The term “Ambulatory Assessment” encompasses a wide range of methods used to study people in their natural environment, including momentary self-report, observational, and physiological methods. In the following, I will illustrate four major advantages of this approach: (a) Real-time assessments increase accuracy and minimize retrospective bias; (b) assessments in real-life situations enhance generalizability, (c) repeated measurements allow tracing of dynamic processes; and (d) multimodal assessments integrate psychological, physiological, and behavioural data. Regarding “promises”, I will report on studies assessing dynamic processes using Ambulatory Assessment. Several of these studies did reveal added value in comparison to more traditional assessment methods in predicting future outcome. Regarding “challenges”, I will demonstrate that the time based design is a special challenge and that the sampling strategy must fit the psychological process of interest. I will outline elaborate strategies, like activity-triggered e-diaries or experimental manipulation in everyday life, to increase within-subject variance in everyday life. To conclude, I will briefly discuss reactivity, generalizability and privacy issues.