As more and more companies recruit employees internationally, the personality instruments used in the selection process need to be available in various languages. Equivalence between language versions is essential if an instrument is applied in this context but cannot be automatically assumed. Instead statistical methods for detecting differential item functioning (DIF) need to be applied. Results from these analyses can provide evidence of language equivalence on the item level. This paper demonstrates how results of empirical DIF analysis based on two language versions of a personality questionnaire can be used to inform subsequent translations of the respective instrument. For this purpose, findings from a comparison of the English and German versions of the Occupational Personality Questionnaire (OPQ32n; N=967/361) are used. After briefly explaining the concept of uniform and non-uniform DIF, typical reasons for the occurrence of both kinds of DIF are explained. Based on item examples with DIF occurrence it will be shown how a good understanding of the different types of DIF and their causes can inform future adaptation processes and hence increase the quality of new target language versions of the instrument. A structured item review process as part of the translation procedure is suggested. This process includes a careful consideration of the likelihood of uniform and non-uniform DIF for each item based on what has been learnt from previous empirical DIF studies of the instrument.