Amazon Leo recently launched 29 new satellites, bringing the size of its constellation to 396. To the surprise of everybody who follows the industry, the company announced that it has enough satellites to begin initial service later this year.
This has everybody wondering what initial service means. The company hasn’t announced its pricing yet, which has been a closely held secret. The big mystery is how the company could launch service with such a small number of satellites.
When Starlink began beta service, it had over 600 satellites in orbit. Starlink beta customers reported slow speeds and spotty coverage, and said they routinely lost broadband service when there was no satellite in range of their home. Starlink didn’t start getting good reviews on its coverage until it had over 2,000 satellites in orbit.
A recent article in PC Magazine speculated that Amazon might launch by serving a narrow geographic band that would extend from Phoenix, across the U.S., and that enters into Canada in Michigan. The company might be able to use its small number of satellites to offer service in this narrow band as a way to work out any bugs. Amazon is not using the term beta test, but that’s what a limited launch of this nature would realistically be.
It’s hard to see Amazon Leo being a serious competitor to Starlink for the foreseeable future. By the end of this year, Starlink will have nearly all of it’s original constellation of 12,000 satellites in orbit, and Starlink currently has FCC approval to launch 15,000 total satellites. SpaceX recently asked the FCC for permission to grow the constellation to 100,000 satellites. By contrast, earlier this year, Amazon Leo asked the FCC for a two-year delay for meeting the requirement to have over 1,600 satellites in orbit this summer. When its first constellation is complete, Amazon Leo will have just over 3,200 Gen1 satellites in orbit, with an additional 4,500 that was approved by the FCC earlier this year.
Amazon Leo is starting with one edge, in that its first satellites have a lot more capacity than the current Starlink satellites already in orbit. But Starlink says it will begin launching the more powerful V3 satellites sometime in the second half of this year. The V3 satellites are said to have ten times the capacity of current V2 satellites. The V3 launch has been delayed, with the original plan to start the improved satellites in the first quarter of this year. The delay has mostly been due to the need to use the larger Starship rocket rather than the current Falcon 9 rocket to launch the heavier V3 satellites. It’s going to take a number of years for Starlink to replace the older V2 satellites, with an announced capacity of 60 V3 satellites per launch using the Starship rocket.
Perhaps the early announcement from Amazon Leo of pending service is meant to mollify the FCC and NTIA, which are both anxious to see satellite competition. Amazon Leo won preliminary awards to serve over 300,000 locations awarded in the BEAD grant program, and states and the federal government have to be worried about the company being able to fulfill that obligation.
The whole industry is waiting to see Amazon’s pricing and bundling strategy. There has been a lot of speculation that Amazon would bundle Prime memberships, AWS cloud services, or other bundles of non-traditional market goods and services. Such bundles would be unique, and something that Starlink and other ISPs could not match. Picture buying satellite broadband and getting free shipping and access to a huge video library as part of the bundle.
The biggest question is if Amazon Leo will offer prices below Amazon’s standard $130 per month (recently increased from $120). Doing so would put a lot of pressure on Starlink to lower its prices. Unfortunately, there is also the possibility that the two companies will have similar pricing and will form a duopoly of expensive satellite broadband.
The phrase “…and will form a duopoly of expensive satellite broadband” assumes broadband should continually get cheaper.
I’d argue the opposite: it will form a duopoly of realistically priced satellite broadband.
Nearly everything required to deliver internet service is becoming more expensive—labor, insurance, equipment, fiber construction, utility poles, electricity, taxes, and financing. Electricity itself, another service many consider essential, hasn’t become cheaper over time either.
People often claim broadband is getting cheaper because the cost per Mbps has fallen. I don’t find that metric meaningful for consumers.
My T-Mobile plan (4 phones) costs about $216/month and routinely delivers around 1 Gbps. By the “$/Mbps” logic, that’s just $0.216 per Mbps. But I only need about 50 Mbps most of the time. If internet were truly “getting cheaper,” shouldn’t my bill be around $10.80/month?
Consumers don’t pay their bills in Mbps. They pay them in dollars. If my monthly bill keeps increasing, it’s difficult to argue that connectivity is becoming cheaper simply because I’m receiving far more capacity than I actually use.