For institutions intent on improving their research student outcomes, it is important to identify the variables most strongly associated with timely or tardy completions, which the university has the potential to influence or amend. For this to occur the analyses of doctoral completion times need to be conducted at an institution or discipline level. However, conducting meaningful analyses of doctoral completion data is a complex undertaking fraught with potential problems. This paper uses an intra-institution case study to reflect on and illustrate how different meanings can emerge from data analysis, depending on the statistical approach used to analyse the data. The paper outlines the lessons learned during data analysis, and the implications for doctoral education. The findings also point to potential marketing opportunities for research institutions.