When they were talking about the Starship launch the other day, it hit home for me how much more practical knowledge you can learn about your device when it survives use and you can disassemble it to study. Especially in these days where a device that can analyze metal for microscopic cracks is cheap enough that every airline maintenance hub has one on hand.

This fact essentially, and I think somewhat ironically, makes single use rockets harder to do design iteration on than ones meant to last for years.

Presumably the newest rockets will have higher thrust to weight ratios, or lower risk of failure, so the older ones will be priced as the cheap seats for as long as they last and the opportunity costs aren’t too high. Then they’ll either be parted out or put on display - land an EOL stage at every new spaceport and put that out front as a billboard.