Background: Precise size measurement of colorectal polyp is essential prerequisites for colonoscopy surveillance interval and resection strategy of polyp. However, both visual and pathologic estimation of polyp size have been known to be inaccurate. We evaluated the usefulness of the colon cap and external grid application on monitor (gCAP) for the size measurement of colon polyp. Methods: This study was a prospective randomized trial at a single tertiary hospital for 3 years. The accuracy and the availability (measuring time) of gCAP for polyp size measurement were compared with that of visual estimation (VE). Results: A total of 280 polyps were enrolled from 111 patients.Mean sizes of polyps measured were 4.0±1.7 mm by gCAP and 4.2±1.8 mm by VE (p＝0.368). Comparing with endoscopic forceps method, mean discrepancies (MD) and mean absolute discrepancies (MAD) were lower in gCAP group than in those of VE group. Measuring time of gCAP group was 4 seconds longer than that of VE group. However, the differences of measuring time between each method and gold standard forceps methods were more than 20 seconds. On subgroup analysis by size, gCAP was more accurate for the measurement of polyps ≥ 5 mm in size. On the other hand, however, gCAP was not superior for size measurement of polyps ＜5 mm in size. Conclusions: gCAP was a more accurate method for polyp size measurement compared with visual estimation. In addition gCAP did not take long time as forceps method did, especially for polyps ≥ 5 mm in size.