The documentation habit
For a long time, I used a laboratory notebook. I was taught that a lab notebook has two critical features. First, what goes in stays in. The notebook is a one-way record. It's not to be edited or "cleaned up". Second, things go in chronological order.
These two rules served me well during my undergraduate and graduate research training in biology and chemistry, and in my own research as a graduate student, and also in my job as a protein engineer. However, I've realized that the lab notebook is only one piece of the puzzle, if your goal is to become a highly effective scientist and engineer.
Write down how to do it first
The major thing I have learned to do—here at the 10-year mark in my scientific career—is to write down what I am about to do before I do it. And when I'm done, revise the plan so that I can do it again.
I call this "the documentation habit".
I started doing this primarily during my PhD training, where I worked in a diverse research lab with both experimental and computational scientists. I marveled at the similarities between a wet lab protocol and a data analysis script: both hard-earned, well-described algorithms for doing something specific with some inputs.
So I started writing a procedure for computational work too. In the software world, we call these tutorials, documentation, or user guides.
Now, whenever I have to do something more than once, I have a written document describing how to do it. A list of the inputs. A step-by-step description of exactly what needs to get done.
You can't improve what you don't write down
I've found this process critical for ensuring reproducibility, increasing my ability to teach and mentor, and enabling me to be a more efficient and effective communicator in the world of biology where things can get a little hectic.