This summer I got to apply what I learned at Booth at my summer internship with Hyatt Hotels Corporation in their marketing division. My principal task was to identify the factors that drive guests to promote the Hyatt brands. Hyatt uses Net Promoter Score which defines consumers to be a promoter, passive or detractors of a brand based on their self reported likelihood to recommend. I used information from guest satisfaction surveys and other sources to create a data set of 180 variables and 350,000 observations. I applied the data mining methods I learned in Matt Taddy's Data Mining class such as Random Forest, Principal Component Analysis and Penalized Multinomial Logistic Regression to model the high dimensional data set. Hyatt was very excited about this analytical approach to consumer insights as predictive modeling could be used for several applications in marketing, operations and finance. For example, when considering a new innovation to implement, you can use the model to predict what impact it will have on profits. In fact, one of my recommendations was on a new innovation they were considering and I was able to use the model to quantify the benefit that they would gain from brand health relative to the cost of the innovation. You are able to learn which aspects of the hotel are more important to certain segments of guests and thus are able to do more effective targeting of promotions. I am very thankful for being at Chicago Booth which is leading research in this growing field.