In the oxidative coupling of methane (OCM), the activation of methane and the suppression of deep oxidation are in a persistent trade-off relationship, and a catalyst design strategy that balances the activity and the selectivity is desired. In this study, we analyzed a random catalyst dataset for OCM that was earlier obtained by high-throughput experimentation, and extracted heuristics such as elements, supports, and their combinations related to methane activation at a low temper-ature and selective formation of C 2 compounds at a high temperature. The obtained heuristics were used for catalyst development. The most effective was the use of a mixed support between La 2 O 3 and BaO, which improved the lowtemperature activity, the high-temperature selectivity, as well as the maximum C 2 yield. It was considered that La 2 O 3 acted as a heater and helped low-temperature operation of BaO, which is highly selective but not active at a low temperature.