The blast furnace (BF) process accounts for over 90% of global iron ore-based steel production and remains a major contributor to CO2 emissions. This study investigates the variability in operations across the global BF fleet and its impact on CO2 emissions. Using plant-level data from 2011 to 2021, we find that variation in the composition of inputs results in emissions ranging from 1.20 to 2.08 tons of CO2 per ton of hot metal (tCO2 tHM−1) globally, driven by differences in fuel and reductant consumption. Our estimates suggest global emissions could have been lower by 2.8%—and up to 9.6%—if each furnace operated at its historical or national minimum CO2 emissions intensity. We use furnace-level data to find even greater variation within and across individual furnaces at the national level, estimating potential for CO2 emissions reductions of 5.7%–15.5% in the United States. These findings suggest substantial latitude to incentivize plants to reduce CO2 emissions without new capital investment, by up to an estimated 1.35 billion tons of CO2 cumulatively across the global BF fleet from 2011–2021. It further underscores the limitations of relying on ‘best available technology’ or country-specific emissions factors in policy and planning efforts.