What is wrecking your data plan? A measurement study of mobile web overhead

The growing popularity of smartphones and continuous user demand for a rich web experience has resulted in an exponential surge in cellular bandwidth requirements. Cellular providers have struggled to keep pace with the new requirements while users often face a monetary cost associated with the data downloaded to their device. While many modern websites have adapted to the new mobile habitat, they often take shortcuts to transition from their desktop to mobile versions, many times carrying redundant content that is never utilized. Moreover, mobile users are effectively paying for certain undesirable content, such as advertisements, in the form of their bandwidth costs. In this paper, we study the composition and complexity of modern websites, from both a mobile and desktop perspective, to identify sources of wasted bandwidth. We developed a custom crawler-based framework to perform an in-depth analysis of the top 100,000 popular sites ranked by Alexa. Our results show that 23% or more of the content size on an average website is unnecessary, unused or redundant. Our results serve as a motivation for developing optimized websites and enhancing the web infrastructure to better suit the mobile environment with emphasis on reducing bandwidth costs, while also improving performance and efficiency.