Virtual Stars, Real Fans: Understanding the VTuber Ecosystem

Yiluo Wei, Gareth Tyson · 2025

Read Virtual Stars, Real Fans: Understanding the VTuber Ecosystem by Yiluo Wei, Gareth Tyson (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-01553.txt.

Introduction

Yiluo Wei and Gareth Tyson study the viewers behind VTubers, streamers who perform as animated avatars run by hidden real people. Despite hiding their faces, VTubers draw huge, loyal audiences and earn serious money. Who these fans are, and how they behave, was poorly understood.

The pair built a data set of 2.7 million streaming sessions from Bilibili, the main home of VTubers in China. They profiled viewer traits, engagement habits, and influence inside fan communities. Then they flipped the usual recommendation logic: instead of suggesting streams to viewers, their tool suggests likely future fans to the streamer. Knowing which viewers to court could help small VTubers grow a core audience faster.

Catalog record

CREATORYiluo Wei, Gareth Tyson
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE67 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-01553

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Virtual Stars, Real Fans: Understanding the VTuber Ecosystem Yiluo Wei, Gareth Tyson Abstract . Livestreaming by VTubers — animated 2D/3D avatars controlled by real individuals — have recently garnered substantial global followings and achieved significant monetary success. Despite prior research highlighting the importance of realism in audience engagement, VTubers deliberately conceal their identities, cultivating dedicated fan communities through virtual personas. While previous studies underscore that building a core fan community is essential to a streamer’s success, we lack an understanding of the characteristics of viewers of this new type of streamer. Gaining a deeper insight into these viewers is critical for VTubers to enhance audience engagement, foster a more robust fan base, and attract a larger viewership. To address this gap, we conduct a comprehensive analysis of VTuber viewers on Bilibili, a leading livestreaming platform where nearly all VTubers in China stream. By compiling a first-of-its-kind dataset covering 2.7M livestreaming sessions, we investigate the characteristics, engagement patterns, and influence of VTuber viewers. Our research yields several valuable insights, which we then leverage to develop a tool to “recommend” future subscribers to VTubers. By reversing the typical approach of recommending streams to viewers, this tool assists VTubers in pinpointing potential future fans to pay more attention to, and thereby effectively growing their fan community. Conference: Proceedings of the ACM Web Conference 2025; April 28-May 2, 2025; Sydney, NSW, Australia Proceedings of the ACM Web Conference 2025 (WWW ’25), April 28-May 2, 2025, Sydney, NSW, Australia DOI: 10.1145/3696410.3714803 ISBN: 979-8-4007-1274-6/25/04 CCS: Human-centered computing Empirical studies in collaborative and social computing 1. Introduction Live streaming has gained significant popularity globally, encompassing a wide range of themes, including personal experiences ( Lottridge et al., 2017 ; Lu et al., 2019a ; Tang et al., 2016 ) , artistic creation ( Fraser et al., 2019 ; Lu et al., 2019b ) , educational content ( Lu et al., 2018b ; Faas et al., 2018 ; Lu et al., 2018a ) , and gaming sessions ( Hamilton et al., 2014 ; Deng et al., 2015 ) . As part of this trend, we have recently witnessed the emergence of a novel type of streamer, referred to as VTubers (Virtual YouTubers) ( Lu et al., 2021 ) . A VTuber is an animated 2D or 3D virtual avatar that performs in live video streams. These streams are produced using tools like Live2D ( Live2D Inc., 2023 ) that capture the actor’s movements and control the avatar accordingly. The avatar is therefore voiced and controlled by a specific person. VTuber content covers a wide range of topics, including anything that does not require a real human to present. Although seemingly fringe, VTubers have been gaining significant popularity, with dedicated fan bases and corporate sponsorship deals. There are tens of thousands of highly active and influential VTubers — the most popular ones have millions of followers ( Virtual-youtuber.userlocal.jp, 2024 ) . Thanks to the various monetization mechanisms offered by live streaming platforms, VTubers earn substantial incomes. Among the top 50 highest-earning YouTubers based on all-time superchats (viewer donations), 31 are VTubers, each earning between $1.1 million and $3.2 million USD ( playboard.co, 2024 ) . In contrast to prior studies showing that a sense of realism is a crucial factor in audience engagement ( Tang et al., 2017 ; Haimson and Tang, 2017 ) , VTubers intentionally conceal their real identities and present themselves to viewers through virtual avatars. This approach completely reconstructs the self-representation ( Wan and Lu, 2024 ; Chinchilla and Kim, 2024 ) of the streamer, and also the relationship between the streamer and their fanbase ( Xu and Niu, 2023 ; Lu et al., 2021 ; Turner, 2022 ; Lee and Lee, 2023 ) . This is driven by an intersection of Japanese otaku culture and idol culture ( Lu et al., 2021 ; Ferreira et al., 2024 ) , with the core viewer communities of VTubers exhibiting similar traits. As a result, these online viewer communities play a much more significant role compared to those of other streamers. The core viewers of a VTuber usually create and share their own subculture, and are active and passionate. Core viewers engage not only in live streaming but also contribute in various other ways to expand the VTuber’s influence. For instance, there are over 420,000 dedicated fan artworks on Pixiv ( Pixiv, 2024 ; Wei and Tyson, 2024 ) for VTubers affiliated with Hololive (a “celebrity” VTuber agency). Consequently, building such a core viewer community is essential to a VTuber’s success, both in terms of fame and monetization. Yet, we still possess only a limited knowledge of the characteristics and behaviors of VTuber’s viewers. We argue that gaining a deeper understanding is crucial for VTubers to enhance engagement with their audience, foster a stronger fan community, and thus attract more viewers (thereby generating greater monetary income). This insight can be especially beneficial for self-operated and startup VTubers, who do not belong to a management agency, as it will help them better understand their audience and identify potential core viewers. To bridge this gap, we conduct the first large-scale measurement study of VTuber core viewers, focusing on Bilibili as an exemplar platform. Bilibili is a major website for user-generated video and live streaming. It currently hosts the largest number of VTubers in China, with almost all Chinese VTubers streaming on the platform. Importantly, on Bilibili, core viewers can be easily identified through a membership system, whereby users can pay a monthly fee to gain membership for a streamer. That is, “members” are the core viewers. This membership is similar to a paid subscription on Twitch but more expensive and with additional privileges e.g., participation in a private chat group. Consequently, purchasing a membership signifies a strong commitment to the streamer, making members of a VTuber and their community an ideal target for our study. To investigate the members of VTubers on Bilibili, we compile a dataset encompassing all live streaming records (2.7 million live records with 3.6 billion chat messages in total) of 4.9k VTubers, along with 21k other streamers on Bilibili, from June 2022 to August 2023. This dataset includes almost all streamers with a moderate fanbase on Bilibili, representing the largest dataset of VTubers available, to the best of our knowledge. Exploiting this dataset, we focus on the following research questions: • RQ1: How do the behaviors of members differ from those of non-members? How do these patterns change before and after they become a member? Understanding these behavioral variations can assist in identifying potential future members and improving strategies for member promotion and retention. • RQ2: How do the chat messages sent by members differ from non-members? How do these chat patterns change before and after they become members? Besides identifying potential members, analyzing these communication patterns can offer insights into the unique culture of the community among members, guiding efforts in better community building. • RQ3: Leveraging the findings from RQ1 and RQ2, can we develop a tool that assists VTubers in identifying viewers likely to become members in the future? While numerous studies have explored recommending streams to viewers, the inverse — identifying valuable viewers for streamers — remains unexplored. By flipping the

Read the complete text