Poor retention of participants in weight management interventions is frequently recognised as an issue hindering study completion and power for statistical analyses, potentially leading to Type II error. This study aimed to explore retention issues in a gestational weight gain (GWG) intervention to identify opportunities to improve participant retention. Seven hundred and fifty-two pregnant women were approached at their booking appointment at a large antenatal clinic Melbourne, Victoria between August 2011 and August 2013. Two hundred and sixty-one women (34.7%) agreed to take part in this randomised controlled trial to prevent excessive GWG comparing health coaching (HC) to education alone (EA). At 18 weeks gestation, women were randomised to participate in HC/education (n=130) or EA (n=131) sessions. During the study period, retention strategies included the provision of a participant resource pack, frequent contact via phone and text message, and flexibility of appointment times. At 32 weeks gestation (post-intervention follow-up), 35.4% of the HC group and 26.7% of the EA group had dropped out of the study. Reasons given were: lost contact (HC: 67%; EA: 68%), not enough time (HC: 17%; EA: 7%), lack of desire to continue (HC: 7%; EA: 0%); parking/travel issues (HC: 4%; EA: 0%), personal reasons (HC: 0%; EA: 14%), medical reasons including premature delivery and pregnancy loss (HC: 2%; EA: 6%), and moved interstate (HC: 2%; EA: 0%). Future research should consider the modifiable reasons for participant dropout during an intervention (i.e., loss of contact and practical issues). Additionally, understanding the motivations of women to remain in obesity prevention interventions may lead to the development of additional retention strategies. Considerations include personal reasons for weight loss/maintenance; support from significant others; meeting participant expectations in intervention delivery and outcomes; management of competing demands on participants’ time and energy; and key stakeholder investment in the intervention.