Open data science approaches to pig welfare are enabling previously impossible analyses of welfare drivers, intervention effectiveness, and geographic variation in welfare outcomes.
Open data science represents a force multiplier for pig welfare improvement. Analyses possible only with large-scale open datasets reveal patterns invisible to individual researchers: the farm characteristics most strongly predictive of poor welfare, the interventions with largest effect sizes across diverse farm contexts, and the geographic clusters where targeted support would have most impact. Normalising open data in welfare research creates cumulative improvement in knowledge quality.