In the case of NASA, they could stand to shed a lot of the institutional structure they’ve built up. When institutional knowledge becomes more about navigating bureaucracy than doing engineering, something needs to break.

My dream outcome would be that they shrink down the bureaucracy massively, retire a lot of the career middle managers who make sure nothing ever moves too fast and no project is allowed to run lean, and then raise pay for the remaining people to attract some of the private sector talent that could infuse some new knowledge. The few people I know with SpaceX experience are not at all happy with the work environment there, but the pay and pace of a place like NASA aren’t an option for them.