This paper sets out to test three hypotheses aimed at identifying features of efficiency and non-efficiency in interpreters’ notes for consecutive – defined here as note-taking features that correlate with high target-text quality, on the one hand, and with poor target-text quality, on the other. The hypotheses, which were generated on the basis of empirical findings from a pilot study, are as follows: (1) the more the notes, the better the target text, (2) the more the abbreviations/the fewer the full words, the better the target text, and (3) the more the notes in the source language/the fewer in the target language, the better the target text. The hypotheses are tested on data comprising the target texts and notes produced by five interpreters working from Spanish into Danish.