There is general acceptance in corporate and institutional quality assurance standards, such as ISO 17100, of process-reliant translation, which incorporates revision. There is nevertheless a general lack of consensus on whether revision by someone other than a translator improves a translation, and scarce research on how visible and valuable are the effects of revision to the end-user. Revision is a key step in quality assurance, if only because every translator makes mistakes. The invisible reviser and their influence on final translations are nevertheless absent from research and theoretical discussions, to the extent that the reviser, more so than the translator, has been described as the “truly invisible” participant in the translation process. This research surveyed Australian-based translators in the Japanese-English language pair to find out their experience in and attitudes to revision. Ten translators from among over 50 respondents were then asked to complete a revision task, the results of which were rated and ranked by over 60 English first language readers. The interim results provide some insight into unanswered questions about the efficacy and visibility of revision, and enable some comment to be made on approaches to revision taken by same and opposite language revisers and on translator and end-user perceptions of what is important in translated text. While not part of the original research, the sudden prominence of ChatGPT and its rewriting ability in the latter stages now provoke the question whether the technology will extinguish the reviser before her or she has had a chance to become visible.