Background: Despite the growing interest in school-based mindfulness training (SBMT) and its evaluation, evidence on the effectiveness of SBMT is still limited and somewhat inconsistent. Further, knowledge on the role of implementation quality, which is essential for a more widespread use of SBMT, is scarce. Objective: This study examined effects of two established 8-week SBMTs and the role of implementation quality. Interventions were hypothesized to enhance emotion regulation, social well-being, and emotional well-being. Furthermore, higher responsiveness of participants and quality of delivery were hypothesized to be associated with better program effectiveness. Method: Nineteen elementary school classes (second through sixth grade) from German-speaking regions of Switzerland, with 246 students (aged 9 to 12 years), chose to implement either intervention and were then randomly assigned to the intervention or a waitlist control group (cluster-randomized). Results: Multilevel modeling revealed that SBMT enhanced some aspects of social well-being (self-reported social participation), and further partially stabilized emotion regulation (self-reported anger control). However, there were also unexpected effects (enhanced self-reported stress vulnerability and hiding of emotions, as well as reduced parent-reported social participation and prosocial behavior). Effects were small, quite robust when controlling for covariates, and not consistent across informants and SBMTs. No robust associations between higher quality of delivery and better effectiveness of SBMT were found, and higher responsiveness of participants was only associated with higher anger control after the intervention. Conclusions: We discuss in-depth the reasons for unexpected findings and provide directions for future research.