Google trends is an online tool that provides access to daily and weekly data on the frequency of individual search keywords, subjects and phrases in a given time period. Due to the timeliness Google trend, a number of studies have shown that this information can be used to monitor economic trends at the time of their creation, thus avoiding the time lag that is a feature of official statistical releases. The aim of this paper is to examine the extent to which information obtained through the Google trends can serve the purpose of predicting tourist arrivals in Croatia. We analyzed the period from January 2005 through December 2014, where there were used monthly data on arrivals of foreign tourists, and information about searching the term "Croatia" .Because data on the Internet search feature strongly pronounced seasonality, as well as tourist arrivals, all the variables were seasonally adjusted previously conducting regression analysis. By comparing the results analyzed autoregression models, it was shown that the search term "Croatia" in the first two weeks, and the search in the previous month are statistically significant predictors of tourist arrivals during the month. Thus, the results of the analysis in favor of studies that highlight the value of the information obtained through the Google trend, it can be concluded that data on internet search can be used in predicting tourist arrivals in Croatia.