Brand image measurement is the most fundamental building block of strategic marketing decisions in branded industry. Therefore, brand image studies are both regularly conducted and well researched. Nevertheless the measurement tools used are typically not constructed with the scientific rigour needed to generate the most informative results. The aim of this article is to evaluate the effect of one potential weakness: questionnaire length. It is investigated whether questionnaire length influences (1) the initial response rate of assigning attributes to brands, (2) the repeat rate of doing so, and (3) the empirical generalisation proposed by Dall’Olmo Riley et al. (1997) according to which there is a constant relationship between repeat rates and the initial response level.