2 TRACKING AUDIENCE
BEHAVIOR
2.1 Audience Viewing Habits
Audiences tend to watch movies online and are
fragmented (Miles, 2019). At the beginning of the
epidemic outbreak, Chinese cinema attendance
dropped by 68%, but online viewing surged by 70%.
Audiences favored short content (e.g., film reviews
on Tik Tok) and often watched classic films (e.g., The
Shawshank Redemption) multiple times.
Audiences watch movies through social media.
Viewers interact with each other through barrage (e.g.
BiliBili) or live streaming with microphones (e.g.
TikTok), creating a "cloud movie watching"
community. Users on Weibo and TikTok shared their
opinions about Detective Chinatown 3, which became
one of the key factors for the film's first-day box
office to break 1 billion yuan.
2.2 Audience Consumption Behavior
The audience's behavior of paying to watch movies
presents different wills. 70% of online viewers are
willing to pay for the films they want to watch, but
they want the films to be of high quality (e.g. original
Netflix films). "Ne Zha" has also increased, resulting
in total sales of more than $200 million for the
peripherals products. The subscription model of OTT
platforms allows viewers to choose multiple sets and
pay flexibly according to their needs (Chen, 2023).
Therefore, the value of content to viewers is also
related to the role of certain channels, such as social
media and internet media, especially as viewing
experiences shared on social media platforms can
significantly influence the consumption decisions of
other viewers. Audiences are more cautious about
buying tickets and watching movies offline. 59% of
viewers say they are saving money on trips to the
cinema and instead investing in home cinema tools,
such as projectors, and private on-demand services).
2.3 Audience Content Preferences
The current state of viewers is that they are
increasingly demanding diverse content, mainly
localized and regionalization content (Mugil &
Kenzie, 2025). OTT platforms cater to the different
needs of viewers: they support multilingualism and
multiculturalism. Viewers like content that is relevant
to their own cultural backgrounds around the world.
In addition, people want interactive and participatory
experiences, so the use of virtual reality and
augmented reality technologies brings new forms of
viewing. In addition, viewers show a strong interest
in action, science fiction and superhero films, as
evidenced by the global popularity of Marvel films.
2.4 Audience Viewing Mechanism
In the new media era, criticism of films is no longer
focused on professional critics, but shifted to the
public, with an emphasis on popularity,
fragmentation and real-time feedback. Therefore, the
ecology of film criticism is diversified. In the age of
digital communication, the film viewer's experience
does not end with the closing credits. Viewers can go
to social media platforms such as Weibo or Tik Tok,
or use streaming services such as Tencent Video, and
post a large number of short comments, emojis, or
barrage, creating what is known as "instant word-of-
mouth". To illustrate, the film "Hi Mom" reached
billions of readers on Weibo on its first day of release.
The "mother-daughter love" video created by users in
the style of “capsule" triggered emotional resonance
and indirectly boosted box office growth.
The platform presents differentiated
characteristics. The success of art films and the
audience's decision on which film to watch is directly
related to Douban's in-depth and lengthy review and
rating system. For example, Return to Dust rating on
Douban largely helped viewers choose the film,
which had a rating of 8.4, allowing it to transition
from niche to mainstream. A film's highlights or
shortcomings quickly go viral through short 15-
second videos. By highlighting certain details in
micro formats, short videos have an impact on
potential viewers. For example, the topic "Moon
Man" has over 2 billion views, directly influencing
people's decision to watch the film. BiliBili, the
culture of screen comments brings people together,
creating a collective atmosphere for watching films.
The implementation of this "barrage" reinforces the
audience's emotional engagement. For example, in
the screen comments of The Wandering Earth 2, high-
frequency words such as "tears in my eyes" and "the
rise of Chinese sci-fi" reinforce the audience's
emotional identification. Words reinforce the
audience's emotional identification.
Algorithmic Recommendation and Film Review
Visibility. Positive feedback loops and information
filtering: Streaming media platforms use data
algorithms to automatically present and filter certain
film reviews, creating a "review filter bubble". Based
on monitoring the behavior of Netflix viewers,
similar movie reviews are recommended to viewers,