Matching Mobile Applications for Cross Promotion∗†

Mobile ecosystem is one of the most successful markets in recent years with billions of mobile device users and millions of mobile applications (apps) in multiple app platforms. As the market grows, the challenge is how app developers advertise their apps to the right users and how customers search the right apps that fit their needs. Cross promotion, advertising a mobile app in another established app, is introduced as a new promotion framework. The performance of this emerging ad framework has not been well studied in the literature. Using data from 1,011 cross promotions that ran between September 2013 and May 2014 in Korean app markets involving one million consumers and 325 mobile apps, we evaluate the effectiveness of cross promotion in comparison with existing ad channels such as mobile display ads. While cross promotion, on average, is still suboptimal as compared with display ads, we find evidence that a careful matching of mobile apps can significantly improve the effectiveness of cross promotions. We model the ad placement in cross promotions as a matching problem and identify significant factors that contribute to better app matching. Results show that app similarity, measured by topic models from apps’ text descriptions, is a significant factor that increases the user engagement in cross promotions. With the observations, we propose a matching mechanism to generate app matches with stability and improved effectiveness.