In recent years, grassroots environmental justice activists have increasingly used big data techniques for monitoring, recording, and reporting toxic environmental exposures. Despite the promise of big data for environmental justice, there is a need to address structural barriers to making toxic environmental exposures visible, and to avoid over-relying on new digital methods and techniques as a panacea for problems of voice. The emphasis of real-time analysis in crowdsourced and participatory big data is good at tracking the immediate aftermath of environmental disasters, but it misses slower-burning environmental problems that emerge over time. While big data more generally may have implications for understanding toxic exposure landscapes across different temporal and spatial scales, it is complex, difficult to analyze, and faces significant problems of reliability. There are three key blind spots of the ethos and practice of big data in relation to environmental justice: voice, speed, and expertise. In the context of increasing pressure to embrace new tools and technologies, it is also important to slow down and to reflect on the wider implications of the age of big data.