OBJECTIVE : We assess outcomes of endoscopic orbital decompression for Graves' ophthalmopathy. METHODS : A review of endoscopic orbital decompressions of the medial and partial inferior wall between July 2004 and July 2017 was carried out. Outcome was assessed by comparing pre- and post-operative measurements of exophthalmometry and visual acuity. Results were evaluated by repeated measures analysis of variance. RESULTS : A total of 41 orbits in 25 patients underwent endoscopic orbital decompression for Graves' ophthalmopathy in the time period; however, six orbits in three patients had insufficient data for inclusion. Eleven patients required concurrent septoplasty to allow access. Measurements were taken at a mean of 11 days, 32 days, and 95 days post-operatively. Reduction in mean proptosis was 2.81 mm at 1-month post-decompression and 3.26 mm at 3 months. There was no significant difference between those treated for compressive optic neuropathy compared with those treated for cosmetic reasons. Colour vision by Ishihara plate improved significantly by a mean score of 2.67 post-operatively. Using LogMAR conversion for visual acuity, measured by a best-corrected Snellen chart, improvement of 0.18 was achieved at 1-month post-decompression, equivalent to approximately two lines on the Snellen chart. There was minimal (0.04) further improvement at 3 months. The improvement in visual acuity was greater in cases treated for compressive optic neuropathy than cosmesis, but this did not reach statistical significance (p = 0.06). Three cases required revision surgery. Diplopia disimproved or developed in four cases and squint surgery was required in three cases. CONCLUSIONS : Endoscopic orbital decompression offers an effective, safe and minimally invasive treatment for Graves' ophthalmopathy. There is a trend towards continued improvement in outcomes over the course of 3 months post-operatively.