Nitrous oxide is produced in soil during several processes, which may occur simultaneously within different micro-sites of the same soil. Stable isotope techniques have a crucial role to play in the attribution of N2O emissions to different microbial processes, through estimation (natural abundance, site preference) or quantification (enrichment) of processes based on the 15N and 18O signatures of N2O determined by isotope ratio mass spectrometry. These approaches have the potential to become even more powerful when linked with recent developments in secondary isotope mass spectrometry, with microbial ecology, and with modelling approaches, enabling sources of N2O to be considered at a wide range of scales and related to the underlying microbiology. Such source partitioning of N2O is inherently challenging, but is vital to close the N 2O budget and to better understand controls on the different processes, with a view to developing appropriate management practices for mitigation of N2O. In this respect, it is essential that as many of the contributing processes as possible are considered in any study aimed at source attribution, as mitigation strategies for one process may not be appropriate for another. To aid such an approach, here the current state of the art is critically examined, remaining challenges are highlighted, and recommendations are made for future direction.