The purpose of this study was to compare the two methods-guidewire localisation and the radioguided occult lesion localisation-used in the localisation and surgical removal of non-palpable breast tumours. This retrospective study enrolled patients diagnosed with nonpalpable malignant breast tumours. In this study either guidewire localisation (GWL, n = 69) or radioguided occult lesion localisation (ROLL, n = 321) was used for the detection and removal of the tumours. The two methods were compared with regards to preoperative localisation time, operating time, removed specimen volume, the pathological tumour size, the presence of positive surgical margins and postoperative complications. Furthermore, we have also investigated other factors that could have an impact on the frequency of positive resection margins. The localisation time was significantly shorter in the ROLL group, both with ultrasound guidance (5.7 ± 1.44 min vs. 21.6 ± 2.37 min, p = 0.05) and with radiographic guidance (21.8 ± 3.1 min vs. 41.6 ± 3.75 min, p = 0.021) as well. No significant difference was observed between the two methods in terms of operating time, removed specimen volume and pathological tumour size, or the presence of positive resection margins, or the occurrence of postoperative wound infections. The size of the tumour (ROLL, GWL grps), the presence of a multifocal tumour (ROLL grp), the presence of an extensive in situ breast carcinoma around the invasive cancer (ROLL, GWL grps) and the volume of the removed breast specimen (GWL grp) significantly increased the frequency of positive resection margins. We recommend the use of the ROLL method for the removal of nonpalpable breast tumours as it has a much shorter localisation time, and it is a simpler surgical technique as well.