Official monthly unemployment data is unavailable in China, while intense public interest in unemployment requires timely and accurate information. Using data on web queries from lead search engines in China, Baidu and Google, I build two indices measuring intensity of online unemployment-related searches. The unemployment-related search indices identify a structural break in the time series between October and November 2008, which corresponds to a turning point indicated by some macroeconomic indicators. The unemployment- related search indices are proven to have significant correlation with Purchasing Managers’ Employment Indices and a set of macroeconomic indicators that are closely related to changes in unemployment in China. The results of Granger causality analysis show that the unemployment-related search indices can improve predictions of the macroeconomic indicators. It suggests that unemployment- related searches can potentially provide valuable, timely, and low-cost information for macroeconomic monitoring.