The efficacy of deep brain stimulation of the subthalamic nucleus (STN) is dependent on the accuracy of targeting. In order to reduce the number of passes and, consequently, the duration of surgery and risk of bleeding, we have set up a new method based on direct magnetic resonance imaging (MRI) localisation of the STN. This procedure allows a short duration of the neurophysiological session (one or two initial tracks). Whenever a supplementary track is needed, the stimulation-induced side effects are analysed to choose from one of the remaining holes in Ben's gun. A good knowledge of anatomical structures surrounding the STN is mandatory to relate side effects to the actual position of the track. In our series of 11 patients (22 sides, 37 tracks), the most common and reproducible side effects were those characterised by motor, sensorial, oculomotor and vegetative signs and symptoms. Moreover, the therapeutic window (distance between the current intensity needed to obtain the best clinical effect and the intensity capable to induce side effects) predicted clinical efficacy in the long-term, and contributed to the choice of which among the examined tracks had to be implanted with the chronic macroelectrode.