The purpose of this paper is to study how related variety influences firm sales. We apply an instrumental variable method (as well as the most recent plausible instrumental variable method in robustness tests) to analyze more than 600,000 firm observations in all of Japan's 47 prefectures. We find that related variety, as a kind of regional industrial structure, has a U‐shaped relationship with firm sales. This finding enriches the related variety perspective by supplementing micro‐level evidence, revealing that the “related variety–firm sales” relationship is not linear, as most prior studies have suggested.