Web-based graphical user interfaces (GUIs) are mostly not tailored for small devices with touchscreens, such as smartphones. There is little scientific evidence on the conditions where additional taps for navigation are better or scrolling. Therefore, we conducted a user study in which we evaluated different ways of tailoring a GUI for a smartphone. Each participant performed the same task with two different layouts of the same GUI. We collected quantitative data through measuring task completion time and error rates, as well as qualitative data through subjective questionnaires. The main result is that minimizing the number of taps is important on a smartphone. Users performed significantly better when they could scroll (vertically), instead of tapping on widget elements (tabs). This preference was also reflected in their subjective opinions.