The　increasing　demand　for　hydrogen　production　has　driven　interest　in　ammonia　decomposition.　Iron-based　catalysts,　widely　used　for　ammonia　synthesis,　exhibit　suboptimal　performance　in　the　reverse　process　due　to　their　tendency　to　form　iron　nitrides.　Recent　experiments　have　shown　that　alloying　iron　with　cobalt　enhances　the　catalytic　activity　(Chen　et　al.,　Nat.　Commun.　15,　871,　2024),　yet　the　microscopic　origin　of　this　promotional　effect　is　not　fully　understood.　To　address　this,　we　leverage　recent　developments　in　machine　learning-based　molecular　dynamics　simulations　to　investigate　the　key　reactions　of　the　catalytic　cycle,　fully　accounting　for　dynamical　lateral　interactions　on　the　catalyst　surface.　Our　simulations　reveal　that　cobalt　alloying　provides　a　dual　promotional　effect:　it　slightly　lowers　the　free　energy　barrier　for　nitrogen　recombination,　which　is　the　rate-determining　step　for　ammonia　decomposition　on　iron,　while　significantly　suppressing　nitrogen　migration　into　the　bulk,　thereby　preventing　nitride　formation.　These　insights　are　supported　by　complementary　transient　decomposition　experiments　and　desorption　measurements,　which　confirm　the　enhanced　activity　and　resistance　to　nitridation　in　FeCo　alloys　compared　to　monometallic　iron　catalysts.　Furthermore,　long-term　stability　tests　demonstrate　that　the　FeCo　catalyst　sustains　high　ammonia　conversion　over　extended　time　scales.　By　capturing　the　complex　interplay　of　competing　dynamical　processes　at　the　atomic　scale,　our　results　highlight　the　importance　of　going　beyond　static　structure-property　relationships　to　gain　mechanistic　insights　that　can　guide　the　rational　design　of　more　robust　and　efficient　catalysts.