Hacker News Reader

Ridiculous. SpaceX offers a product that costs far less than its competitors while being as good or better in most respects. Their profit margins on launches must be enormous at this point.

That in turn enables Starlink. They can put up thousands of satellites very cheaply. Then they can turn around and sell subscriptions. Starlink has about 8 million active customers. At $40+/month, that's at least $4 billion/year in revenue. Probably a lot more. Given their launch costs, that's a ton of profit.

"not that Starlink hasn't been tried 10 times before" is just... not true. Nothing like it was ever tried before. Iridium is the only one that came even vaguely close, and it was still a radically different type of service. Iridium was extremely low capacity phone service, then low-bandwidth (it made dialup look super fast by comparison) data, with a network of a few dozen satellites covering the globe. It could not support many customers because it had few satellites. It also had to pay for launches in the 1990s, so an order of magnitude or more costlier. That means that it was enormously expensive, for a product few people actually needed. Handsets cost thousands of dollars, then you got to pay several dollars per minute on top of that.

Iridium was basically space dialup, and extremely expensive space dialup at that. Starlink is space broadband, and their cheap launch costs and other technological advancements mean the service is profitable at a competitive price point.

> Blue Origin has beat SpaceX to Mars

As you said, not especially relevant to a financial discussion.

> as for profitability

SpaceX is profitable.

> US government cannot risk having SpaceX as a single option to get to orbit, so it has no choice developing a publicly funded rocket program

Being the U.S. government's prime contractor while it keeps ULA on life support is a great deal. Same for Europe and Arianespace.

> SpaceX is profitable.

Strange. For me profitable means money_out > money_in, over the whole company history. OpenAI stock is worth 137 billion. Of this, at least 20B was actual money put in (some "free" by the US government). So let's say $18B investment, although that should really be increased by whatever investments were made in Starlink, which is also >$10 billion.

Revenue for SpaceX, with starlink split out (because starlink pays in equity). Obviously, these are guesses.

2024: $14 billion revenue, of which $10b is Starlink investment

2025: $15 billion revenue, of which $11b is Starlink investment

Starlink, when some figures were publicly discussed, in 2023, had $55 million profit on $1.5 billion revenue (but that was counting Starlink shares as cash), about 3.7%

Let's very generously say they doubled that in the past 2 years, and hey, we're being generous, round up. Say they're at 10% margin (Musk's claim, of course, is 60%).

So:

2024: $140 million profit, of which $10b was invested in Starlink

2025: $150 million profit, of which $11b was invested in Starlink

Cashflow:

2024: $-10 billion

2025: $-11 billion

So I think I can be very comfortable in saying that SpaceX is not profitable, it's deep in the red, getting worse over time, only supported by Starlink valuation. With somewhat less confidence I can say it's actually getting less profitable over time, rather than more.

They would need to at least 5x the launches they did in 2024 to just breakeven (while potentially reducing starlink launches to zero, which would make it close to 15x), and they only increased the launches by 20% (138 in 2024 vs 129 launches in 2025 up to october). 20% is generous. There is the question of how much of these launches were Starlink, and for 2025 there is no good data, but up to 2025 Starlink launches increased by over 40%.

My assessment is unless Starlink takes over global internet, SpaceX is bankrupt and will have to sell it's designs and launch technology for pennies on the dollar.

Oh and can I just add, I've interviewed at Starlink and received a (pathetic) offer, which people on reddit claim was pretty typical. Which told me one thing: Starlink is already very aware of the need to drastically save money.

That is not a standard definition of profitability, and basically a version of the sunk cost fallacy. you dont go bankrupt from sunk costs.

> For me profitable means money_out > money_in, over the whole company history

One, this is a nonsense definition. Two, SpaceX meets it. It’s why it hasn’t fundraised for years.

> I've interviewed at Starlink and received a (pathetic) offer, which people on reddit claim was pretty typical. Which told me one thing: Starlink is already very aware of the need to drastically save money

Wat. Based on this metric, Amazon and Walmart are broke.

> One, this is a nonsense definition. Two, SpaceX meets it. It’s why it hasn’t fundraised for years.

First, It's a private company. You can't know that.

Second, even if true, massive investments were made into Starlink and paid into SpaceX, as I pointed out into my post. That counts as fundraising.

> Wat. Based on this metric, Amazon and Walmart are broke.

Not really. Amazon pays a lot, and for IT Manager roles, so does Walmart.

> It's a private company. You can't know that

Private companies still have shareholders and audited financials.

> massive investments were made into Starlink and paid into SpaceX, as I pointed out into my post. That counts as fundraising

No. If I invest $1 in SpaceX and it invests that in Starlink, that’s $1 of fundraising.

> Amazon pays a lot, and for IT Manager roles, so does Walmart

They also pay other people peanuts. I know plenty of people whose net effectives at SpaceX are seven to eight figures. That doesn’t obligate them to pay top dollar for every role.

> No. If I invest $1 in SpaceX and it invests that in Starlink, that’s $1 of fundraising.

Okay, I'll bite. And what if you "invest" $1 in Starlink, and Starlink immediately "invests" that $1 in SpaceX, while also giving SpaceX starlink shares?

I guess I don't get to make the "wait, isn't that Elon Musk paying Elon Musk" argument, despite it being true?

I mean I even wish you were correct, but by my estimates the most optimistic view of SpaceX is that they're still about 400% too expensive to get global satellite internet working, when launching at cost. That's a remarkable achievement ... but not good enough. It's so close it really sucks. On the other hand, I hate AT&T as much as the next guy but I also don't want a global mobile internet monopoly replacing them. And if Starlink is going to have competitors the value drops by half easily.

> They also pay other people peanuts ...

Well my point is, of course, that SpaceX pays very little compared to Amazon and Walmart for roles that pay a lot at Amazon and Walmart. SpaceX pays only a little better than NASA, and NASA pays about double what you get in an academic position.

Oh, and yes, double what you get in an academic position is pretty badly paid, especially considering what most NASA engineers are capable of.

TDLR: A link to bad short-seller video from 4 years ago and a lot of unsustainable claims.

Option 1 isn’t really an option, unfortunately. There are no viable single launch options using it. So it’s really SLS x 2. But building and launching one SLS at a time is almost too much as it is. If that’s the only option, I think Artemis is dead and we should start over.

> Plus Blue Origin has beat SpaceX to Mars

What about that Tesla that regularly crosses Mars orbit? Ok, it's not on Mars, but it was just about calculating an orbit. They could have smashed it on Mars as well.

Injecting a dummy payload into an eccentric helicentric orbit which periodically crosses' Mars' orbit /= a Mars mission. The complexity and relevance to future human presence on Mars isn't close

(Though tbf the choice of launch vehicle isn't that relevant to whether the ESCAPADE mission succeeds, and missions involving Mars flybys like Hera which are lot more serious than the Tesla one have been launched on SpaceX rockets)

If injecting a payload into an eccentric heliocentric orbit doesn't count, then why does injecting a payload into MEO count, just because that payload is then capable of getting to Mars from there? BO didn't launch it to Mars, they launched it to orbit, and ESCAPADE is now getting itself to Mars.

Because ESCAPADE is a mission to Mars which will orbit Mars and send back data from Mars and the dummy payload released by the Falcon Heavy test launch as a fun little meme which orbits the sun millions of miles from Mars isn't (It'd be easier to claim the SpaceX-launched deep space missions getting Mars gravity assists as Mars missions, at least those missions have actual success criteria related to Mars proximity and might take some photos on the way past).

A Mars mission is more of a Mars mission than launching something which isn't on a mission and doesn't go to Mars isn't very difficult to understand, unless you're actively trying not to. I'd already pointed out that the choice of launch vehicle is largely irrelevant to ESCAPADE's success; no shit they've got their own propulsion. The roadster would have needed it (and a suitable launch window or funky itinerary) to actually get to Mars too. I don't think it's crazy that despite having the engineers and cash to have got to Mars already SpaceX are being beaten by a payload built by one of their newspace contemporaries and launched by another, but a certain SpaceX chap does keep insisting that Mars is the all important destination.

If the choice of launch vehicle is largely irrelevant then how does this make any sense? We’re talking about the launch companies. The claim was “Blue Origin has beat SpaceX to Mars.” But they did no such thing. BO launched a payload into Earth orbit. They have nothing to do with where it goes from there. SpaceX didn’t get to Mars either but at least they were the ones who actually got the Roadster into heliocentric orbit.

The OP made the point (amongst many more dubious claims) that SpaceX was one of the few established launch providers that hadn't sent a payload to Mars. This is true (it says more about Mars not actually being a priority for SpaceX than it does about their technical capability, but that in itself is interesting). Another person suggested that we should count the Tesla roadster as a Mars mission; I pointed out that it shouldn't count as a Mars mission because it wasn't a Mars mission (and injecting a meme into a random deep space orbit isn't a more impressive display of technical prowess than deploying an actual mission payload to an L2 staging orbit; the whole point of the Roadster stunt was they didn't particularly care if the payload survived never mind where it ended up). It's not rocket science! Most stuff a Falcon 9 launches reaches its final orbit with its own propulsion; it still gets credited with the launch.

And I’d certainly credit BO with launching this payload to orbit. But saying they sent a payload to Mars doesn’t make any sense to me. It’s like saying a taxi took me to Europe, because they took me to the airport. A critical part of the journey, certainly, but only part.

Well if no launch provider gets credit for launching Mars missions unless they're also supplying the payload/propulsion this whole argument is moot (just leaves SpaceX behind Berkeley/Rocketlab and ISRO rather than Blue Origin and ISRO in the Mars race...)

If we're doing analogies to holidays, the Starman Tesla Roadster definitely hasn't proven its worth for visiting London, even though it's gone beyond it (and SpaceX could have made it hit London if they'd really wanted to, just ask Werner von Braun :-)

Quibbling about whether parties involved in conveying tourists to actually visit London receive undue credit for their segment of the journey, or whether the London Underground is more important than Gatwick flights which don't even reach City boundaries seems like pointless pedantry on the other hand, particularly if deployed in defence of the claim that flying away from London at escape velocity counts as a London visit.

Pretty clear SpaceX could have sent something to Mars for about 10 years, had they so wanted.

And also pretty clear that if Mars colonization was as imminent and important as an typical Elon comment on the subject suggests, they would have bothered with a probe by now...

Nice, the wording of "smashed it". Because, the point of getting to Mars or Mars orbit is that you need rocket burns to insert and to land on Mars. Getting to places in space is a delta-V game and paying only half your delta-V costs doesn't count because it doesn't work.

If that's why it matters, then why are you crediting BO for that when they didn't have anything to do with the parts which will be performing those burns?

I guess they get partial credit if those parts and payload are heavier than the car.

Alas for BO’s credit, the two spacecraft together are only a little over one ton.

Yep, he is desperate for cash, he is leveraged to the hilt on his shares which is why he desperately begs his fanboys for more. His empire is a house of cards.

> as has Blue origin in the US

Blue Origin is losing many billions every year and has only survived thanks to a hobby project. And even if they continue, to get to SpaceX like cadence is a long way away, and many more billions in investment.

Europe is a decade plus behind and has no way ever to get to the launch cadence. And even then they have 0 chance competing for international launch. And a true Starlink competitor out of Europe is fantasy.

> Also the EU has setup a working Starlink competitor

No it doesn't.

> .. not that Starlink hasn't been tried 10 times before, with the most spectacular crash being Iridium

If you really think Starlink and Iridium are comparable, you should get your head checked.

And just because something hasn't worked before, doesn't mean changing technology doesn't change that.

The question you should ask is "Are there historical example where a 10x drop in cost allowed for a new much larger volume in an industry". And if you look at it that way, its patently obvious.

SpaceX doesn't even need to pay itself margin, if SpaceX had to fully buy SpaceX launches, the economics would be a lot worse.

> TLDR: SpaceX is bankrupt, Starlink is a pets.com

This is analysis where my only conclusion is that you just hate Musk and SpaceX for political reasons.

Did pets.com make like 10 billion in revenue and had many major militaries as costumers? I must have missed that.

> so it has no choice developing a publicly funded rocket program.

It does have a choice the US doesn't need to publicly fund anything. They already have ULA, SpaceX, BlueOrigin, Rocket Lab, Stoke space, Relativity.

> Everyone always makes the point that SpaceX is cheaper than SLS. However ... this fails to correctly compare prices for the only options the US government has:

You act as if SLS is the law of the universe, but it isn't. Anybody with a brain has known for 10+ years that eventually the US will switch to commercial rocket launch. As NASA has already mostly done, and DoD has done 100% already.

SLS is the last vestige of a dying system of cronies from the Shuttle days. It has not future. Only a long political fight to suck up as many resources at can be extract from congress before it inevitably dies.

The future in the US is clear, competitive launch with SpaceX as the leading provider and ULA, Blue and friends competing for contracts.

Every time someone mentions Eutelsat as a competitor I'm reminded that my own friend group has multiple people who can simply buy the entire company, which sort of describes how successful it is.

Option 1 isn't an option, really. NSSL policy is to ensure that there are two independent providers so that Assured Access To Space can work.