<![CDATA[Human resources play a crucial role in determining the success of government organizations, where employee performance is a key factor in achieving goals and improving public service quality, including at the Regional Revenue Agency of North Sumatra Province. However, observations show that the performance of some employees is still not optimal, both in terms of quantity, quality, and timeliness. This condition is influenced by several factors such as competence, loyalty, motivation, and work-life balance. Limitations in technical skills, digital proficiency, low levels of loyalty and motivation, and a lack of balance between work and personal life have been found, which ultimately affect employee work productivity. This research aims to complement previous studies by simultaneously examining the impact of these four variables on employee performance through a more comprehensive approach, relevant to the demands of digitalization in government agencies. The population in this study includes all employees of the Regional Revenue Agency of North Sumatra Province, totaling 139 individuals. Since the population is relatively small, the sampling technique used is a saturated sample, where all members of the population are treated as respondents. The research findings show that competence, loyalty, motivation, and work-life balance each have a positive and significant impact on employee performance. Among these four variables, competence is the most dominant factor in improving performance. Employees with sufficient knowledge, skills, and abilities tend to work more effectively and efficiently. High loyalty also encourages a sense of commitment and responsibility toward the organization, while strong work motivation increases enthusiasm in achieving established targets. Additionally, the balance between work and personal life plays an important role in helping employees work more focused and productively. Simultaneously, these four variables have proven to significantly contribute to improving employee performance.]]>
