Past studies of performance indicators for libraries have concentrated on theoreti cal considerations at the expense of prac tical measures applicable without undue effort in normal working situations. Many indicators advocated are highly valid but are too expensive in relation to their util ity. Libraries in some countries have tried to agree a total set of indicators before starting to apply any of them, while many individual libraries have instituted their own measures, which are unlikely to be totally consistent with those used by other libraries. A pilot project, conducted at the Copenhagen Business School Library, Denmark (CBS) aimed to approach the matter empirically, starting with a set of measures, testing them in one or more libraries, and then seeking agreement between libraries of the same type, in order to establish a standard set of mea sures that may be applied throughout the country. The pilot evaluation tested most of the proposed indicators satisfactorily; indicated those that were most promising; and confirmed many of the suspected strengths and weaknesses of the CBS Library, but putting them on a firmer quantitative footing. The study also pro duced unexpected findings and pointed to remedial action needing to be taken at once. A full report of the evaluation has been issued separately in Danish and English versions (Copenhagen Business School Library, 1992).