This paper considers two critical issues of platform design: personalized recommendations and drip pricing. In many online markets, consumers often have little ex-ante knowledge of product features. In these markets, how does platform design impact consumer welfare, seller outcomes, and platform profits? I answer this question in the context of the online market for hotel rooms using data from Expedia Group, an online travel agency (OTA). I present evidence that in this market, consumers do not ex-ante know product features, including price. On the demand side, this paper proposes an optimal sequential search model where consumers have rational expectations of the joint distribution of product features, form consideration sets through page turns and clicks, and make a final purchase decision from their consideration set. As for the platform, I use LambdaMART, a machine learning algorithm, to create a mirror of the platform’s recommendation system. The search engine mirror allows for the supply side model where profit-maximizing firms consider how changes in price impact position on the page in search results. The proposed model and estimation strategies allow for platform design counterfactuals, including changes to the order of search results and feature emphasis. The paper concludes by evaluating the welfare effects of personalized recommendations and drip pricing.