Results from IP5 suggested that a mixed mode approach affects response rates. A small qualitative follow up study was undertaken to understand the role that mode played in panel members’ decisions about whether to continue to take part and in particular the role of the web instrument in this process. A range of factors were found to influence IP5 panel members’ decisions to continue to take part in the study. Some factors were ‘generic’, however there were other factors that were specific to the sequential mixed mode design of IP5 and the initial invitation to take part by web.