Reproducible workflow

How will you arrange your workflow such that all (major) decisions and research steps taken are documented and accessible for others? Being transparent about your own work can be challenging, but very important for later reuse, verification or replication. Examples include documenting deviations from planned procedures (log book), decisions to change procedures, annotating (in English) analytical syntaxes or other computer code.