Asking citizens “the way democracy works” is the basis of a wide literature on the support citizens have for their political institutions and is one of the most common survey items in political science. Moreover, it is a key indicator for the purported global decline in legitimacy. Yet, its trends, levels, and dynamics are still debated, and conclusions may be erroneous. In this paper, we compile a unique global dataset between 1973 and 2018 encompassing all major cross-national datasets and national election studies in twelve countries to study the dynamics and consistency of “satisfaction with democracy” (SWD) measures globally. Our results show that while trends and between-country differences in democratic satisfaction are largely similar, the levels of satisfaction vary substantially between survey projects, and both trends and levels vary significantly in several widely studied countries. We show that this has consequences at the individual level: opting for one survey over another may alter our conclusions about the relationship between key demographics and SWD. Thus, researchers studying SWD should endeavor to consult diverse survey sources and should be cautious about their conclusions when they do not, especially when it comes to making claims about changes in SWD over time.