Interview assisted data collection modes are most often used to assess ego-centered networks owing to complex structure of such questionnaires. Web surveys, on the other hand, offer a variety of efficient solutions with sophisticated questionnaire interfaces and time-and-cost effectiveness. Single name generators for evaluating ego-centered networks can be effectively applied also withinWeb surveys, as shown in our experiments. However, visual elements of the measurement instrument have to be designed very carefully. In this study several experiments on Burt’s (1984) name generator were applied within Web questionnaire. Visual characteristics of name generator and instructions for respondents were varied. Effects of characteristics of measurement instrument and experimental instructions on number of listed alters, composition of network, and dropout rate, were studied. For collecting the data on alters two different techniques for measuring name interpreters were applied: alter-vise (complete information on name interpreters is obtained for each alter separately and the variable context is changing) and variable-vise (complete information on name interpreters is obtained for each variable and the alter context is changing). Both approaches were evaluated with regard to composition of networks and dropout rate.