While the Woven EndoBridge (WEB) device has transformed the treatment of wide-neck intracranial aneurysms, incomplete occlusion remains a significant challenge requiring better understanding of contributing factors. A retrospective analysis was conducted on multicenter data from patients who underwent WEB device treatment for intracranial aneurysms between January 2011 and December 2022. Using machine learning models, Cox regression, and time-stratified analyses, we evaluated factors associated with persistent incomplete occlusion, defined as non-improving Raymond-Roy Occlusion Classification grade 2 or 3 at final follow-up. Among 813 patients (607 with < 24 months follow-up, 206 with ≥ 24 months), machine learning analysis identified aneurysm height, Acom location, neck diameter, and pretreatment mRS as predictors of persistent incomplete occlusion. On Cox regression. larger aneurysm neck diameter (HR 1.13, 95% CI 1.01-1.27, p = 0.027) and height (HR 1.14, 95% CI 1.02-1.26, p = 0.017), and radial access (HR 2.68, 95% CI 1.76-4.07, p < 0.001) increased, while posterior circulation location (HR 0.56, 95% CI 0.37-0.84, p = 0.005) decreased the risk of persistent incomplete occlusion. Time-stratified analysis revealed that in short-term follow-up (< 24 months), larger aneurysm neck diameter (OR 1.28, 95% CI 1.08-1.52, p = 0.004) increased the risk of incomplete occlusion. In long-term follow-up (≥ 24 months), smoking (OR 2.69, 95% CI 1.04-7.00, p = 0.04), higher pre-treatment mRS (OR 1.78, 95% CI 1.15-2.76, p = 0.009), and immediate flow stagnation (OR 0.33, 95% CI 0.11-0.96, p = 0.04) increased, while older age (OR 0.94, 95% CI 0.90-0.98, p = 0.002) and WEB-DL (OR 0.06, p < 0.001) and SLS devices (OR 0.02, p = 0.003) decreased the risk of persistent incomplete occlusion. Aneurysm characteristics and device type significantly influence long-term WEB treatment outcomes.