AIMS To compare the efficacy of remnant ablation following a single low dose (specific activity of 131I administered, 1074–1110 MBq) vs. a single high dose (mostly 2775–3700 MBq) of 131I in patients with differentiated thyroid cancer and to determine whether or not the extent of surgery influences outcome. METHODS Nineteen studies have reported the results of low dose 131I ablation. Of these, 11 met our criteria for a comparative analysis. Two additional cohorts of ours were added and these were analysed in two groups based on the extent of surgery (near‐total NT; Woodhouse1 vs. sub‐total ST; Woodhouse2). There were 518 low dose and 449 high dose patients in all. RESULTS The average failure of a single low dose was 46 ± 28% (SD). Meta‐analysis revealed a statistically significant advantage for a single high over a single low dose and a pooled reduction in relative risk of failure of the high dose of about 27% (P < 0.01). From this we estimate that for every seven patients treated one more would be ablated given a high rather than a low dose (assuming a low dose failure risk of 50%). Also, a significantly greater proportion of patients are ablated after a single high or low dose, if they underwent near‐total as opposed to sub‐total thyroidectomy (summary relative risk (RR) 1.4; P < 0.05). CONCLUSION High dose 131I is more efficient than low dose for remnant ablation particularly after less than total thyroidectomy. Results suggest that patients with differentiated thyroid cancer should routinely have a total thyroidectomy followed by high dose 131I (2775–3700MBq) for ablation of the remnant.