Representative bureaucracy studies have explored the potential impact of symbolic representation on clients' attitudes and behaviors, even in the absence of direct action by bureaucrats. Through four original conjoint experiments conducted with a nationally representative sample of 1600 participants in China, we analyze the symbolic representation effects of five most commonly observable identity signals, including gender, age, accent, party affiliation, and human voice (versus AI chatbot), across two types of bureaucrat‐initiated contacts and two types of client‐initiated contacts. Our findings contribute to the existing literature by demonstrating that identity congruence has heterogeneous symbolic representation effects; clients' expectations about active representation serve as a significant mediating factor influencing their willingness to cooperate with bureaucrats.