Self administered surveys can be conducted over mobile web capable devices. The literature is scarce on design of mobile web surveys. Often, methods developed for an established mode of data collection are applied to a new mode. However, many of these methods may be inappropriate for this mode. Mobile web surveys have unique features, such as administration on devices with small screens and keyboards, different navigation, and can reach respondents in various situations, factors that can affect response processes and resulting survey estimates. While informed by findings from other modes, experiments are needed to develop methodologies for mobile web surveys. A probability sample was drawn and a household member randomly selected to participate in a mobile web survey panel. Ninety-two participants were provided with a smartphone with voice and Internet plans and asked to complete a short web survey every week. A series of design experiments were launched through the weekly surveys. Experiments address three major objectives. First, we test fundamental findings found robust across other modes, but whose impact may be diminished in mobile web surveys. Second, we test findings from experiments in (computer administered) web surveys. While mobile web surveys share much of the functionality of web surveys, such as the ability to present images and to control the number of questions per page, both situational and device differences can lead to lack of similarity. Third, experiments were devised that directly address the unique display, navigation, and input methods of the mobile device. We conclude with practical implications, cautions, and suggestions for future research.