Nowadays, customers have a variety of options to gather information about products, which can sup-port their purchasing decisions. More and more customers use YouTube reviews or unboxing videos to get a first impression of different products and interact or discuss with other users in the comment section. Automatically analyzing these comments to gain a better insight about the important product aspects remains a major challenge in the field of social media monitoring because the text data is unstructured and noisier compared to conventional review data for example from Amazon. In this study, we focus on the automated aspect extraction task to answer the question, which characteristics of products are important from the (potential) customer view. We show that YouTube comments are a valuable data source for this purpose with an aspect extraction precision comparable to conventional Amazon reviews. To improve aspect extraction in general, we propose a new aspect sorting method based on Google Trends. Incorporating the search volume of products combined with aspects into the extraction procedure improves the precision results especially for noisier text data. To illustrate the analysis results, we choose Amazon reviews and YouTube comments about three exemplary smartphones.