Several viewpoints have been reported regarding the effect of temporary cements, different surface pretreatment protocols before adhesive cementation, and predictive factors. This in vitro study tested if temporary cement, pretreatment of the tooth surface, the size of enamel or dentine influence adhesive cementation to zirconia ceramics. Twenty premolars were prepared for determination of enamel and dentin area, bond strength test and failure analysis. The samples were divided into two groups: untreated prior adhesive cementation (n = 10) and with temporary cementation done, pretreated prior adhesive cementation (n = 10). Zirconia overlays (Katana Zirconia STML) were cemented on the grounded flat teeth surfaces using Panavia V5. An additional six premolars underwent dentine tubule analysis with SEM to detect temporary cement residues after temporary cementation on an untreated tooth surface (n = 3) and on a pretreated surface (n = 3). The independent sample t-test was used to compare the two groups and the means of the total tooth, dentin or enamel areas did not differ significantly between the untreated and pretreated specimens. The mean tensile bond strength was significantly (p = 0.005) higher in the pretreated specimens (337N) than in the untreated ones (204N). The overall multivariable linear regression model with three predictors (surface pre-treatment, enamel area and dentine area) was significant (p = 0.003), among which the size of enamel was the strongest predictor (β = 0.506; p = 0.049), followed by the pretreatment effect (β = 0.478; p = 0.001) and the size of dentin area (β = −0.105; p = 0.022).