Most studies on online marketplaces focus on seller uncertainty and rely on data from online marketplaces in the U.S. This paper extends this literature by focusing on product uncertainty and defining its two dimensions - description uncertainty (identifying the product's characteristics) and fit uncertainty (matching product characteristics with the buyer's needs). It also examines the distinction, relationship, and relative effects of the two dimensions of product uncertainty on actual product returns, and how online marketplaces can use IT-enabled mechanisms to mitigate product uncertainty. The proposed hypotheses are tested with data from 144 buyers in Taobao's online marketplace in China using an econometric model. The results stress the role of product presentations in reducing description uncertainty and of online communication, both between the buyer and the seller and also among buyers, in reducing fit uncertainty. The paper draws implications for reducing product uncertainty in online marketplaces with the aid of IT.