The Minamata Convention on Mercury entered into force in August 2017, committing its currently 92 parties to take action to protect human health and the environment from anthropogenic emissions and releases of mercury. But how can we tell whether the convention is achieving its objective? Although the convention requires periodic effectiveness evaluation (1), scientific uncertainties challenge our ability to trace how mercury policies translate into reduced human and wildlife exposure and impacts. Mercury emissions to air and releases to land and water follow a complex path through the environment before accumulating as methylmercury in fish, mammals, and birds. As these environmental processes are both uncertain and variable, analyzing existing data alone does not currently provide a clear signal of whether policies are effective. A global-scale metric to assess the impact of mercury emissions policies would help parties assess progress toward the convention's goal. Here, I build on the example of the Montreal Protocol on Substances that Deplete the Ozone Layer to identify criteria for a mercury metric. I then summarize why existing mercury data are insufficient and present and discuss a proposed new metric based on mercury emissions to air. Finally, I identify key scientific uncertainties that challenge future effectiveness evaluation.