Elon Musk is serious about building a global satellite internet broadband business at SpaceX.
On June 4, 2020, the 10th anniversary of the original Falcon 9 launch, SpaceX delivered its eighth batch of "Starlink" broadband internet satellites to orbit. Barely a week later, it launched the ninth batch on Saturday -- 58 Starlinks, with a trio of Planet Labs' "Skysats" tagging along for the ride.
Assuming most Starlink launches carry 60 satellites as payload, and subtracting a few for attrition, there are probably close to 540 Starlink satellites now in orbit around Earth. This means that, as of today, Elon Musk is two-thirds of the way toward his goal of getting an initial 800 satellites orbiting Earth -- the benchmark for being able to deliver "moderate" internet coverage all around the world.
Ultimately, of course, SpaceX wants to have as many as 42,000 satellites in orbit, but ... baby steps. First he's got to get to 800, then prove that the satellite internet system works as promised -- providing seamless coverage of internet service to every square foot of Earth's surface, at low latency speeds comparable to what you would expect from a cable internet provider.
SpaceX's Starlink internet constellation could one day provide internet service to anywhere on Earth. Image source: Getty Images.
The future is internet
Why is this important to SpaceX? Two reasons:
First, because when Starlink is up and running, SpaceX expects the revenue from operating its internet broadband satellites will eclipse the revenue it gets from launching other companies' satellites. As early as next year, SpaceX internal documents show satellite internet service could generate more than 50% of total SpaceX revenue. By 2025, the company expects to get 85% of its money from internet service -- at operating profit margins of 60%!
But second, even before this happens, SpaceX has a chance to win part of a $16 billion Rural Digital Opportunity Fund (RDOF) being handed out by the Federal Communications Commission.
Designed to provide "gigabit speed broadband networks" to "unserved rural areas" of America, RDOF will allocate its $16 billion among bidders that could include both traditional wired telecoms such as Comcast (CMCSA -5.82%) and CenturyLink (LUMN -6.20%) as well as satellite providers such as SpaceX or Amazon.com's Project Kuiper. FCC intends to dole out the money over 10 years to winners of a reverse auction, in which bidders offer to provide low-latency service for low prices. The lower the price offered, and the lower the latency a company can provide, the better that company's chances of winning.
How SpaceX's chances just improved
In proposed reverse auction rules released last month, the FCC expressed some doubt about SpaceX's ability to offer low-latency service, noting that other satellite operators (albeit operators that orbit their sats much farther from Earth than SpaceX's proposed 340-miles-up altitude) suffer as much as 600ms in lag.
Latency, or "lag", is the time between when you send a command on your computer and when you see something happen in response to that command. When latency is 100 milliseconds or less (i.e. one-tenth of a second), "interactions feel instantaneous," says Gmail creator Paul Buchheit. Thus, 100ms is the standard FCC has set for good broadband.
With its Starlink satellites orbiting so much closer to Earth, though, SpaceX is promising latency "below 20 milliseconds," on par with latency from a wired internet provider on Earth. The FCC initially doubted this assertion, but last week the FCC confirmed it will give SpaceX a chance to prove its capability.
In a statement, FCC Chairman Ajit Pai promised to apply "very close scrutiny" to SpaceX's claims, and said the FCC will not allocate money "to fund untested technologies." Nevertheless, he confirmed that "LEO service providers [such as SpaceX can] apply to bid in the low-latency tier instead of limiting them to the high-latency tier."
In short, SpaceX will get its shot at winning some or all of the $16 billion.
Clearing the hurdle
And this is where SpaceX's rapid-fire satellite launches become important. The faster Musk, and SpaceX, can get these satellites into orbit, the more and better data they can generate to prove Starlink's ability to provide low-latency broadband internet service -- and SpaceX's right to compete for the $16 billion.
In this regard, SpaceX noted on a recent Reddit discussion that it was clocking "5 trillion bytes of data" (five terabytes) exchanged between Earth ground stations and its Starlink network daily even before its latest launch, generating more and more data on the satellite constellation's performance -- and its lag rates. The company is also continually updating software on the satellites, and getting opportunities to both "debug" the software and optimize the satellites' performance.
What comes next
SpaceX intends to begin "private" beta testing of the Starlink network "later this summer, followed by public beta testing" later this year. Depending on how that goes -- and how quickly it goes -- it sounds like SpaceX could very well have all the evidence it needs to prove its capability in time for the FCC's auction, which is scheduled to begin on October 29.
At this point, I think it's safe to say the race for the FCC's $16 billion has begun -- and SpaceX is very much a part of this race.
