PURPOSE. The aim of this study was to determine differences in shear bond strength to human dentin using immediate dentin sealing (IDS) technique compared to delayed dentin sealing (DDS). MATERIALS AND METHODS. Forty extracted human molars were divided into 4 groups with 10 teeth each. The control group was light-cured after application of dentin bonding agent ($Excite^{(R)}$ DSC) and cemented with $Variolink^{(R)}$. II resin cement. IDS/SE (immediate dentin sealing, $Clearfil^{TM}$ SE Bond) and IDS/SB (immediate dentin sealing, $Adapter^{TM}$ Single Bond 2) were light-cured after application of dentin bonding agent ($Clearfil^{TM}$ SE Bond and $Adapter^{TM}$ Sing Bond 2, respectively), whereas DDS specimens were not treated with any dentin bonding agent. Specimens were cemented with $Variolink^{(R)}$. II resin cement. Dentin bonding agent ($Excite^{(R)}$. DSC) was left unpolymerized until the application of porcelain restoration. Shear strength was measured using a universal testing machine at a speed of 5 mm/min and evaluated of fracture using an optical microscope. RESULTS. The mean shear bond strengths of control group and IDS/SE group were not statistically different from another at 14.86 and 11.18 MPa. Bond strength of IDS/SE group had a significantly higher mean than DDS group (3.14 MPa) (P < .05). There were no significance in the mean shear bond strength between IDS/SB (4.11 MPa) and DDS group. Evaluation of failure patterns indicates that most failures in the control group and IDS/SE groups were mixed, whereas failures in the DDS were interfacial. CONCLUSION. When preparing teeth for indirect ceramic restoration, IDS with $Clearfil^{TM}$ SE Bond results in improved shear bond strength compared with DDS.