Purpose: To compare fracture strength, failure mode, and chairside time of Class IV fractures restored with CEREC (Chairside Economic Restorations of Esthetic Ceramics) technology or direct composite. Methods: Forty-eight fractured anterior bovine teeth were randomly assigned to three experimental groups (indirect restoration) with margin designs including: A) butt joint, B) short chamfer (one mm), and C) long chamfer (two mm) and a control group (direct composite). Preparations were scanned; restorations were milled from zirconia-reinforced lithium-silicate blocks and cemented. Fracture load (N) and failure mode were analyzed. Techniques were timed from start of margin preparation through finishing. Results were analyzed using one-way analysis of variance or the Kruskal-Wallis test (significance level: P=0.05). Results: Fracture loads (mean±standard deviation) for groups A, B, and C and control group were 2,177±644 N, 2,183±507 N, 2,666±609 N, and 2,358±886 N, respectively (not significantly different; P=0.26). The direct composite was significantly different from all indirect groups (P<0.01) for failure mode. Chairside time was longer for direct restoration. Conclusions: Fracture strength is similar for directly and indirectly fabricated Class IV restorations, with margin design not affecting strength or failure mode. Practitioner's chairside time, but not total time, is reduced when using indirect methods.