and know what's happening

Edit: here's one https://www.reddit.com/r/Bogleheads/comments/1t2msfk/wsj_sto...

[Huppie]

I remember a recent episode of the Rational Reminder podcast about this, it's not as bad as it may look...

https://youtu.be/QaQHdF2DeRY

[dzhiurgis]

Rocketlab market cap is ~$70b (ATH right now tho). Not hard to imagine SpaceX being worth 30x.

[abirch]

When do we call AI bubble? Like 2 trillion value and losing billions.

[aeternum]

No need to worry because bubbles can't exist in space

[Zigurd]

This time it is different. All the best commencement speeches say so.

[maipen]

As if any of the marketcaps actualy reflect a company's true value. It's never just about financials.

[jfengel]

That's kind of the whole point of a stock market. If you already had a solid revenue stream, you wouldn't need investment.

These numbers would be kind of typical for a software play, since the great thing about software is that you write it once and then sell it many times. They're making a similar assertion for hardware: "fund rocket ship design, and sell it many times (i.e. lots of launches)".

The weird looking part to he is cramming xAI into it. It's a completely different business with little overlap that I can see, in a crowded market that they are far from leading.

[kentm]

> The weird looking part to he is cramming xAI into it. It's a completely different business with little overlap that I can see, in a crowded market that they are far from leading.

My personal theory is that Musk wants to roll up all his companies into a mega corporation that he fully controls, and this is part of the process. I expect Tesla and SpaceX to merge years down the line.

Of course, the counter to this thesis is that he didn't roll in Neuralink or Boring Company. But its probably that these three companies + Tesla are the ones he's most passionate about.

[jerlam]

Raising the market cap of Tesla is one of the requirements for Musk's nearly $1 trillion pay package. Merging companies is a lot easier than trying to sell more cars or non-existent robots.

[yibg]

These numbers would be ridiculous even for a software play. < 20B in revenue at almost 2T valuation? That's almost 100x revenue multiples at a not so great revenue growth rate.

[rchaud]

xAI "acquired" Twitter, so rolling it into SpaceX is a way to hide a badly distressed asset inside a potential moneymaker and hope nobody notices. Like mortgage backed securities.

[throwaway85825]

Then keep rolling in more money losers. Like Nortel.

[electriclove]

There were talks in the past about spinning Starlink out. Perhaps the thinking that led them to keep Starlink in is the same thinking about their new data center business (what they got from xAI and will grow in orbit in the future)

[stainablesteel]

putting tesla robots on the moon ran by LLMs seems to be a pretty coherent overall plan, I don't think it's different

[Rebelgecko]

Spacex has been playing fuckfuck games in recent months to boost their subscriber numbers.

The day after I got my dish I got an email that the price of the base plan would double. They also sent residential subscribers "free" dishes, which a ton of people took them up on right before the price change

[ACCount37]

I'm surprised launch is only -$0.65B, given just how much were they sinking into launch infrastructure and R&D for Starship.

Guess Falcon 9 the old reliable is still printing cash in the meanwhile.

[pclmulqdq]

"Printing cash" is generous given the expenses and attempted $2 trillion valuation.

[alopha]

Starlink is a cash machine because the costs are externalised to the rest of the company, all in it's a money pit.

[abirch]

SpaceX itself wouldn't be that bad, it's the xAI. It's going to burn through cash.

I remember Josh Brown talking about Peleton after its IPO: "Great Product, Horrible Investment"

[JeremyNT]

What is the best way to hedge against this turkey being included in my index funds?

[jaccola]

Mistiming is equivalent to being wrong. And Musk is good at keeping valuations high! So who knows when it will tank.

[sebzim4500]

Figure out how much of your fund it is and take a short position to hedge it

[__turbobrew__]

I have been thinking the same thing. I don’t want this turd in my index funds.

[doctoboggan]

short it?

[tobias3]

Choose another index where it is not included?

[thephyber]

It’s so big that it’s going to swing the markets when insiders start to liquidate after it is listed and on some indexes.

[Rury]

That alone will probably do little, due to contagion effects. Simply finding an index which excludes it might not be enough if it still shares other similar underlying stocks with indexes that do include it.

[moralestapia]

Typo: I'm sure you meant >1T.

>ARPU: $99/month in 2023, $81 in 2025, $66 in Q1 2026

Oof, are they already on diminishing returns phase?

While I don't think the financials are bad, I agree, this is definitely not a 1T company (but the market can stay irrational ...).

[tristanj]

Starlink is giving away the satellite dishes for free to grow customers. These dishes are expensive to manufacture and cost the company hundreds of dollars each. The estimated manufacturing cost of a Starlink standard dish is around $400.

[fragmede]

Which is a fine thing to say, but CAC vs LTV (customer acquisition cost vs lifetime value of the customer) is the underlying equation. If it costs them $150 to give away a dish, but they get, say, $300 before the user churns, they still come out ahead.

[golem14]

By making a lot of antennae, they also lower their price in the future due to the learning curve.

[wmf]

That shouldn't be included in ARPU.

[boelboel]

They've been upping the subscription prices recently past few months.

[takinola]

Can someone ELI5 Starlink revenue sources? At the core, it's an ISP (but served from space). What does Starlink have that differentiates from any other ISP? Is it because the TAM is global? That may be true at the margin, but I am sure most eligible subscribers would prefer a land-based system (eg no one in SF is cancelling their Xfinity to use Starlink) so how much is really left for them?

[greycol]

It's because they're good speeds in a lot of places that couldn't get good speeds before. It's also great for mobile work sites, i.e. construction sites, drilling camps, other b2b service businesses where a bunch of portacoms rock up to a site. Anywhere it's mildly hilly you can't actually assume you'll get a signal outside of town but a satellite dish basically guarantees that. Even if you can guarantee your in a spot long term the upfront cost of fibre or a tower may not balance out as cheaper than just eating the higher bandwidth costs.

It's also worth remembering that in a lot of places with low density it isn't appealing for competitors to build out to, so there's a lot of markets where it's a no brainer to switch from the local monopoly to starlink because the price was already inflated and it was worse service.

[jcattle]

There's a huge number of people in developing countries (think Indonesia, Brazil, Nigeria, Kenya, Mexico) where the country just lagged on internet build out. Now in these countries the price of a subscription will be a lot lower, than the >100$ you pay in the US, but since (simplified) the only additional cost per customer is the cost of the end terminal, it's still worth it for the ARR. The break-even point per customer will just be further in the future.

Also I wouldn't underestimate the amount of people living in rural areas of the US, Canada, Australia or Germany.

[stefan_]

I think the ISP story is a bit shitty outside some niche markets (mobile internet on airplanes and the like), but the military applications are massive. Everything is moving into drones and Starlink terminals are small enough to fit on them all, while being essentially unjammable, giving you a coarse location signal on top and plenty of bandwidth.

[kristiandupont]

For one thing, just about every airline in the world is a potential customer.

[dzhiurgis]

In NZ you can buy add-on so you can text/fb messenger/maps/weather and even apple music thru your phone via Starlink. I can't imagine this being super popular add-on, but if telco just rolls that into main offering I'd assume eventually that will be new norm.

[porphyra]

It's pretty much expected that a rapidly growing high tech company is gonna have a lot of losses and debt right? They're just spending huge amounts of money on capex. Not doing so would be like floating minerals in Starcraft: symptomatic of bad macro.

[bombcar]

Now you have me imagining uThermal as CEO of SpaceX and I think it'd be worth it to see the cheese runs.

[zelias]

"Watch me PLANETARY FORTRESS RUSH the war in Ukraine"

[Spartan-S63]

If they cleaved off xAI and let it die, they'd be in much better shape!

[electriclove]

Did you see that they are getting $15B/year from Anthropic because of what xAI built?

[wongarsu]

They are switching from digging for gold to selling shovels (capacity in their DCs). And it looks like they managed to make hay at the perfect time

[redox99]

xAI is by far their most profitable segment, receiving 1.25B a month from Anthropic.

[lubos]

That 1.25B per month is not profit

[redox99]

Assuming renting their datacenters doesn't cost them any more than running them for themselves, and plugging 15B a year of revenue (which ignores X entirely and other forms of revenue) you get 5.4B income, more than Starlink 4.4B income (which is slightly subsidized by the launch segment)

[tasty_freeze]

electricity is a major cost. that 1.5B/mo is part for renting the hardware, part paying the lease on the building, and a significant part (I have no idea what %) is paying for the electricity.

[jpkw]

Depreciation should be quite substantial - I recall reading that the starlink sats have a 5 year life expectancy?