This paper explores a method of analysing large volumes of geographic data by combining visual and numerical components. For great parts of the work, we use an instance of CartoDB, a tool that en- ables the handling and analysis of such data with only little programming knowledge, and tailor it to our needs. Our method has applications in researching the behaviour of the mari- time transport sector, e.g. in response to policy changes. In this instance, we examine the MARPOL regulations for emission control areas that came into effect in North America in August 2012. We identify the num- ber of ships that is affected by the regulations and whether there occur any significant operational changes using speed as a measure. For traceability and transparency, we walk through the individual steps required to produce those results, which adds much needed weight when used in policy debates.