A lot of the invested time is 'pre-loaded' to an earlier point in time and if you continously apply them it should not feel like a huge effort at the end ('growing pains' when changing the infrastuctre)
If it takes more time, but the studies are objectively better the time investment should not matter that much; might be good to have an example ready.
You can use extra output (e.g., published datasets) as an academic output on your CV
It is part of your work as a researcher (might not work).