>>12987090>Does anyone know if $99 a month is actually profitable considering a decent adaption rate and sat replacement costs?
In terms of satellites and ground maintenance yes, it'll be (mildly) profitable. The big economic challenge right now actually is the terminal. Initial estimates were that it probably cost around $2000, not $500, though I remember seeing they'd managed to bring it to around $1000 as they scaled up and refined it. But they really need to get that down farther in mass production. It's not an issue in beta since the absolute volume is so limited, but if they want to avoid a cell subsidy model (which, granted, obviously can work) of "$500 with 2 year contract" it's an issue. I don't know how the whole chip and fabrication shortages might fuck things up for them also.
Some of the REALLY huge profit centers though are yet to come, since they require intersat optical links which are only at v0.9 and only deployed on solar sats right now. Those include commercial marine shipping, aerospace, mobile military, etc., places that right now pay through the nose for MEO/GEO sat links at best which are complete shit and that's it, they literally have zero other options. SpaceX could charge $1000-10000/month or more for some of these and it'd be a fucking BARGAIN vs what they pay now. Mature Starlink will also do lower latency between continents, so they'll be able to dip into the immensely profitable HFT sector.
Finally there's the long tail of poorer countries. Starlink is fixed cost and fixed maintenance for coverage, covering a continent/poles is the same as covering the world. Once they're up and running in America/Canada/Australia, maybe parts of Europe, every marginal dollar turns pretty pure profit after that if terminals get covered. Even if they charge $5/month to 3rd worlders (with some charity footing the bill for the terminal, or maybe a bunch of mudhuts split the cost or something) they'll still make money.