Acquiring accurate and timely market share information is very important for producers to arrange producing plan and design marketing strategy. However the high cost and long period of collecting survey data in survey-based method make it much difficult to easily get latest market shares data. Recently, the emerging online web systems provide users with new and convenient ways of searching, learning, experiencing and buying products. The users’ activities data captured by these web systems can reflect users’ buying intentions and behaviours very well, and contain very valuable information for predicting market shares. In this study, the correlation between Google search index and market shares of mobile phones is analyzed with time series analysis technology. The experiment result shows the statistically significant relationships exist between search index and market shares. This indicates the easily got search index data with low cost has the power of timely forecasting market shares. This study opens a door to apply users’ online activities data to accurately and timely predict market shares, which will bring many benefits to producers and customers.