The purpose of the present study was to assess the proper time to perform a restoration (immediately or delayed) after placement of two calcium silicate-based cements (CSCs) and to test the performance of two different restorative protocols regarding shear bond strength (SBS). Seventy-five acrylic blocks were randomly divided into five groups ( = 15). Specimens were filled with either ProRoot MTA (Dentsply Tulsa Dental) or Biodentine (Septodont). The restoration was performed at an immediate (12 min) or delayed (seven days) timeframe, using a resin-based flowable composite (SDR) (bonded to the CSC using a universal bonding system) or glass ionomer cement (GIC) as restorative materials. SBS was measured using a universal testing machine. Fractured surfaces were evaluated, and the pattern was registered. Statistical analysis was performed using the Dunn⁻Sidak post hoc test ( < 0.05). Biodentine/immediate SDR showed the highest mean SBS value (4.44 MPa), with statistically significant differences when compared to mineral trioxide aggregate (MTA)/GIC (1.14 MPa) and MTA/immediate SDR (1.33 MPa). MTA/GIC and MTA/immediate SDR did not present significant differences regarding SBS. No statistical differences were verified concerning mean SBS between both CSCs within the 7 day groups. MTA/delayed SDR (3.86 MPa) presented statistical differences compared to MTA/immediate SDR, whereas no differences were observed regarding Biodentine performance (Biodentine/immediate SDR and Biodentine/delayed SDR (3.09 MPa)). Bonding procedures directly on top of MTA might be preferably performed at a delayed timeframe, whereas Biodentine might allow for immediate restoration.