Every year the Norwegian Directorate of Immigration (UDI) conducts approximately 10,000 interpreter-mediated asylum interviews in Norway. Each asylum interview ends with the interpreter providing a sight translation of the draft interview report (an oral translation of the text) to the asylum seeker. The UDI wished to find out why some interpreters took so much more time than others for sight translation. In response, Oslo and Akershus University College of Applied Sciences initiated a pilot project. The project’s working hypothesis was that slow sight translation was due mainly to a combination of three factors: interpreters’ poor reading skills; interpreters’ poor interpreting techniques; and/or interpreters’ unfamiliarity with the appropriate genre/style. A more complex picture emerged, however, from an analysis of 13 asylum interviews; group work involving 108 UDI interpreters; and discussions with UDI specialists. An asylum report is the joint product of all the participants in an asylum interview, who are typically the asylum seeker, the interviewer and the interpreter. The pilot study showed that each participant could contribute in various ways to slowing the speed of sight translation. Contributing factors identified that were unrelated to the interpreter included the interviewer’s competence in interview techniques; the interviewer’s competence in creating a written record of speech; the asylum seeker’s narrative ability; and the asylum seeker’s ability to correct mistakes in the draft report. All these factors may have just as much impact on the speed of sight translation as the interpreter’s competence. In addition, the physical strain caused by the length of asylum interviews (six hours plus) and the emotional stress potentially caused by such institutional interviews may affect the communicative abilities of all participants. As sight translation in the public sector is an under-researched area, the article starts with a discussion of terminology and an explanation of the UDI context. It then presents a possible explanation for slowness of sight translation and concludes that there is a need for more research into the various factors relevant to sight translation in the public sector and that there is a need to provide systematic training for both interpreters and interpreter users.