Electrocatalyst discovery is an inherently multiobjective challenge that can benefit from closed-loop approaches towards acceleration. However, previous computational closed-loop efforts for electrocatalysis have often focused on a single objective to be optimized. Here, we propose a multiobjective closed-loop strategy driven by sequential learning (SL) that employs a product of normalized property metrics to score candidates. In each iteration, a candidate catalyst system is autonomously selected via the multiobjective score, as implemented in our AutoCat software, and evaluated using a high-throughput density functional theory (DFT) pipeline. As a demonstration, we apply this scheme towards a model problem of searching for single-atom alloy (SAA) electrocatalysts for nitrogen reduction, balancing three targets: activity, stability, and cost. We limit our search to dopants on close-packed surface facets of simple transition metals, resulting in a total of 441 SAA systems in our design space. We show that our proposed formulation of the multiobjective scoring system and the SL framework efficiently explore the SAA design space to find optimal candidates. We also propose a ranking scheme that quantifies the effectiveness of an identified candidate in balancing all the target objectives, taking into account the uncertainty in the preliminary evaluation method (DFT) itself. Based on this scheme, we identify a few top-performing SAA candidates—Zr₁Cr, Hf₁Cr, Ag₁Re, Au₁Re, and Ti₁Fe—for further investigation.