Things I wish I knew before I started mentoring students

As newly minted scientists, we are not trained in how to be effective mentors. Nevertheless, in almost every research environment, mentoring becomes a significant portion of the job. This is especially true of TT positions in higher ed. Alas, my only training in being an effective mentor came on the job. I was thinking about this recently, since it is student recruiting season at ProdigalU. So here are some things I wish I knew when I started:

1. So many of my early decisions would be reactions to things I did not like about the way I was mentored. Sometimes consciously, sometimes not.
2. Even though #1 is true, it was hard not to replicate aspects of my advisors' mentoring styles, since that is all I knew when I started.
3. Although I knew (and was reminded by many, many people) that Prodigal as a student is not a good model to use in deciding how to mentor, it is really, really hard to act on this knowledge.
4. My first students (brave as they were to pick an advisor with no track record and no one to ask about) would have a HUGE impact on my mentoring style. (So recruit wisely).
5. Even though few people ever really have an idea of how to set up a group culture (and I was certainly in that boat), one will form anyway (even without my input) and it will stick around a long time. I really lucked out that my first set of students were serious, hard working, and easy going. They set the tone for the next rounds of students and on to today. I should have paid more attention to this, though I am happy with how things turned out in the end.
6. My students are my best recruitment tool (luckily for me, since I think I have great students!)
7. Just as my students will be linked to me forever, I will also be linked to them (see #5--hooray for great students). 
8. Mentoring is much harder than it looks. It takes quite a bit of experience to figure out the best way to mentor a particular person, and no one strategy works for everyone. I am still learning. Some personality types will never click, and that is no one's fault. If that happens, it is even harder to be a good mentor.
9. People need what they need, and sometimes that isn't me as a mentor, no matter what their science abilities are. If the mentor-mentee relationship is not working, it is best for all concerned to resolve the situation quickly. It does NO ONE any favors to pretend things are OK when they aren't, or that someone will get a PhD when they won't (at least not with me). I let things go on for way too long when this happened in my group.
10. Recruiting is just as much making sure I can work with the student as it is attracting students to my group. 
11. I ended up doing nearly as much mentoring about things outside the lab as inside (and not just career stuff either). I hadn't expected this.
12. That I would be so excited to hear from group alumni (undergrads too). I wish I stayed in better contact with my own mentors, now that I see how nice it is to hear about how things are going with my former students. I also wish I had told more of my students this when they were in my lab.