Abstract: This study investigates the alignment between government subsidy policies and corporate behavioral preferences in blockchain-enabled agricultural assistance. It aims to reveal how cost subsidies versus sales-based subsidies differentially influence the decisions of self-interested and altruistic enterprises, thereby optimizing policy efficacy and sustainable agricultural development. A tripartite Stackelberg game model involving the government, enterprises, and farmers is constructed to analyze equilibrium outcomes under four scenarios. Sensitivity analysis and strategy comparisons are employed to validate the results.The findings demonstrate that sales-based subsidies effectively incentivize self-interested enterprises to enhance traceability efforts, expand market scale, and significantly increase farmer income. In contrast, cost subsidies are more suitable for altruistic enterprises facing high R&D costs, as they alleviate financial pressures. Further analysis shows that when technology costs are low and consumer traceability preferences are high, sales subsidies achieve multi-win outcomes; conversely, excessively high costs coupled with low market demand may lead to a "prisoner’s dilemma." The study innovatively proposes a "policy-preference" matching mechanism, providing theoretical and practical insights for designing targeted blockchain agricultural subsidy policies.