New technology allows us to collect rich and voluminous data about refereeing--something that was not possible before. In this quantitative study, we consider data about the dates of sending (1) suitable papers to "BJET," (2) the papers to referees and (3) when the referees' reports are returned to the editor. To do this, we examined data from over 11 000 reports collected between 2005 (when "BJET" began using ScholarOne as its online editorial manager) and October 2014. We found that most of the submissions and responses were made in the earlier part of the week, with some evidence of academics (and the editor) working on Sundays. We also found modest agreement between referees on the quality of individual papers.