Is SpaceX Its Own Best Customer?
SpaceX is famous for landing rockets.
But landing a rocket creates another problem.
What do you do with it next?
A reusable rocket is only valuable if you have something to launch.
And SpaceX found an unusual solution.
It became one of its own biggest customers.
In 2026, roughly 79% of Falcon 9 launches have been dedicated to Starlink, SpaceX's own satellite network.
That means most Falcon 9 launches are no longer primarily carrying satellites for NASA, governments or outside companies.
They are carrying SpaceX's own satellites.
That may explain something important about SpaceX that the famous rocket landings do not.
SpaceX didn't just figure out how to reuse a rocket.
It built a business that gives the rocket something to do.
Reusable Doesn't Automatically Mean Cheap
Imagine an airline buys an aircraft that can fly for 30 years.
That sounds like a valuable asset.
But if the aircraft flies twice a year, the economics are terrible.
The airline still paid for the aircraft. It still needs pilots, maintenance crews, airports and infrastructure.
The more frequently the aircraft flies, the more opportunities the airline has to spread those costs across flights.
Rockets have a similar problem.
Developing a rocket requires enormous amounts of engineering, factories, launch pads and testing.
Making the rocket reusable solves only part of the problem.
You also need launch frequency.
SpaceX has achieved both.
In 2025, it launched 165 Falcon 9 missions.
Only eight used a new first-stage booster.
The other 157 used flight-proven boosters.
By March 2026, SpaceX said it had completed approximately 650 orbital launches across its vehicles, with more than 540 flown by flight-proven Falcon rockets.
The rockets are reusable.
But just as importantly, SpaceX keeps finding reasons to reuse them.
Then SpaceX Built Starlink
Originally, the obvious customers for a rocket company were outside organizations.
NASA needs something launched.
A military needs something launched.
A satellite company needs something launched.
They pay the rocket company.
The rocket flies.
Then SpaceX created Starlink.
Now SpaceX manufactures satellites, launches them on its own rockets and uses them to operate an internet network.
That changed the economics.
Instead of waiting for someone else to need a rocket, SpaceX could create much of its own launch demand.
And that internal demand has become enormous.
Starlink accounted for about 54% of Falcon 9 launches in 2020.
By 2026, it was approximately 79%.
The rocket company had effectively created a customer inside itself.
The Customer Became Bigger Than the Rocket Business
Then something even more interesting happened.
Starlink stopped looking like a side business.
In 2025, SpaceX's Connectivity segment generated approximately $11.4 billion in revenue.
Its Space segment generated approximately $4.1 billion.
SpaceX's own investor materials say the Space segment includes external customer launches but does not record internal revenue when SpaceX launches its own satellites.
So SpaceX does not pretend that one division "sold" a Starlink launch to another division and count that as revenue.
Instead, the costs associated with launching Starlink satellites are capitalized as part of the satellite assets.
The accounting reflects the underlying business logic:
The rocket is helping build the internet network.
And that network has become a much larger revenue business than external launch services.
Starlink Gives Falcon 9 Something to Do
The relationship works in both directions.
Falcon 9 made it possible to deploy thousands of Starlink satellites without buying a launch from another company every time.
But Starlink also created enormous recurring demand for Falcon 9.
SpaceX builds rockets.
Those rockets launch Starlink satellites.
The satellites build a larger network.
The network generates connectivity revenue.
And maintaining and expanding that network requires more satellites and more launches.
So Starlink did not simply benefit from cheap rockets.
The rocket business benefited from having Starlink to launch.
SpaceX Can Even Fill Empty Launch Slots With Itself
There is another advantage that is easy to miss.
Rocket customers sometimes get delayed.
Their satellite may not be ready.
Testing may take longer.
Manufacturing problems happen.
A normal launch provider may have a rocket and launch infrastructure ready but no payload ready to fly.
SpaceX has another option.
Its investor materials explain that if an external customer's payload is not ready for a scheduled launch opportunity, SpaceX may instead deploy its own satellites.
In other words, SpaceX can replace missing external demand with internal demand.
That's an extraordinary advantage for an infrastructure business.
Imagine a shipping company whose trucks rarely need to sit empty because whenever outside customers have nothing to ship, the company has millions of its own packages waiting.
That is increasingly what Starlink does for Falcon.
Now There Is Almost Too Much Demand
The relationship has become so powerful that it is creating the opposite problem.
SpaceX's own satellites are consuming so much Falcon capacity that some outside customers are struggling to find launch slots.
Reuters reported that at least seven space companies had been told Falcon 9 availability was effectively booked until 2028 or 2029.
Starlink is part of the reason.
The transition toward Starship also matters. As SpaceX prepares to move more of its own satellite deployment to the much larger vehicle, Falcon 9 capacity has become caught between strong outside demand, heavy Starlink usage and a launch system that is itself entering a new generation.
SpaceX increasingly has to decide what deserves a Falcon launch slot.
And the economics can favor SpaceX's own payload.
A third-party customer pays SpaceX once for a launch.
A Starlink satellite can potentially help generate connectivity revenue for years.
That changes what a Falcon launch slot is worth to SpaceX.
Cheaper Rockets Don't Necessarily Mean Cheaper Launches
This also explains a strange feature of the launch market.
Reusing boosters should reduce SpaceX's internal costs.
SpaceX itself says frequent reuse, manufacturing efficiencies and engineering improvements have reduced its internal launch costs.
But lower costs do not automatically mean SpaceX must give all those savings to customers.
Those are two different numbers:
What a launch costs SpaceX.
And:
What a customer pays SpaceX.
If demand for Falcon 9 is strong and launch slots are scarce, SpaceX can keep some of the economic benefit created by reuse.
Reuters reported that Falcon 9's advertised price increased from roughly $54 million in 2013 to around $74 million today, even as SpaceX became dramatically better at reusing its hardware.
That doesn't mean reuse failed.
It may mean the opposite.
SpaceX created a more efficient product while also creating enough demand that it did not need to pass every efficiency gain to customers.
This Is Why Copying Falcon 9 Isn't Enough
Suppose another company successfully builds a reusable rocket tomorrow.
It lands perfectly.
It can fly again.
Has it copied SpaceX?
Not really.
It still needs customers.
It still needs launch pads.
It needs manufacturing scale.
It needs enough missions to keep its rockets flying.
It needs enough experience to make turnaround routine rather than experimental.
And ideally, it needs something valuable to launch when outside customers don't show up.
SpaceX already has that something.
Starlink.
This is why SpaceX's advantage cannot be understood by looking at Falcon 9 alone.
The rocket and satellite network reinforce each other.
SpaceX Built Both Sides of the Market
In 2025, SpaceX completed 170 missions across Falcon and Starship.
It launched more than 2,200 metric tons into orbit.
According to SpaceX, that represented more than 80% of all mass launched to orbit worldwide that year.
That is an extraordinary number.
But it becomes easier to understand once you realize SpaceX is not simply competing for the world's existing launch demand.
It helped create a huge new source of demand itself.
Most rocket companies sell access to space.
SpaceX does that too.
But SpaceX also owns one of the largest reasons to go to space.
That changes the business.
A competitor cannot necessarily reproduce SpaceX's advantage by building a cheaper rocket.
It may also need enough demand to keep that rocket flying.
SpaceX spent years building both.
The rocket.
And the reason to launch it.
Starlink did not simply benefit from cheap SpaceX rockets.
It helped make frequent SpaceX launches economically useful.
SpaceX didn't just build a reusable rocket.
It built its own reason to keep reusing it.
BEYOND THE OBVIOUS.
Sources
SpaceX — SEC Registration Filing
SpaceX — 2026 Investor Materials
SpaceX — Segment Financials and Investor Presentation
Reuters — SpaceX's Satellite Ambitions Squeeze Out Rivals Reliant on Its Rockets