The Covid-19 pandemic has had many negative consequences on the general public mental health. The aim of this study was to test the effectiveness of and satisfaction with an app with gratitude exercises to improve the mental health of people with reduced mental well-being due to the Covid-19 pandemic, as well as potential mechanisms of well-being change and dose–response relationships. A two-armed randomized controlled trial design was used, with two groups receiving the 6-week gratitude intervention app either immediately (intervention group, n = 424) or after 6 weeks (waiting list control group, n = 425). Assessments took place online at baseline (T0), six weeks later (T1) and at 12 weeks (T2), measuring outcomes (i.e., mental well-being, anxiety, depression, stress), and potential explanatory variables (i.e., gratitude, positive reframing, rumination). Linear mixed models analyses showed that when controlled for baseline measures, the intervention group scored better on all outcome measures compared to the control group at T1 (d = .24–.49). These effects were maintained at T2. The control group scored equally well on all outcome measures at T2 after following the intervention. Effects of the intervention on well-being were partially explained by gratitude, positive reframing, and rumination, and finishing a greater number of modules was weakly related to better outcomes. The intervention was generally appealing, with some room for improvement. The results suggest that a mobile gratitude intervention app is a satisfactory and effective way to improve the mental health of the general population during the difficult times of a pandemic.