Background The standard for sentinel lymph node biopsy (SLNB), the dual technique (radiolabelled tracer and blue dye), has several drawbacks. A novel magnetic technique without these drawbacks has been evaluated in a number of clinical trials. It uses a magnetic tracer and a handheld magnetometer to identify and excise sentinel lymph nodes. A systematic review and meta-analysis was performed to assess the performance and utility of the magnetic in comparison to the standard technique. Methods MEDLINE, PubMed, Embase and the Cochrane online literature databases were used to identify all original articles evaluating the magnetic technique for SLNB published up to April 2016. Studies were included if they were prospectively conducted clinical trials comparing the magnetic with the standard technique for SLNB in patients with breast cancer. Results Seven studies were included. The magnetic technique was non-inferior to the standard technique (z = 3·87, P < 0·001), at a 2 per cent non-inferiority margin. The mean identification rates for the standard and magnetic techniques were 96·8 (range 94·2–99·0) and 97·1 (94·4–98·0) per cent respectively (risk difference (RD) 0·00, 95 per cent c.i. –0·01 to 0·01; P = 0·690). The total lymph node retrieval was significantly higher with the magnetic compared with the standard technique: 2113 (1·9 per patient) versus 2000 (1·8 per patient) (RD 0·05, 0·03 to 0·06; P = 0·003). False-negative rates were 10·9 (range 6–22) per cent for the standard technique and 8·4 (2–22) per cent for the magnetic technique (RD 0·03, 0·00 to 0·06; P = 0·551). The mean discordance rate was 3·9 (range 1·7–6·9) per cent. Conclusion The magnetic technique for SLNB is non-inferior to the standard technique, with a high identification rate but with a significantly higher lymph node retrieval rate.