Two different hearing instrument fitting strategies were compared in the laboratory and in a field test with regard to the benefit for hearing aid users and their satisfaction with the fitting. DSL [input/output (i/o)] fittings based on hearing threshold and uncomfortable levels for the subject were evaluated vs a prescriptive fitting method based on unaided loudness scaling. Twenty-one subjects were fitted diotically with both fitting strategies implemented in a digital hearing instrument (Siemens Prisma). The patients tested both fitting strategies sequentially in a 4 + 4-week field trial using a crossover study design. Speech recognition threshold measurements, sound quality ratings and paired comparisons of sound quality were performed for all conditions (unaided, DSL[i/o] fitting and loudness-based fitting). In addition, subjective benefit and preference were assessed with questionnaires. Speech audiometry did not reveal significant differences between the two fittings. DSL[i/o] fittings showed superior results in most sound quality tests and in the self-assessment of communication abilities while the loudness-based approach was slightly preferred in noisy environments. The results seemed to be influenced by the higher gain predicted by DSL[i/o]. This study provides no evidence that effort spent on loudness scaling leads to improved fitting results.