Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

SpaceX Internet Satellite Network: Starlink

This site may earn commission on affiliate links.
How is that assumption 'quite fair'?

When you objectively analyze things like the split of legacy satellite services for USA vs ROW, the regulatory *sugar* show of ROW for a LEO (and a US company), inevitable competition, and the fundamental logic of who can actually afford FSS, it’s actually quite apparent. I appreciate not all those details are readily obvious to everyone, but that does not mean that are not so.

Ignoring the competition is coming argument:
US: 5 million out of 330 million people (FCC pegs unserved at 5.2 million)
Rest of world: 5 million out of 7.5 billion people
You could make a claim the monthly cost may be adjusted in developing regions, but a 0.07% take rate seems pessimistic.

Except that simply doing population math doesn’t all all reflect reality. Again, not even considering competition, it’s a function of prosperity (= high), density (= low), existing options (= shitty), and regulatory approval (= US friendly and without competing interests). Look at the world population distribution and find a non-US country anywhere near the top of the list that checks those boxes for Starlink.

Plan for India next year is 200k terminals.

Sure. Could happen. But maybe not. While that could of course just be posturing from The Indian Man, one would be remiss to not at least consider Bharti’s influence…

Then there are military, airline, cruise ship, and back haul service revenue streams.

Current satellite users is not a valid figure to base numbers off of beyond setting the minimum addressable market size. DSL and cellular are both better than current geostationary satellite service but worse than Starlink.

Current FSS subscriptions is in fact the best metric from which to build a Starlink user model. As repeatedly pointed out, near as makes no difference there’s nobody choosing to go dark while they’re waiting for good internet. Everyone in the US who 1) wants and 2) can afford internet is already paying for it. So when it comes to consumer level subscriptions, Starlink is, again near as makes no difference, explicitly in the game of stealing customers from the competition. That of course includes folks with shitty terrestrial solutions (so, on top of the 10M FSS bucket) but that larger potential base is also countered by having to compete with evolving terrestrial and space solutions.

Airlines, military, and cruises are not a slam dunk. Cruises certainly are the easiest to close, and certainly commercial level customers are generally A Good Thing, but it’s not like the couple hundred cruise ships around the world are going to represent significant revenue so we can move on…

While the US military could represent a much larger chunk of revenue there are plenty of security issues to work out. Given the speed at which The Man works and is willing to change “that’s the way we’ve always done it”, don’t expect anything huge in any short amount of time.

Airlines might seem like a strong opportunity (nobody has said “my gogo is plenty good”), but of course strapping tens/hundreds of bored people to one sat link is also a perfect storm of steaming congestion, so there’s plenty of work to do on service level, customer pricing, etc. Airlines are absolutely a race to the bottom dollar, and so any added budgetary line item needs to be wholly accounted for [in the black] by the bean counters. And let’s face it, “better internet service than the other guys” is something few are looking for when they book a flight, so there’s a lot of work to do on the customer side too. People are either loyal to their airline or they’re loyal to Expedia. The ability to steam the new season of tiger king from 30000 ft isn’t going to change that.

All that is why,IMO cell backhaul is likely the biggest commercial opportunity for Starlink. Rather than a new or materially revolutionary service, It’s an expansion of an existing and otherwise ubiquitous service, it can be funded by proven marketing strategies (“can you hear me now?”), and it enables Starlink to be a partner with the connectivity players rather than a direct competitor.

Finally, Do not underestimate human nature when it comes to products vs services. Starlink is not like a Tesla 2.0. With Tesla, there’s a tangible set of differentiators in the product. Internet service is much more abstract—people don’t care where it comes from, just that it works and is cheap. As competition evolves (GEO operators, LEO operators, and most importantly, terrestrial solutions) it’s simply going to come down to cost. To wit, to the average human, $75for 75meg comcast >> $100 for 500meg Starlink.
 
When you objectively analyze things like the split of legacy satellite services for USA vs ROW, the regulatory *sugar* show of ROW for a LEO (and a US company), inevitable competition, and the fundamental logic of who can actually afford FSS, it’s actually quite apparent. I appreciate not all those details are readily obvious to everyone, but that does not mean that are not so.



Except that simply doing population math doesn’t all all reflect reality. Again, not even considering competition, it’s a function of prosperity (= high), density (= low), existing options (= shitty), and regulatory approval (= US friendly and without competing interests). Look at the world population distribution and find a non-US country anywhere near the top of the list that checks those boxes for Starlink.



Sure. Could happen. But maybe not. While that could of course just be posturing from The Indian Man, one would be remiss to not at least consider Bharti’s influence…



Current FSS subscriptions is in fact the best metric from which to build a Starlink user model. As repeatedly pointed out, near as makes no difference there’s nobody choosing to go dark while they’re waiting for good internet. Everyone in the US who 1) wants and 2) can afford internet is already paying for it. So when it comes to consumer level subscriptions, Starlink is, again near as makes no difference, explicitly in the game of stealing customers from the competition. That of course includes folks with shitty terrestrial solutions (so, on top of the 10M FSS bucket) but that larger potential base is also countered by having to compete with evolving terrestrial and space solutions.

Airlines, military, and cruises are not a slam dunk. Cruises certainly are the easiest to close, and certainly commercial level customers are generally A Good Thing, but it’s not like the couple hundred cruise ships around the world are going to represent significant revenue so we can move on…

While the US military could represent a much larger chunk of revenue there are plenty of security issues to work out. Given the speed at which The Man works and is willing to change “that’s the way we’ve always done it”, don’t expect anything huge in any short amount of time.

Airlines might seem like a strong opportunity (nobody has said “my gogo is plenty good”), but of course strapping tens/hundreds of bored people to one sat link is also a perfect storm of steaming congestion, so there’s plenty of work to do on service level, customer pricing, etc. Airlines are absolutely a race to the bottom dollar, and so any added budgetary line item needs to be wholly accounted for [in the black] by the bean counters. And let’s face it, “better internet service than the other guys” is something few are looking for when they book a flight, so there’s a lot of work to do on the customer side too. People are either loyal to their airline or they’re loyal to Expedia. The ability to steam the new season of tiger king from 30000 ft isn’t going to change that.

All that is why,IMO cell backhaul is likely the biggest commercial opportunity for Starlink. Rather than a new or materially revolutionary service, It’s an expansion of an existing and otherwise ubiquitous service, it can be funded by proven marketing strategies (“can you hear me now?”), and it enables Starlink to be a partner with the connectivity players rather than a direct competitor.

Finally, Do not underestimate human nature when it comes to products vs services. Starlink is not like a Tesla 2.0. With Tesla, there’s a tangible set of differentiators in the product. Internet service is much more abstract—people don’t care where it comes from, just that it works and is cheap. As competition evolves (GEO operators, LEO operators, and most importantly, terrestrial solutions) it’s simply going to come down to cost. To wit, to the average human, $75for 75meg comcast >> $100 for 500meg Starlink.
I think what you're forgetting is that the Terrestrial solutions will have less incentive to spend the enormous amount of money required for a few rural customers because the number of potential rural customers will now decrease due to Starlink.
 
Current FSS subscriptions is in fact the best metric from which to build a Starlink user model. As repeatedly pointed out, near as makes no difference there’s nobody choosing to go dark while they’re waiting for good internet. Everyone in the US who 1) wants and 2) can afford internet is already paying for it. So when it comes to consumer level subscriptions, Starlink is, again near as makes no difference, explicitly in the game of stealing customers from the competition. That of course includes folks with shitty terrestrial solutions (so, on top of the 10M FSS bucket) but that larger potential base is also countered by having to compete with evolving terrestrial and space solutions.

People only have what is available. For 33 million Americans, that's DSL About 4,860,000 Added Broadband From Top Providers in 2020 | Leichtman Research Group

Which is being phased out with no replacement AT&T shelving DSL may leave hundreds of thousands hanging by a phone line

And that is for sub 1Mb up, 8 Mb down in my area.
At this point for many, there is no low latency alternatively besides possible cellular coverage; and there is no evolving competition in the useful future. Rural infrastructure built out is not profitable for cable/ fiber. No other major constellations are launching at this point.
So 33 million underserved DSL subscribers + 10 million geo sat users + some number of unserved people (yrs, some people have nothing rather than geo sat), new construction, cable converts = 45million+.
Even at 30% conversion, that's over 10 million just in the US.

As to the world, there is a large unserved opportunity. China and India each have over 500 million unconnected people. With government provided downlink sites, users would be virtually free after terminal purchase. Any revenue is incremental profit.
These are the countries where internet access is lowest
 
People only have what is available. For 33 million Americans, that's DSL About 4,860,000 Added Broadband From Top Providers in 2020 | Leichtman Research Group

Which is being phased out with no replacement AT&T shelving DSL may leave hundreds of thousands hanging by a phone line

And that is for sub 1Mb up, 8 Mb down in my area.
At this point for many, there is no low latency alternatively besides possible cellular coverage; and there is no evolving competition in the useful future. Rural infrastructure built out is not profitable for cable/ fiber. No other major constellations are launching at this point.
So 33 million underserved DSL subscribers + 10 million geo sat users + some number of unserved people (yrs, some people have nothing rather than geo sat), new construction, cable converts = 45million+.
Even at 30% conversion, that's over 10 million just in the US.

As to the world, there is a large unserved opportunity. China and India each have over 500 million unconnected people. With government provided downlink sites, users would be virtually free after terminal purchase. Any revenue is incremental profit.
These are the countries where internet access is lowest
Lots in Australia, NZ, and Germany also have terrible options. In my small area of a Washington DC exurb ..still somewhat rural county I can assure you that every single home is on the waiting list. 100% We have no major company offering high speed internet. It is a Tmobile or brand X cell phone or nothing and ...the cell coverage sucks. I can't believe that there is any cap on the addressable market that is relevant to spacex. I would imagine rather the opposite. It is simply a matter of procuring and delivering devices and establishing ground stations and big box's in space.
 
. And let’s face it, “better internet service than the other guys” is something few are looking for when they book a flight
I don't know about others, but I wouldn't mind paying something like $10 for each domestic flight and $25 per Atlantic/Pacific flights for a decent continuous internet connection. The current one time charges and the monthly subscriptions that Gogo and others offer are atrocious and favor frequent fliers only.
 
  • Like
Reactions: Dave EV
Sure. Could happen. But maybe not. While that could of course just be posturing from The Indian Man, one would be remiss to not at least consider Bharti’s influence…
FFPDoXLaUAAgn1-
 
  • Like
Reactions: house9
So 33 million underserved DSL subscribers + 10 million geo sat users + some number of unserved people (yrs, some people have nothing rather than geo sat), new construction, cable converts = 45million+.
Even at 30% conversion, that's over 10 million just in the US.
Glad that you mention "cable converts". We have cable-based Internet service at our primary residence in a semi-rural neighborhood. The service quality is lousy and the cable company doesn't seem eager to stay on top of maintenance. While we have Powerwalls and can keep running during power outages, the cable system's backup batteries go dead after a few hours (and cellular service also degrades greatly). We will gladly pay $99/month, instead of ~$60/month, for the reliability of Starlink and the ability to function off the grid through outages and potential natural disasters! From speaking with neighbors, I know that others feel the same way. Tech-savvy people who've left the cities will love Starlink!
 
I think what you're forgetting is that the Terrestrial solutions will have less incentive to spend the enormous amount of money required for a few rural customers because the number of potential rural customers will now decrease due to Starlink.

Its actually a little disappointing that this is the takeaway for so many people. Clearly the above describes the primary consumer customer base and nothing I've said suggests I'm "forgetting" that; what I'm trying to do is move the conversation beyond a rudimentary "Starlink is going to be so awesome and I can't wait for it and everyone is going to love it " toward something that attempts to analyze starlink in a more practical, realistic manner. If folks would prefer to just High Five each other over Starlink though, I'll slip into the shadows here.

Interestingly, it seems as though lots of folks here are forgetting that everyone is in this for the money, and everyone (save for maybe Kuiper) needs their service to be a profit center. That goes for terrestrial wired solutions, terrestrial wide area LAN type solutions, MNOs, LEO constellations, and GEO constellations. Starlink is absolutely disrupting the status quo and as a result not all competitors are going to make it, but don't for a second think that any of them aren't spending big bucks on proper analysis of Starlink (and everyone else) relative to their offering.

Similar to Tesla, Starlink's biggest advantage is its speed--not the speed of the service, mind, but their ability to bring a product to market. None of the other competitors will be able to keep up in that respect. But, like we're seeing in the auto industry where other mfg's are starting to offer compelling products rather than simply roll over and let Tesla win, don't think for a second Comcast or Viasat or Verizon or anyone else is simply going to curl up and let Starlink take all. If there's money to be made on rural customers, those competitors are going to go after it.

People only have what is available. For 33 million Americans, that's DSL About 4,860,000 Added Broadband From Top Providers in 2020 | Leichtman Research Group

1. Don't confuse people with subscribers.
2. Don't dismiss the fact that all of those people have access to satellite now, and they're not paying for it.
https://www.leichtmanresearch.com/about-4860000-added-broadband-from-top-providers-in-2020/
As to the world, there is a large unserved opportunity. China and India each have over 500 million unconnected people.

And China is a non-starter for Starlink and India is far from a sure thing.

Again, in no particular order:
--Prosperity
--User Density
--Competition
--Regulatory Approval

I don't know about others, but I wouldn't mind paying something like $10 for each domestic flight and $25 per Atlantic/Pacific flights for a decent continuous internet connection.

Sure, as would many of us here I suspect. Again, there's clearly a market there. The question is just how big is that market? We don't need to live in binary here...


Again, sure. But again, in the spirit of actually moving beyond "Starlink is going to be so awesome", how about you try to address:
1. The fact that a government agency (the DOT) had to tell people "Don't sign up for starlink"?
2. What role Bharti will play.
 
I don't see the concern here. Indian govt is asking SpaceX not to sign up any customers until they are approved for offering their service. And they are educating their citizens not to put the cart before the horse.
 
Again, sure. But again, in the spirit of actually moving beyond "Starlink is going to be so awesome", how about you try to address:
1. The fact that a government agency (the DOT) had to tell people "Don't sign up for starlink"?
2. What role Bharti will play.
I agree that the people here are not representative of overall customer base. Keep in mind thought that current satellite based solutions do not compete well with even lower end DSL offerings for many uses. The longer term addressable market for StarLink will be quite large, the bigger question needs to be what is the addressible market per generation of the constelation, and how the profitability of that constelation will keep up.

I also can see that both SpaceX and the Indian DOT are posturing against each other. SpaceX is trying to get a lot of per-subscription signups, so that they can go to the Indian government and demand the license get approved due to how many citizens want the service. The DOT is trying to warn citizens that it is not an officially approved service yet, and do not "waste" your money on something that may not happen, and have to get a refund in months/years if it does not get approval.

Both want to be able to say that they are looking out for the Indian consumer.

Overall I expect SpaceX to eventually move towards an opportunity cost based pricing model in parts of the world that can not support the full cost subscription, but that won't be for a while, and has to be done within the power duty cycle of the satellites.

Eventually SpaceX will have an array of service offerings, that range from low cost, standard consumer, back-haul, low latency, and evn isolated (middle of the ocean), all of which will need to be priced and prioritized correctly.

-Harry
 
Eventually SpaceX will have an array of service offerings, that range from low cost, standard consumer, back-haul, low latency, and evn isolated (middle of the ocean), all of which will need to be priced and prioritized correctly.

That's the key here. People want to do math on American pricing of consumer level Starlink and apply it blindly to the rest of the world's population.

In reality, the overwhelming majority of the world gets their internet solely from their mobile devices, because they can't afford fixed connectivity or the devices to operate on that service. India, like most of the world's population, is both a) super dense in population, and b) super poor (their GDP is something like ~3% of the US). Both are terrible for a business plan to provide consumer level FSS, the former because most of the rich people in India live in dense areas (and so any consumer level satellite service would be density limited, similar to Starlink in the 'burbs here in North America) and the latter because most of the people couldn't dream of affording satellite service in the first place. Perhaps that's all moot, as its also worth noting that out of the 1.3B people in India, there's only something like 20M fixed service subscriptions.

So...for Starlink, the real opportunity in India is backhaul for the Indian MNOs in areas where it is unprofitable for those MNOs to currently offer service via terrestrial solutions. That necessarily means that the MNOs need to be able to, at a minimum, cover the commercial cost of a Starlink backhaul network through increased subscriber revenue at a price that still returns acceptable profit to Starlink. Given the population distribution of India (= almost everyone lives where there's already coverage) that's not exactly a foregone conclusion. And then similar to the inevitable ebb and flow of rural America service options, as terrestrial companies see those rural areas returning profit and as they improve their infrastructure $/bit, they're going to start pushing back into those rural areas, reducing the effective footprint of satellite services. Everyone is in this for the money, and nobody is simply going to let the other person take all the cookies.

Then, of course, is Bharti. Despite folks hand-waving past this one (or I suspect more succinctly, not understanding why I keep bringing it up), it could be a real turd in the Starlink punchbowl. Certainly given their 40% stake (or whatever) in Oneweb, Airtel is going to at a minimum lock out their ~1/3 share of the 1B+ mobile subscriptions in India. That's maybe not the end of the world since Jio and Vodaphone still have 2/3 of the subscriptions....but...it is also very plausible that Bharti has the right influences within the GoI to materially limit or even lock out an American telecom company completely. While the latter is admittedly a bit town crier, one would be foolish to think that Starlink will be able to walk into regulatory approval completely unchallenged.
 
  • Like
Reactions: petit_bateau
But, like we're seeing in the auto industry where other mfg's are starting to offer compelling products rather than simply roll over and let Tesla win, don't think for a second Comcast or Viasat or Verizon or anyone else is simply going to curl up and let Starlink take all.
If Tesla's recent experience in the auto market is any indication, Starlink will indeed run well ahead of the competition for years to come, at least in areas with lower population densities. There is still not a single non-Tesla EV on the market that's a great value proposition relative to Tesla vehicles. Even the relatively good non-Tesla EVs, such as the Ford Mustang Mach-E, are not produced in sufficient volumes to truly take down ICE vehicles. This is relevant here because Tesla and SpaceX have much in common in their respective corporate cultures. As you mentioned, speed of innovation is a key advantage.
 
If it’s actually true (I’d like to see the story picked up by a bigger outlet like SN or Ars), Elon 1) pretty much perpetually runs his companies financially ragged so this wouldn’t exactly be a shocking scenario and 2) is quite a drama queen when it comes to what he thinks are motivating statements.

Given the amount of flight time, gut feel says raptor production is just slow and not plagued by some deeper issue. It could be as simple as some exec head nodding to some impossible Elon demand and letting it snowball out of sight until some untenable impasse. Elon can ultimately be set straight, so in that scenario it’s just a matter of resetting expectations and moving on.

I also think there’s a bit of nuance in the ‘v1 is financially weak’ statement. IMO what that’s really tied to is the recurring overhead of F9 vs their aspiration on Starship. In other words, hucking up equivalent mass of V1 sats on Starship as v2 sats (which, it’s no secret, is going to be physically bigger, and so will benefit from some combination of upsides that I described recently upthread) would almost certainly end up as a slightly more expensive exercise with slightly lower constellation level performance. But not like a make or break difference. More like a ‘maybe Elon will have to worry a teeny bit more about Jeff in the #2 position’ difference.
 
Maybe just a roadblock waiting for the right people to pay off the right people...but maybe a worse sign for Starlink in India.

To turn a phrase: Space Is Easy...getting local regulatory approval to offer satellite service around the world is The Hard Part. (And yes, it absolutely involves actual bribes.)
Or just the same deal as the FCC; you can't sell or take orders for a product until it's certified. The rules on certification changed in 2021: What is the WPC ETA Certification - India Certification – Your expert for India Certifications

How do I get the WPC license?​


The WPC certification process is relatively simple compared to other certification types and can be completed within 2 to 3 months. After the application, the applicant has to submit RF (Radio Frequency) Test Reports which have to be issued by an accredited testing laboratory. Until 2020, it was possible to conduct these tests abroad and use these reports. However, since the beginning of 2021, international test report are no longer accepted and hence, tests have to be conducted locally in India. A factory inspection is not necessary for the WPC (ETA) Approval.
 
  • Like
Reactions: petit_bateau
Or just the same deal as the FCC; you can't sell or take orders for a product until it's certified.

India is exponentially harder to secure approval to offer satellite service than most countries...or at least American based service, since that's my major data point. Does that mean India is a non-starter for Starlink (like China)? No. It means that the duck will be paddling full out, compared to a leisurely swim for the US and many friendly nations.

What's concerning is that dude was in his job for 3 months and then left. Its hard to imagine he's resigning simply because some process might take a bit longer than expected. Best case, he just couldn't deal with the relentless pressure of being an exec in Elonco.

Unfortunately, any way you skin it this is also a troubling sign that Starlink might not know what they're doing with respect to navigating foreign regulations. Whether they don't know how to hire the right person (and/or got fleeced in the process of hiring someone), haven't done basic diligence on actual regulatory processes, or simply think they can bulldog through foreign bureaucracy with a typical SpaceX MO of ask forgiveness...hopefully they're learning some lessons here and are changing tact for other markets.