Assessment of platelet count is an important diagnostic parameter in haematology. Automated blood cell counters have largely replaced the manual method. However, all abnormal platelet counts are verified in Leishman's stained peripheral blood smear. Platelets also can be identified in the unstained blood smears. The objective of this study is to compare the unstained peripheral blood smears with the stained smears and determination of the effectiveness of unstained smears in the estimation of platelet count. 250 Venous blood samples sent for blood cell counts were analyzed. Platelets were counted in the unstained smear under 100× objective in 10 fields without placing immersion oil and the average number of platelets was calculated. Same smear was stained by Leishman's stain and platelets were counted under 100× objective after placing a drop of immersion oil. Collected data were analyzed for intra-class correlation coefficient (ICC). ICC showed excellent agreement (ICC > 0.85). The unstained smears were found to be as effective as stained smears for platelet count in most of the cases. However, in case of doubt a stained preparation has to be made to confirm the count. The turnaround time was 3-5 minutes compared to 15-20 minutes by stained smear technique, thus this technique may be used as an initial screening method whenever there is large sample load.