In this project we investigate how online reviews can influence product performance in the music and video games industry. We scrape data from Metacritic, Billboard, and VGChartz for our analysis. We employed Latent Dirichlet Allocation to identify latent topics from the online reviews and investigate if these topics can be related to product sales. Using these technique, we contribute to the industry to identify topics that can relate to sales. Such topic can change across time and/or industry, but the same technique can be applied in future research to conduct new investigation in a different industries.