Observer-based studies often underestimate key ecological parameters. Here a fresh approach was used to analyse six years (2006–11) of attendance cycles to estimate foraging trip lengths of a lactating flipper-tagged otariid: subantarctic fur seals at Marion Island. Multi-state mark-recapture models were used to calculate detection failures of females, correct estimates accordingly, and investigate the effects of year, season, pup sex and the presence of a telemetry device on attendance cycle parameters. There were no differences between corrected and uncorrected attendance data. This is attributed to the high capture probability across all seasons (range: 83–98%). This illustrates that observer-based studies are useful to augment telemetry studies. Only season and pup sex had a significant impact on female provisioning rates. In winter, foraging trip durations were longer (t-value=25.22, P<0.0001) and attendance durations shorter (t-value=-2.15, P=0.01) than during summer. Females with female pups spent a higher proportion of their time on land (χ2=6.6, P<0.05). Male pups have higher growth demands and are larger which suggests they can deplete female milk-stores faster.