Seabird populations breeding in the UK Overseas Territories remain relatively understudied compared to UK seabird populations, despite their international importance. Here we present results from one of the first seabird tracking studies in the Caribbean region, of Brown Boobies Sula leucogaster breeding on the Important Bird Area (IBA) of Dog Island, Anguilla. Birds were tracked for 5–7 days during the chick-rearing period using GPS data loggers. We assess how representative the at-sea areas of use (utilisation distributions) identified from our sample of 16 birds are likely to be of those of the whole breeding colony, and examined the effect that grid cell size used in the ‘time-in-area’ analytical approach has on these predictions. We also assess the effectiveness of the BirdLife International’s seaward extension approach to marine IBA designation, where terrestrial IBAs are buffered a set distance using existing information on the foraging radii of the same or similar breeding species. Foraging trips were 125.3 ± 54.4 (SD) km long and lasted for 5.6 ± 1.95 hrs on average. Birds travelled into the waters of four neighbouring territories; Saint Martin, Saba, Saint Eustatius and Saint Barthelemy. Our models suggest that many more individuals would need to be tracked to fully identify important at-sea areas for this colony, although this depends on the scale that important areas are defined. Whilst a smaller grid cell size may be necessary for assessing fine-scale habitat use, a larger grid cell size may be more appropriate for marine spatial planning processes. Although the BirdLife Seaward extension approach using maximum foraging distance recorded from Brown Boobies at a Mexican colony predicted a smaller foraging area than that used by Dog Island birds this approach still incorporated at least 99% of their 50% UD, 98% of their 75% UD and 86% of the 95% UD.