Using online sales data of listed companies in the consumer sector, we examine the predictive effect of online sales data on future stock returns. Our results show that the year-on-year growth rate of monthly online sales can predict the stock returns of the next month, and the calendar portfolio constructed based on the growth rate of online sales has significant excess returns. This article further empirically analyzes the two conditions for online sales data to have predictive effects: online sales data contains information related to the company's operating performance, and stock prices cannot respond to these information in a timely manner. We find that the year-on-year growth rate of online sales can significantly predict the growth of the company's operating income and the change in profitability, and the return rate of the calendar portfolio is highest in the second week after the portfolio is constructed, that is, investors’ response to online sales information has been delayed. In addition, this article also finds that online sales data contains new information beyond the traditional financial information that affects stock pricing. The conclusions of this article are of great significance for understanding the role played by alternative data such as online sales data in the capital market and the effectiveness of the market.