Building many identical satellites in an assembly-line like fashion cuts down on costs significantly. Large geostationary satellites are currently based on standard buses (like the Boeing 702 and its variants), but each satellite is generally a bespoke product. Larger quantities of the same thing like the second generation Iridium network and the o3b MEO satellites are very interesting from a per-unit cost perspective.
Right. It definitely looks like we could be in for a shakeup in the industry. The question still remains though; what is the total size of the market for all of those satellites? How many could we place in MEO/GEO/GSO before the market is completely saturated? I honestly have no idea. I would think the LEO satellite market will eventually be dominated by less powerful vehicles, although if SpaceX can survive long enough to open up other markets (like Mars tourism and mining), maybe those smaller rockets get cannibalized by payloads being co-launched on larger vehicles.
