Researchers often use police-recorded crime data to study offenders’ criminal behaviour, although police data do not include all offences nor all offenders: many crimes are not known to the police and only a small proportion of all registered crime are cleared. Several researchers have expressed concern about the generalizability and applicability of research findings only based on known offenders. Empirical research into unknown offenders is limited, precisely because of the difficulty of studying these offenders. However, forensic DNA databases offer opportunities. DNA profiles found at crime scenes can distinguish (not yet) arrested offenders and their DNA profiles can be linked to multiple crime scenes or co-offenders. This makes it possible to study the serial and co-offending behaviour of unknown offenders, with is impossible with police data. By using both police data and DNA data, I aim to advance the current understanding of serial and co-offending behaviour. This objective is achieved by addressing three research questions: (1) To what extent do unknown offenders have an impact on the features of serial co-offending networks? (2) To what extent does the spatial spread of crimes committed by serial co-offenders change when crimes of unknown offenders are taken into account? (3) To what extent does the temporal spread of crimes committed by serial co-offenders change when crimes of unknown offenders are taken into account?