We think there are four steps to using PRECIS-2, which may be iterative depending on what you discover after going through the steps.
Your first step is to be clear about why you are doing your trial. Are you:
The first would be an explanatory approach to designing a trial, the second a pragmatic approach. (Although this is not a dichotomy, there will be a continuum from pragmatic to explanatory). Both types of trials have their place but trialists should be clear which path they are on. As Schwartz and Lellouch pointed out, trialists have often taken the first approach by accident.
This step is explained in more detail for each domain later on.
A table can be used in conjunction with the PRECIS "wheel" or instead of the wheel to give rationale for scores. You can use this to assist discussion with trial collaborators.
Review your design choices (Step 2) on the PRECIS-2 wheel to see whether they will produce a trial that will support the aim identified in Step 1. Go back to Step 2 and modify your choices if required.