Researchers sometimes view clinical trial registration as a mere formality rather than scientific protection. Late registration after enrolling participants, even by mistake, undermines the credibility of the research because readers cannot know whether the results or methods were modified before data collection. In the study (doi:10.1136/bmj-2025-086467), the authors found that journals do not always publish information on whether a study was retrospectively registered, and authors often do not report it. Many readers are unaware of this shortcoming or unable to assess its impact on the interpretation of the study. Registration is an essential part of the integrity of the research process. As of 2018, the BMJ has kept a record of submitted studies that appeared to be unregistered, retroactively registered, or similar.