What makes the mean uniquely useful compared to other measures of central tendency, and what is a primary limitation researchers must consider before relying on it exclusively?