Skip to content

Comment on Ask HN: Did Google/Alphabet ever try to build a Starlink?

Comments

The bottleneck was always getting satellites up in space. SpaceX has an effective monopoly on launches at this point, so it isn't a surprise that they were the first ones to establish themselves in the satellite internet market.

It has nothing to do with the monopoly. They built rockets that cost about 20 million to launch and everyone else's rockets cost about 60 million to launch. This is internal costs for the launch providers. On pricing, what customers pay, the difference is hardly so stark because SpaceX doesn't leave money on the table without good reason given their superior reliability and human capability.

That cost win was because of reuse and reuse gave them a fleet. The launch market was inelastic and they had all this launch capability and not enough customers so they became their own customer with their excess capacity and now Starlink is successful enough that commercial launch will soon take a back seat to Starlink internal launches.

The ~$40 million dollars of savings that SpaceX saves per launch is only truly significant because SpaceX was so successful at trimming costs elsewhere.

That SpaceX was able to launch 60 satellites per launch vs competitor's 6 and was able to build them for $1/4 million per satellite vs a competitors $10M made a lot more difference in terms of constellation viability than the launch cost.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.