This study investigates work and non-work antecedents for the work–life balance (WLB) development of remote employees during the COVID-19 pandemic. Longitudinal data (N = 1,146, T1; N = 737, T2) was collected in May–June 2020 and December 2020 in one multinational company (MNC) in Finland. In data analysis, structural equation modeling (SEM) with a cross-lagged panel model was utilized. The results revealed that during the pandemic, WLB slightly decreased. The quantitative job demands increased and predicted a decreased WLB at T2. Job autonomy decreased but did not have an effect on WLB development or buffer the negative effect of quantitative job demands on WLB. Time saved from commuting was positively related to WLB development, but the number of children living at home and the age of the youngest child had no statistically significant link to WLB development over time (similar finding for men and women). Although care responsibilities from the gender perspective is not the focus of our study, the additional analysis show that WLB at T1 was more challenging for women the more children they had, or the younger the youngest child was. For men, children did not make a difference for their WLB at T1. This finding indicates that WLB has been more challenging for mothers compared to fathers already when our first data had been collected, and the continuance of the pandemic did not change the situation in any direction. This research contributes to the knowledge about work and non-work related demands and resources as antecedents for WLB development during the pandemic. As a practical implication during the pandemic, the authors suggest that employers should follow development for employees’ WLB as a measure of well-being in remote work. In addition, the workload of remote employees should be followed, and time saved from commuting should be preserved as employees’ non-work time and protected from work-related tasks.