Cancer registries are used to compare incidences between regions, plan for service provision, and to assess the impact of health interventions. Significant variation in data capture for skin cancers and reporting of results was evident between regional cancer registries in the UK when assessed in 1991. Using a postal questionnaire we sought to document methods of recording skin cancer incidence in the UK in 2000, and to assess if practice has changed from 1991. All UK cancer registries were asked for details of their method of skin cancer case registration and latest available incidence figures. Methodology was assessed against recently implemented national standards. All registries responded to the survey. Sources of data were more uniform than was the case 9 years ago. All registries except one attained national standards for basal cell carcinoma data collection, but only half of the registries attained standards for squamous cell carcinoma. Ten of the 12 correctly recorded numbers of malignant melanomas, but three still failed to record the Breslow thickness or Clarke's level. Wide variation is evident in the recorded incidences for each of the malignancies, and the efficiency at which figures are made available. Thus, although there has been improvement since 1991, variability still exists between UK registries in methods of data capture, the data recorded, and efficiency of data processing in skin cancer registration.