Sometimes research is hard to scale. What would you do differently? Build variations of devices ahead of outcome, so that additional experiments can be execute

Sometimes research is hard to scale. What would you do differently? Build variations of devices ahead of outcome, so that additional experiments can be executed almost in parallel? Prioritize the research equipment delivery on a planetwide level? Get the state to sponsorthose cushy wages and pensions nuclear scientists had, back in the 70s?