Objective: To compare the effect of traditional and ¿baby-led¿ breastfeeding advice on early infant weight gain and exclusive breastfeeding rates. Design: Longitudinal cohort study: part prospective, part retrospective. Setting: One UK general practice. Participants: 63 exclusively breastfed infants in two cohorts: 32 babies born before and 31 babies born after a change in breastfeeding advice. Intervention: A change from baby-led to traditional breastfeeding advice. Main outcome measures: Primary analysis: comparison of the effectiveness of the intervention (ie, weight gain expressed as standard deviation score gain (SDSG) between birth and 6¿8 weeks) and exclusive breastfeeding rates between babies whose mothers received traditional advice and those whose mothers received baby-led advice. Secondary analysis: relevance of feed length (ie, weight gain expressed as SDSG between birth and 6¿8 weeks in babies feeding for 10 min or less from the first breast and those feeding for longer than 10 min). Results: The two groups were equivalent with respect to birth weight, gestational age, and parity. Primary outcome: babies whose mothers received the traditional advice were more likely to be exclusively breast fed up to 12 weeks (log rank ¿2¿=¿9.68, p¿=¿0.002) and gained more weight up to 6¿8 weeks than those given baby-led advice (mean SDSG 0.41 (95% CI 0.13 to 0.69) vs ¿0.23 (95% CI ¿0.72 to 0.27)). Secondary outcome: irrespective of feeding advice given, babies feeding for 10 min or less from the first breast gained more weight by 6¿8 weeks than babies feeding for longer than 10 min (mean SDSG 0.42 (95% CI 0.11 to 0.73) vs ¿0.19 (95% CI ¿0.64 to 0.26)). Conclusions: In this study, traditional breastfeeding advice resulted in increased weight gain and increased exclusive breastfeeding rates compared with baby-led advice. Exclusively breastfed babies who had shorter feeds (10 min or less from the first breast) gained more weight.