INTRODUCTION: The manufacturer of subcutaneous implantable cardioverter defibrillators (S-ICDs) acknowledges that 'deep implants' may fail to elicit a magnet response, however, does not define 'deep implant' or recommend a maximum implant depth. This study aims to systematically evaluate the effect of subcutaneous tissue depth and magnet types on evoked magnet response. METHODS: Sunshine Coast University Hospital's S-ICD cohort underwent magnet response evaluation; where bar and donut magnets were compared and the evoked magnet response was recorded in three separate zones, guided by a template. Ordinal regression (OR) models assessed the relationship between the evoked magnet response and tissue depth (TD), measured via post-implant X-Ray. The patient's ability to hear the magnet response audible tone was recorded. RESULTS: Patients (n = 39) with measurable TD (n = 30) were analyzed. The bar magnet evoked a magnet response in all zones in 53% of patients, compared with 73% of patients with the donut magnet (p = 0.18). The relationship between bar magnet response and TD showed the odds of an evoked magnet response decreased by 11% every 1 mm increase in TD (OR of 0.89, p < 0.01), whereas the donut magnet decreased by 16% per 1 mm (OR of 0.84, p < 0.01). Directly over the S-ICD was the most effective in evoking magnet response with the bar (85% of patients), and off-centre was most effective for the donut magnet (100%). BMI and Praetorian score were not significantly associated with magnet response. We found 23% of patients were unable to detect the audible tone. CONCLUSION: We observed a statistically significant association between TD and ability to evoke magnet response. The bar magnet was less reliable than the donut magnet for therapy inhibition in deep implants.