There are many things that are much easier the second time around (and yes, teaching tops that list). It is also easier to write talks, abstracts, and parts of papers/proposals now that I have some already worked material to draw from for background. Doing slides for talks takes half the time, since I have my cartoons done (for the most part), and can just tweak and update the data. I am much, much faster at reviewing stuff, since I have less time to do it in. I am much better about declining to review the truly crappy papers (now that I am better at spotting them from the abstract alone). This is more thanks having to grade students' written work, than to more experience with reviewing, since I did that too at National Lab.
There are also many things that are harder. I am teaching a full load now, and have more service obligations, so time management is an issue. I am finding it really hard to balance being a good mentor to my four students with all the other demands on my time. There is a HUGE difference between two students and four students. I know I need to take one more next year (especially if one of my proposals in a new area for me comes through), but I don't know where the time to mentor them will come from. Happily, my "senior" students are really strong, and can take on much of the nitty gritty training stuff, but that only buys so much time... I am freaking out about money, since I need more and it is hard to get. I worry about what happens when my "starter" grants run out, since it is even harder to compete with the full pool (rather than the newbies).
Things that ended up working better than expected:
1. Taking undergrads over the summer. Not only were they very productive in the lab, but they let my grad students get their feet wet with training and mentoring in a low stakes, controlled way.
2. The burn the ships strategy with my startup funds. We have all our essential equipment up and running, and I have enough money to cover (barely) operating with four students for next year. My students are looking quite productive lately, so it looks like sprinting will pay off in publications. I don't have much of a monetary safety net, so perhaps the jury is still out on whether this was good long term.
3. Serving on a review panel. This really, really helped my proposal writing, in ways I never expected. I always knew, but never really appreciated fully, how important clarity is in grant writing. Clarity beats everything else (except for good ideas).
Things that didn't work out as well as I'd have liked:
1. Taking undergraduates during the academic year. I am not sure that the effort we put in is worth the productivity we got back. It takes a lot of time to train and mentor someone, and students taking courses have a LOT more pulling on their time than summer students do. It is really hard for them to balance time in the lab vs. their academic work, especially since the academic work comes with deadlines, while the research does not.
2. Relying on collaborators to provide essential materials. Some of these collaborations had been ongoing for years, so it certainly wasn't malicious or deliberate. It is just that no one else cares as I do about making sure we have materials in a timely manner. A minor glitch in a collaborator's lab might lead to a 3-6 month delay for them. Not a big deal (or at least it wasn't when I was at National Lab, with lots of resources). In a limited resource situation (like when starting up a lab), this could have been fatal. My students sit idle, no work gets done, and that means no publications!
3. Being an accessible professor. Ate up a bunch of my time, didn't improve student outcomes OR my teaching evaluations. Next year, I will stick strictly to my scheduled office hours for helping students outside of class, and I will limit the time I spend on emails for the class.