The objective of this vitro study was to determine the effect of the polishing performance of a universal ceramic polishing kit on the surface roughness of lithium disilicate glass-ceramic and translucent zirconia. The ceramic specimen of size 5 x 7 x 4 mm of lithium disilicate glass-ceramic (IPS e.max CAD, Ivoclar Vivadent, Liechtenstein) and translucent zirconia (VITA YZ XT, VITA Zahnfabrik, Germany) (n=5) were fixed with clear resin in PVC block size diameter 14 mm. The specimens were ground with fine diamond bur for 15 seconds to simulate clinical gross contouring. The two-step polishing process started with coarse polishing (EVE Diacera H2DCmf) for 45 seconds and was followed by fine polishing (EVE Diacera H2DC) for 45 seconds. The surface roughness (Ra) of ground and polished specimens were measured every 15 seconds of coarse and fine polishing steps. The Ra measurement was analyzed using a non-contact optical profilometer (Alicona infinitefocusSL, Graz, Austria) at 50X magnification for five areas (0.4x0.4 mm), one at the center, and 1 mm apart from the center in four directions. The area of measurement was perpendicular to the polished direction, provided 4 mm in evaluation length according to ISO 4288. The force in the polishing process was controlled to 1 N by a customed-made device. One-way repeated measures ANOVA and Post-hoc comparisons by Bonferroni test were used to determine the effect of each polishing duration on the Ra in each step. An independent t-test was used for the mean comparison of the Ra in all polishing steps between both ceramic groups. Paired samples t-test was used for comparison in Δ Mean Ra between both polishing steps in the same material type while an independent t-test was used for comparison in Δ Mean Ra between material types within the same polishing step. The result showed that when compared with the grinding step, the Ra of IPS e.max CAD was significantly lower after being polished by a coarse polisher at 45 seconds, while VITA YZ XT was at the first 15 seconds. The fine polishing reduced Ra with no significant difference from final coarse polishing in both ceramics. However, the ceramic types had no significant effect on Ra when compared with grinding and each polishing step. The result of Δ Mean Ra showed that coarse and fine polishing bur was more effective in VITA YZ XT than IPS e.max CAD.