Institutional on-site visit can reflect the focus of institutional investors in a timely manner. This article starts with the short-term and long-term impacts of institutional on-site visit, using abnormal return models and factor model to study the value of institutional on-site visit. It is found that institutional on-site visit has certain information value in the short term, but the abnormal return is not large, and long-term abnormal returns are not obvious. Further information processing is needed in conjunction with other information to find effective investment methods. Research has found that: (1) The results of the CAPM model show that cumulative abnormal returns are significantly different from 0 during the impact window period, and the t-values are very high. The abnormal returns of the Fama French model are only significant within the two intervals of [T0-1, T0+1] and [T0+1, T0+5], indicating a certain level of significant abnormal returns within the week after the event announcement date. However, the average daily abnormal return within the interval is less than 1%, indicating that the abnormal returns are not significant. (2) During the financial reporting period, the abnormal returns were significantly negative in the 5 trading days before the research, while the abnormal returns were significantly positive in the 5 trading days after the on-site visit. On the survey day, the abnormal returns were not significant. In the non financial reporting period, the sample size accounted for 28.11%. Although abnormal returns were positive before and after the survey, their impact during the time window was not significant. (3) There were significant negative abnormal returnsii before the survey, while there were significant positive returns after the survey. (4) Institutional on-site visit cannot predict excess returns for individual stocks, and this effect is valid within 3 months, 6 months, and 1 year.