<p>There is extensive debate about the effects of social media use on well-being. However, evidence central to this debate is lacking because of an inability to confirm a causal link through experimental research. The rise of social media intervention experiments to tackle this issue represents a positive change for the field. However, these studies are encumbered by methodological challenges that prevent a clear take-home message from this literature. In this Perspective, we discuss key methodological considerations for experimental intervention studies on social media use and provide recommendations for how researchers could design their research and interpret their results to mitigate or avoid these challenges in future.</p>