Understanding the impact of video quality on user engagement

Abstract

As the distribution of video over the Internet is becoming mainstream, user expectation for higb quality is constantly increasing. In this context, it is crucial for content providers to understand if and bow video quality affects user engagement and bow to best invest tbeir resources to optimize video quality. Tbis paper is a first step toward addressing tbese questions. We use a unique dataset tbat spans different content types, including sbort video on demand (VoD), long VoD, and live content from popular video content providers. Using client-side instrumentation, we measure quality metrics sucb as tbe join time, buffering ratio, average bitrate, rendering quality, and rate of buffering events. We find tbat tbe percentage of time spent in buffering (buffering ratio) bas tbe largest impact on tbe user engagement across all types of content. However, tbe magnitude of tbis impact depends on tbe content type, witb live content being tbe most impacted. For example, a 1% increase in buffering ratio can reduce user engagement by more tban 3 min for a 90-min live video event.

Real-Time Streaming Data Delivery over Named Data Networking

“…The chasing phase must be completed before data can be rendered for the user; thus, its duration directly affects QoE [41] and should be kept as small as possible. A series of experiments was conducted in order to explore how parameters of the stability estimator introduced in Sect.…”

QoE prediction model for mobile video telephony

“…There have been quite a few interesting works ( [9,22] and [7]) on characterization of smartphone apps. To the best of our knowledge, smartphone video telephony, the new data hog in mobile apps, has not been addressed so far in terms of it's video quality in the literature.…”

On-road video delivery with integrated heterogeneous wireless networks

“…Song et al [12] studied how zooming and ROI-enhancement coding can affect the user experience of mobile sports videos but they did not pay much attention to networking or communication issues. Dobrian et al mainly analyzed the impact of video quality on user engagement based on large-scale user data [1].…”

“…This (denoted as jt i for u i ) is defined as the waiting time before the videos start to play [1] (which is actually related to video caching mechanism). Here, we define y i = 1 À (jt i / t i ) to model the user experience related to join time.…”

“…However, it is a non-trivial challenge to guarantee satisfactory user experience for interactive services like video and gaming applications, due to limited bandwidth, coverage, and other limitations of wireless and mobile communications. Besides the traditional aspects used to evaluate the user experience of on-line video services, e.g., video interruptions and join time (i.e., the waiting time before videos start to play) [1], the user experience of on-road video services also depends on some unique performance aspects. As an example, in on-road video services, it is essential to consider whether a user can enjoy the requested videos during his/her trip which has a limited duration.…”