Background: Overactivation of the salience network (SN) causes hyperarousal in insomnia patients and is associated with sleep-onset insomnia (SOI). Resting-state microstate 3 (RS-MS3) duration is closely related to SN overactivation. However, whether RS-MS3 duration is a biomarker for SOI has not yet been reported in the literature. In addition, SN activity is also associated with efficiency. However, it is not clear whether there are individual differences in the neural mechanisms of SOI in different efficiency groups. Methods: Considering that RS-MS3 duration characterizes the stability and persistent activation of the SN in the resting state, the current study investigated the link between SOI measured by sleep latency of Pittsburg Sleep Quality Index (PSQI), efficiency measured by Kirton Adaption-Innovation Inventory (KAI), and RS-MS3 in a Chinese healthy (subclinical) student population, using electroencephalography (EEG) microstate analysis. Results: We found that RS-MS3 duration was positively correlated with sleep latency and efficiency. The interaction between sleep latency and efficiency was significant. Simple slope analysis showed that high sleep latency was positively correlated with longer RS-MS3 duration in participants with higher efficiency scores. This correlation did not exist in participants with low efficiency scores. Conclusions: RS-MS3 duration may serve as a biomarker for SOI. There is heterogeneity in the relationship between SOI and RS-MS3 duration between individuals with high and low efficiency.