Staphylococcus aureus is a major source of both hospital and community-acquired infections, and is the leading source of skin and soft tissue infections worldwide. The increased application of whole genome sequencing (WGS) has resulted in large volumes of S. aureus WGS data. The timely classification of S. aureus using genomic typing technologies has the potential to describe detailed genomic epidemiology from small to large scales. In this study, a multilevel genome typing (MGT) scheme was developed for S. aureus and was applied to 50,481 publicly available genomes. The MGT nomenclature consisted of a series of eight levels with progressively higher resolution. The application of MGT was showcased in three scenarios using previously published data. Firstly, the performance of a low-resolution MGT level (MGT2) was compared to a standard molecular typing method in S. aureus, Spa typing. Both methods provided equivalent levels of resolution and identified many of the same major types. Secondly, MGT was used to characterise a set of isolates from a study of a common methicillin-resistant S. aureus (MRSA) subtype (ST8-USA300) that had colonised multiple body sites of the same patient. STs from low to mid-resolution MGT levels (MGT2 to MGT6) uniquely described isolates from the same patient, and the highest resolution level (MGT8) could distinguish between isolates within the same patient that had different antimicrobial resistance profiles. Lastly, MGT was used to describe the transmission of MRSA in a hospital. Low to mid-resolution MGT levels (MGT2 to MGT5) were able to track isolates that had spread between wards and also isolates that had colonised different reservoirs within a ward. The S. aureus MGT describes large- and small-scale S. aureus genomic epidemiology with stable and standardised ST assignments and is publicly available (https://mgtdb.unsw.edu.au/).