According to several authors, it is important for survey respondents to have an understanding of their progress within a questionnaire. This can be achieved by displaying a "progress indicator" indicating how much of the questionnaire has been completed. Empirical evidence on the effectiveness of such timers remains scarce and mixed (Conrad et al., 2003). A web experiment was set up to test the effects of progress indicators (n=2,520). The experiment included a number of factors that could potentially interact with the presence or absence of a timer (specific vs. vague time indications in the e-mail invitation and allowing a random subset of respondents to select whether or not to display the timer). A few observations could be made.Presenting the timer significantly reduced the item missing data rate, but break-off rates were not decreased. It was also found that it could be hazardous to allow respondents to select whether to display the timer, for not choosing the timer results in significantly higher break-off rates. Also, it was found that displaying a timer reduces the likelihood of participants saying the survey was too long, which in turn is negatively associated with the inclination to cooperate with a future study. Analyses of more exploratory nature revealed that timers tend to reduce the probability that respondents interrupt filling in the survey, which could indirectly increase data quality. Overall, it could be concluded that timers have limited abilities to increase data quality. However, more than three quarters of the respondents given the choice regarding timer display opted to have it displayed. This conveys that most respondents have come to expect timers and wish to receive feedback on their progress. It therefore urges future research to investigate how progress bars are best used in web surveys.