In this study we investigated the robustness of the WelFur welfare assessment system for farmed mink (Neovison vison) to date of assessment in the winter and growth assessment periods. The prevalence of occurrences of certain measurements was hypothesised to increase with date of assessment (too thin, fur-chewing and stereotypic behaviour in the winter period and injuries, diarrhoea and exploratory mink in the growth period). The welfare was assessed on eight Danish mink farms according to the WelFur-Mink protocol. Each farm was assessed once in the nursing period (to be able to calculate WelFur-Mink scores), four times in the growth period and three times in the winter period. WelFur scores were calculated based on the assessments in the three periods: one calculation for each assessment in the winter and growth periods. The odds of fur-chewing increased with date of assessment in the winter period, and the odds of injuries, diarrhoea and exploratory mink increased with date of assessment in the growth period. The odds of too thin mink in the winter period decreased, ie the change was in the opposite direction to what was expected. The effect of these changes on the aggregated WelFur scores on the higher levels was limited, but could potentially lead to changes in the overall welfare categorisation of farms if the principle scores were close to a threshold between two categories. A potential way to eliminate the effect of date of assessment could be to develop a correction factor for the measurements that can be expected to change within each assessment period.