The Sample of Integrated Labour Market Biographies (SIAB) is the most frequently requested data set provided by the Research Data Centre (FDZ) of the Federal Employment Agency (BA) at the Institute for Employment Research (IAB). However, preparing the SIAB for scientific analysis is a complicated and error-prone task. This article explains steps necessary to prepare the latest version of the SIAB (7519 v1) and gives examples of how the preparation can be done. Among other things, it shows how to impute right-censored wages, deal with parallel employment episodes, and clean up the data set. The supplementary material to this article contains extensively annotated Stata do-files to replicate our data preparation.