Tieba is facing difficulties in the mobile Internet era, with product forms and mobile interactions not matched, and content organization efficiency is low. The rise of the creator economy has made high-quality content producers flow to platforms with commercial monetization models, and interest communities need to create an upward channel for content and people to establish a win-win mechanism.
I saw the news in the first two months, and this year Baidu will re-examine the post. Because search has a declining market share in the face of competition from Xiaohongshu and Douyin, it needs a post bar with 23 million interest themes to serve as Baidu’s core UGC content pool and activate user-generated content to supplement the search ecosystem.
Thinking of the Vientiane Conference three years ago, I asked the then head of Baidu MEG in person, and now Baidu’s content supply and consumption system is mainly around Baijiahao, and in the future, Baidu will do UGC creation, whether to use the account system of Tieba or return to the same account system.
At that time, it was said that Tieba was a spark for Baidu UGC, and it had to be preserved first and then slowly grown. Baijiahao is still in the stage of making up for shortcomings, and it must first concentrate resources to do big things, and first introduce external PGC content through Baijiahao to solve some problems. UGC is a long-term, long-term thing to do, so Tieba has been doing it for so many years and will not give up, which is indeed more difficult.
For example, the form of Tieba must do a lot of iterations, in terms of user experience, remove hostility, and improve the relationship between the bar owner and the bar friends. Break the barriers to the distribution of content between bars and bars, make UGC posts convenient and convenient, so that the information between bars and bars can be exchanged in the square, and at the same time, post bars with high-quality content so that good content can be sent out.
The news I saw three years later is that Baidu Tieba’s strategic focus in 2025 will be on appropriately reducing advertisements to promote the return of old users to Tieba and create independently.
Exactly three years ago, I did a live broadcast with a few friends who worked in the interest community in the early years, discussing the development history, current challenges and future trends of the interest community. At that time, the title of the podcast was “Tieba Mobile, Feed and Creators, and the Future of Interest Communities”, which analyzed the dilemma and transformation of traditional forum models such as Tieba in the mobile Internet era from multiple perspectives.
Guests:
- Song Jian (former general manager of Baidu content ecological platform)
- Shen Zhenyu (Trendy Toys, Founder of Thousand Island Community)
- Albert (former head of the connotation joke)
- Cong Guangle (the first person in charge on the far right)
- Su Qingyang (Former NetEase Cloud Community Product Manager)
- Anchor: Pan Ran (the manager of the random flip book)
After 10 years of interaction design, why did I transfer to product manager?
After the real job transfer, I found that many jobs were still beyond my imagination. The work of a product manager is indeed more complicated. Theoretically, the work of a product manager includes all aspects of the product, from market research, user research, data analysis…
View details >
The core points can be summarized as follows:
- Reasons for the decline of Tieba : The guests agreed that traditional BBS models such as Tieba have encountered challenges in the mobile Internet era due to the mismatch between their product form and mobile interaction, as well as the inefficiency of content organization. Tieba relies on search engine diversion, and its sorting mechanism (time sequence, top post mechanism) cannot meet the needs of efficient content discovery on mobile devices.
- Impact on the creator economy: With the rise of the creator economy, high-quality content producers are more inclined to turn to platforms with commercial monetization models. The key to breaking through the ceiling of interest communities is to create an upward channel for content and people, so that creators can perceive a sense of identity that continues to grow, and at the same time establish a win-win mechanism with creators and users.
- Differentiated value of community ecology: Xiaohongshu’s UGC ecosystem is considered its core competitiveness, which is more open and lower than Bilibili and Douyin, allowing ordinary users to get attention and monetization opportunities. This differentiated value allows it to continue to grow in an environment dominated by short video platforms such as Douyin.
- Differentiated path of community commercialization: Zhenyu shared the successful case of “trendy toy people”, proposing that the key to the commercialization of interest communities is to find vertical categories that combine content and consumption – some products (such as trendy toys and ski equipment) are naturally suitable for sale in the community. “Culture is a commodity, commodity is culture” is an important development direction for the future interest community.
- Future trends in interest communities: Future trends include greater information efficiency, stronger interactive feedback, and vertical tooling. Mainstream platforms have diverted general needs, and interest communities need to focus on solving the special needs of vertical scenarios. Successful communities often develop unique cultures over a long period of natural growth, rather than relying solely on burning money to grow.
The following is a summary of the discussion:
01 From PC to mobile: Tieba’s dilemma and community changes
Pan Lan: Baidu Tieba is the world’s largest Chinese forum in the PC era, what kind of problems does it encounter on the mobile side? Are content organization forms such as topics, forums, and groups still competitive today? What might the future hold for the Interest Community?
Song Jian: I want to analyze it from the perspective of traffic. People like to discuss the demographics and functionality of the community, but often ignore the traffic factor. For example, many people enjoy flowers in the Summer Palace in Beijing, but there are no people in a beautiful green space in Guangxi; Dunhuang Mogao Grottoes were also formed because of the formation of religious art communities on the Silk Road. Baidu has the commanding heights of traffic in the PC era, and the success of Tieba has a lot to do with the accurate distribution of Baidu search. Now that we are making products similar to Tieba on the mobile terminal, it is difficult to form a community atmosphere without such traffic support.
In addition, there is no essential difference between topics and main feeds, and they are both sorting mechanisms in specific functional scenarios. Traditional BBS uses time sequence as a historical inheritance, and if there was a better algorithm at that time, time sequence may not be used. Some particularly long-tail needs (such as “I want to see James’s group”) are difficult to accurately meet by the recommendation system, so topic-specific communities and recommendation feeds will still coexist, but most of the needs are covered by the main feed.
Shen Zhenyu: The product form of Tieba originated from the early electronic bulletin board (BBS). When I was in college, I used the FTerm client to access the forums of Fudan and Peking University, when BBS had no database and was completely based on the file system, and the sorting could only be in chronological order. This product form now seems primitive, it is neither a complete personal expression platform (no personal homepage), nor a good press release system (content can be quickly overwhelmed), nor a high-quality interactive system.
This model will definitely not work in the era of mobile Internet. Baidu Tieba’s product manager must be very entangled because he doesn’t know how to recommend content in the information flow –
Recommend a theme post? Many only have the words “leave to Du Niang”;
Recommend posting? 99% are water stickers. This is completely unsuitable for content distribution on mobile.
Albert: Since Web 2.0, the most frequent demand of users has been content acquisition. Many products that look like communities mainly solve the problem of information distribution. Whether it is a post bar or a channel form, the distribution efficiency is far lower than that of the recommendation algorithm.
Human interests are constantly flowing, and this long-term community needs based on specific interests are constantly being dissolved. I think the form of Tieba is a product of the past era, the efficiency is not high enough, and the needs of solving users’ “consumption relationships” are not perfect enough.
02 Balance between content organization efficiency and community value
Pan Lan: The content organization efficiency of the channel is far weaker than the recommendation, is this the key reason for the decline of successful products on the PC side, such as Tianya, Douban, and Tieba, on the mobile side?
Albert: That’s a relatively major reason. Later platforms have done more in-depth and vertical work in information structuring and user scenario mining. For example, although Douban’s long-tail group remains active, those are relatively low-frequency demands, and mainstream demand has been solved by new products. Coupled with the emergence of short videos, the impact of many factors has been formed.
Cong Guangle: Tieba is a community platform similar to Reddit, with a large number of medium and long-tail ecosystems inside, unlike Discord, which relies on acquaintances and is not completely open. It must rely on an interconnected network environment and search engine, and it is difficult to gather enough users in the closed ecosystem of the mobile Internet.
For content organization, feed flows (whether recommended, subscription-based, or chronological) are more efficient than traditional organization on small mobile screens. The ROI and accuracy of the recommendation system are higher.
Shen Zhenyu: We need to think about what the recommendation solves and what it doesn’t. Tieba is far from just a content distribution platform. In all traditional forums, users are greeted when they post, such as “JRS” for tiger pounce or special titles for Tieba fans. Users don’t just post content, they seek expression, recognition, and interaction. Some people build buildings for celebrities and post every day, and the content value is almost zero, but this is a typical phenomenon of Tieba culture.
Tieba is essentially a form of people’s online expression and community existence. Today’s recommendation engine products (Douyin, Xiaohongshu, Bilibili, Zhihu) are difficult to meet this interactive demand, because users don’t know who they are talking to and feel like they are “talking to big data”. In the age of recommendations, there’s a wall between users and their real audience – recommendation engines.
Our company found that many posts on Xiaohongshu will add a beginning like “Please help me distribute big data to whom, whom, who”, because they don’t know who will see the content. In the era of BBS, we knew exactly who would see the posts, but in the era of big data, this way of communicating with a clear audience was lost.
For PGC creators, they may care more about traffic than who sees their content. But BBS users pursue interaction and communication with specific groups of people, and the number of likes is not important. I’ve been thinking about how to bring back that sense of atmosphere in new products and let users know who they’re talking to.
Song Jian: We should also consider it from the perspective of changes in the creator group. Watching content is to restore energy, sending content is to consume energy, and the cost of interaction is very high. In the past, people created content for free, but now the creator economy has risen, and they can make money on Douyin, Kuaishou, Bilibili, and Xiaohongshu. High-quality creators are chasing business opportunities, which weakens the enthusiasm of those who purely pursue a sense of honor. So BBS now lacks relatively high-quality content creators, even the interesting Tieba meme 10 years ago seems to be dwarfed now.
Albert: These platform ecosystems are actually the result of producers pursuing a better life. Early vertical communities may have high-quality leaders and topic leaders, but as the platform ecosystem becomes richer and richer, they all flow to more rewarding networks, resulting in fewer and fewer high-quality users who can lead the community atmosphere.
03 Innovation of community product forms in the mobile era
Pan Lan: Same is a community product designed entirely on mobile terminals, which customizes different functions for each channel, such as flooding, checking in, sending photos, emotional interaction, etc. What do you think of ways like Same to customize features for specific channels?
Albert: I was a heavy user of Same in the early days. The early group concept of the same was to solve the problem of content diversity, help users quickly find channels suitable for their own content, and promote the desire to produce. But Same is essentially for the purpose of making friends, and its channels are rarely purely interest-oriented, and more channels with stronger dating attributes are more active.
Same users attach great importance to their profiles and constantly refine them, which is completely different from Tieba. When Same appeared on the mobile terminal, it consciously strengthened the image function, and the browsing experience was very good. However, later Same ran a large number of advertisements, attracting many young users and lacking a good crowd distribution mechanism, resulting in a decline in user experience – for example, post-90s users opened the “selfie control” channel and found that it was all post-05 content.
Cong Guangle: Same originally designed the mechanism of “no comments, direct private messages”, which greatly promoted the transformation of the community into social networking. There are a large number of behavioral channels in Same, not nouns like Tieba, but verbs, such as “take a photo of life every day”, such channels are published frequently, and they are the sharing of personal life and data, which promotes social networking well. The same later had a problem because the likes were changed to paid 0.1 yuan, which had a great impact and destroyed the purity and pleasure of social interaction.
Pan Lan: A wrong revision of a product may hurt users’ vitality, just like users have gone to Reddit after the Digg revision. Although there are many topics on Same, Reddit, and Tieba, the top ten topics often account for most of the traffic, such as Li Yuchun, Warcraft, and the rightmost “Let’s be a little joke”, which is a funny topic with a joke. This phenomenon is like a chatter at the end of the world, the topic is generalized, and the positioning is not clear, and this problem seems difficult to solve?
Shen Zhenyu: This problem is indeed unsolvable in the chronological distribution mode. Tieba has always had “four major bars” (namely: Sun Bar, Kou Bar, Hang Bar, Nuclear Bar; The full names are: Sun Xiaochuan Bar, Chinese Bar, Aircraft Carrier Bar, Nuclear War Shelter), which has changed in different periods. As a forum moderator for many years, I’ve observed that no matter how much the layout is designed, traffic always flocks to one place – users will spontaneously run to the most popular page to post, because that’s where they get the most traffic.
This is unreasonable, content distribution does not depend on content quality, but on where it is posted. With the addition of the top post mechanism, high-quality content is easy to sink, and users who can post can maintain visibility. This has led to a lot of noise and bad money driving out good money, which is the “original sin” of Tieba.
Pan Lan: The content community will always face the problem of bad money driving out good money. Even if the team is larger, the moderation is stricter, and there are more good creators, it is difficult to stop the community from going in a bad direction. With the increase of users, more and more content comes from handling and reprinting, differentiation is reduced, and information matching methods are not efficient enough, which naturally leads to user loss and a decrease in original content, further aggravating the dependence on transportation. Is there a solution to this vicious circle?
Song Jian: There are two angles to this: one is that user tastes have improved, and the past content is only good because it is fresh; Second, people will become more and more rational, profit-seeking and lazy, and it is the law of economics to obtain the maximum traffic at the lowest cost. This requires high platform operation capabilities and is not unsolvable, but comprehensive operational capabilities are needed to combat this commercialization trend.
Pan Lan: Some people mentioned that the administrator of Tieba has a lot of power, and traditionally the moderator is responsible for maintaining the order of discussion, selecting high-quality content, and conducting reviews. Today it seems that all of this work has been taken over by the platform. Is it necessary for forum admins to exist today? What is the role?
Cong Guangle: It is difficult to govern a community with only a single character or a few people because of the high complexity of the problem and the lack of a systematic restriction mechanism. Nowadays, new community organizations are more likely to adopt contractual agreement management, with a large number of procedures, judgment criteria, and proof-of-work systems behind them. The community also needs to have an entry barrier, and now many products can be directly joined. The mechanism of human-machine combination and certain rules are more helpful to the overall atmosphere and group collaboration.
Albert: The most important thing is the issue of interests – how are the benefits distributed between the platform and the manager? Why can managers benefit from running their own groups? Looking at the guilds in the live broadcast industry, these organizations evolved from the private sector in the YY era and have a strong relationship with the distribution of benefits.
04 Bilibili and Xiaohongshu: A contest between two community models
Pan Lan: Bilibili and Xiaohongshu are both close to 80 million daily active users (2022), but the patterns are very different. Station B focuses on cultural and spiritual entertainment, while Xiaohongshu focuses on food, clothing, housing and transportation in real life. Spiritual consumption is less frequent and smaller than material consumption, because most people are more concerned about how to improve their lives. What do you think of these two models?
Song Jian: What we do is a “sense of gain”, the crowd is men aged 30-60, who need to talk about capital and social capital, and prefer practical cultural content. There are too many alternatives to culture – you can read “Dream of Red Mansions”, watch TV series, or play games. The practical needs of decoration and buying a house are very clear, and efficiency products can always run out clearly. We are products that do practical knowledge and solve outstanding problems of specific groups of people, not particularly metaphysical.
Shen Zhenyu: Making a product that can realize product ideals and solve user and social problems is already very fulfilling and valuable. Whether it is Xiaohongshu’s idea of building an urban community or Chen Rui’s desire to be a pioneer platform for young people, they have been realized to a considerable extent, and they are all great products.
From the perspective of objective data, after Douyin reached 400 million DAU, Xiaohongshu still grew from 3-40 million to possibly 80 million DAU, indicating that its content has vitality. If the content is completely covered by Douyin, it is impossible to continue to grow.
I think Xiaohongshu’s UGC content ecosystem is the best among all social platforms now. Although Bilibili has strong community attributes, how many people can really create content? How many long videos can be made? The traffic of Bilibili is extremely concentrated in the head UP owners, and the income of non-head creators is very poor. But Xiaohongshu is different, many colleagues in our company casually post some life experience on Xiaohongshu, and thousands of fans pay attention to it, and they can even receive business orders to monetize. It has a very strong UGC ecosystem, which is a scarce resource that Douyin, Bilibili, and Zhihu cannot match, and may become a winner and loser that distinguishes it from other platforms.
05 Real-time interaction needs and the evolution of community forms
Pan Lan: I remember that Douban tried to automatically form a real-time chat room in 2015-2016 to allow users to discuss in real time, but the effect was not good. Would it be better if every topic in the post bar or forum was turned into a chat room? Is this non-friend-driven, interest-based IM group chat viable? Didn’t YY start by teaming up in games? Discord is a similar model, right?
Cong Guangle: Many products have done this kind of experiment before, and Same has also added a live broadcast function that allows users to join the room for real-time discussions. There is a product in Taiwan called BAND, which pairs each community with a real-time group chat. I think it is difficult for a large community to build a simple real-time group chat, because everyone is in the same channel, and it is not the content that is delayed, and the number of people exceeds a certain amount, which is very ineffective.
Discord’s approach is to break up the channel, not only to retain the real-time group chat, but also to turn the group chat into delayed content precipitation and divide it into different sections, so as to support the scale of thousands or even tens of thousands of people. For a large-scale topic community like Tieba, it may be feasible to establish a real-time group chat on a small scale, but it will not work at all on a large scale, unless the structure of Discord is adopted and the real-time discussion mechanism of the live broadcast room is adopted, organized by a small number of people, and a small number of people participate.
Pan Lan: Is it necessary to create a Chinese version of Discord? What should the Chinese version of Discord do? Is QQ channel in this direction?
Song Jian: This type of product is very dependent on timing. YY voice also comes up at a specific time, and there is complex logic behind it. Overseas is multicultural, social interaction is more scattered, and the game’s built-in voice support is not good, so Discord has a chance. But in China, QQ groups are easy enough to use, and the group chat and voice tools that come with the game are also enough, and I don’t think a separate Discord product is needed.
06 The difference between community and community: QQ channel and Discord mode
Pan Lan: Let’s talk about the differences between communities and communities.
Shen Zhenyu: I have an extreme judgment, I think QQ may directly become Discord, and QQ channel may eat QQ’s own chat scene. QQ has been extremely embarrassing in the WeChat era, although we once thought that young people were still using QQ, but in fact, children are now using both QQ and WeChat together – WeChat solves the problem of communication with parents and teachers, and QQ solves the problem of interests and hobbies. QQ has become a solution for online interest socializing, and most of the user’s communication is carried out in group chats.
However, traditional group chats have obvious channel noise problems – in a crowd of 5,000, 5,000 people can hear you say a word, but you will also hear the noise of 5,000 people, and the signal-to-noise ratio is extremely poor. Chat is a very interesting thing, I have considered using the recommended method to solve the signal-to-noise ratio problem in the chat, and the “flying chat” incubated within Byte has also tried to solve it in this way, but it has proven to be unfeasible. Chat is all about real-time interaction, and users need an affirmative or negative response seconds after sending out content, which is an experience that is unparalleled.
Live chat and real-time interaction are devastating in a community or community. The decline of Tieba is largely related to the popularity of real-time chat – posting on Tieba for three days without waiting for a response, and then roaring in the group and immediately someone replying, this experience difference is a dimensionality reduction blow in several dimensions. The problem becomes how to solve the noise problem, because the biggest bottleneck in chat is the maximum number of people. WeChat groups are still limited to 500 people because the noise problem is extremely painful, especially in the era of mobile Internet, active groups will cause mobile phone power consumption problems.
Discord is currently the best solution to the problem of community noise. I have thought about many methods, including recommendations, game rules, etc., but in the end I have been denied because they are either too complicated or cannot solve the problem. Discord largely solves the signal-to-noise ratio problem through permission management, grouping, bots, rules, and plug-in mechanisms. China also needs such a product, and the most likely one to be made is QQ.
Pan Lan: Why is QQ still used by so many people today? Many PC products can only be logged in through QQ, especially Tencent’s historical games such as DNF (Dungeon and Fighter), as well as many mobile and PC games. In addition, teachers and students must use QQ groups instead of WeChat groups, because QQ has richer management permissions. What is the difference between Abot, community and community?
Albert: I think the community is an intermediate form, not on the road to the content platform, but on the road to vertical media. For example, the head UP of Bilibili accounts for a high proportion of video playback, which means that it is developing in the direction of vertical media – a scarce content presentation platform for vertical people. The group is the structure of the consumer relationship, not the structure of the consumption content.
As a loser who started doing real-time interaction in 18 years, we also tried to solve the problem of real-time interaction with the rules of the game. The core of short videos is to solve the problem of connecting strangers. If you want to do a consumer relationship in a stranger scene, you must first solve the problem of efficiently connecting strangers. We tried to expose ourselves more through game mechanics, but it was costly for users and not the best form. One of the things Discord does well is that the screening is done well, and the relationship screening is done in advance, while QQ is very inefficient because of the large traffic.
07 The contradiction between closed communities and content distribution
Pan Lan: A closed circle like Amino, if you don’t enter the circle, there is almost no information, how does it solve the problem of closure and distribution efficiency?
Shen Zhenyu: This is essentially unsolvable. Amino is a very active group chat product, with users spending 50% of their time in live chat channels building their Amino based on vertical interest topics. However, this leads to traffic concentrated in the head sector, and user distribution efficiency is not high. Therefore, Amino mainly spent 2019-2020 doing short video information flow, returning to short video information flow to improve users’ discovery efficiency between different Aminos.
Atmosphere or community belonging is a double-edged sword – if you want a strong sense of belonging, you have to accept closure; to maintain the sense of atmosphere, you have to endure the coldness; To gain more exposure and traffic, you need to break the closed circle and embrace the flow of information. This is unsolvable in terms of mobile phone interaction and distribution efficiency.
Each device naturally determines which distribution form and content form is more suitable. From the perspective of bytes, Zhang Yiming’s biggest discovery is that feed streams are the optimal solution for mobile Internet (note that feed streams are not necessarily recommended). The mobile phone screen is so large that there must be an information flow to solve the complex information distribution problem. The main entrance of the Tieba era is the search box, Tieba is the optimal product of the search engine as a device entrance, and the recommendation and distribution of short videos based on feed flow is the best product of the mobile phone era.
Engaging in a closed community on mobile phones, and opening the channel first every time you consume content, is naturally unreasonable and inefficient. Low efficiency means low reach, low content distribution, and weak overall interaction, so it will not become the mainstream form in the mobile Internet era.
Pan Lan: In the past few years, community products that have been able to achieve large-scale daily activity have finally relied on short videos and recommendation algorithms, and Xiaohongshu has also solved the problem of short videos to a higher level.
08 Exploration of the path of community commercialization
Pan Lan: Whether it is Zhihu, Weibo or other platforms, the commercialization of content communities is a chicken feather. What successful community commercialization cases do you see?
Song Jian: The community does not have such a strong particularity in commercialization. Game products look at user payment, and advertising products look at the media market and advertising originality. Many so-called community problems, such as those in the sports community, may not be large enough (advertising flywheels cannot rise) or advertising is too alienated. The difference between Bilibili and Xiaohongshu is that the video duration is very long, while YouTube is not a strong community but monetization is just like that. I don’t think the word “community” means so much in commercialization, but more importantly, the business form, business attributes, user interaction habits and scale.
Cong Guangle: Is Creative List a community? It can publish UGC content, but it is not purely informational. If you target the community as content, it is the content and advertising model, but this affects the tone of the community. If the community can release matches, manage these flexible workers, or more universal value (such as issuing NFTs) in the future, and let fans or community channels participate in purchases, the commercial value and income of the platform and individuals can be solved.
Communities should not converge around content communities and interest communities, based on the perspective of information content, and the more convergence is inefficient, which is not as good as bilateral platforms. The special feature of the community lies in its group nature and group collaboration, such as “Will the community produce remote work organizations to complete valuable workflows online, such as large-scale design, and share the benefits?” “This is the way out for the community, probably Web3.
Shen Zhenyu: The community is not necessarily a content community, and it may not take content as the core carrier. The core component of a community should revolve around people, not content. Every popular product has a core scene and mentality – Taobao is to buy things, and Douyin is to kill time. Buying things on Douyin and consuming content on Taobao are awkward and unnatural, because the choice is not an efficient solution.
But there are interesting opportunities in the community, and the content scene and consumption scene of some products are combined into one. Some products are naturally suitable for selling in the community, but not on Taobao, such as ski wear and snowboards, because only people who are interested in the community will buy them, and people who love the culture will buy them, and they naturally gather in the community.
Our company (Trendy Toys) was born to answer “how to survive and commercialize interest communities”. We pay attention to the interest of having strong consumer attributes, such as trendy toys themselves are used for consumption. Users in the community are the most direct consumers, and consumption conversion will be direct, natural and fast. Chaowan Qiandao is not only the platform with the highest number of trendy toy enthusiasts in China, but also the largest second-hand trading platform for tide toy and the second largest first-level retail platform for tide toy. These three overlap in our case, because in these interests, consumption and content are naturally one.
Now the interest community is different from back then, and the interest itself has changed – interest has become consumption, or consumption has become interest, which makes commercialization less laborious.
Su Qingyang: The community is divided into three levels: the first level satisfies the content and information value, the second level satisfies the social and companion value, and the third satisfies the cultural value. The live broadcasts, goods, advertisements, and fan monetization that we see every day all revolve around the value of content and information, which is simple traffic monetization. Fan monetization is still classified as a content service because the social relationships established by the content community are recognition of the informative value of the content.
Bilibili, Douyin, Kuaishou and other platforms have not found a business model that transcends traffic itself because it has not yet reached the level of providing social value. The social value here refers to showing off social attributes, such as QQ show and gel. Fashion trends have not declined, but they have not been well accepted in any community. There is huge commercial space for community-based social value and companionship value.
Cultural values also have commercial space. Nowadays, young people’s consumption habits are completely different, in the past, they pursued functional attributes and cost performance, but now there are more and more non-standard products, and the demand for the ostentatious value of standard products is also higher (such as ordinary Nike shoes are not enough, limited editions). Young people deliberately pursue social ostentatious value and scarcity, which requires different cultural identities to give goods, which is Chao culture. Young people need to highlight their values and circles through cultural attributes, and limited editions and collector’s editions can allow them to improve their status and personality in the circle.
Taobao and JD.com can only sell standard products, and a large number of non-standard products cannot be sold, and non-standard products are often subsidiary products of culture and must be combined with community culture. Commodities are culture, culture is commodities, and there will be a new type of community based on the trend culture of young people, and the community carrier is pinned on things. This type of commodity has no advantage in Taobao, JD.com, and Pinduoduo, and must first build a community to sell in the community, relying on community circle attributes and cultural attributes to sell.
At present, most communities are still in the stage of content value monetization, and there is hope to reach the realization of social attributes or even cultural attributes. I am more optimistic about Bilibili because it may derive its own cultural attributes, which is difficult for Xiaohongshu to do.
09 Creator ecology and community development
Pan Lan: How to make creators have a presence in the community? Only by making UGC ecological can the community break through the ceiling. Many community platforms have not solved the creator problem in the end – Zhihu commenters do not have a complete personal homepage and no fixed traffic fans, how to monetize and get positive feedback?
Shen Zhenyu: It depends on what kind of community. If it is Baidu Tieba or Douban Group, the creator has no chance and can only leave in the end. The Douban group has a large number of canceled accounts, and there are many legends of the gods in Tieba, but they are no longer there. The reason is simple: they are not rewarded with effective commercial benefits, and their traffic is completely incapable of competing with “stickers” users who take advantage of community rules. Good content can neither gain higher exposure nor obtain commercial benefits, so it is impossible to retain high-quality content and eventually become a tool for atmospheric feelings.
If Xiaohongshu, Bilibili, and Zhihu are also regarded as communities in a broad sense, especially Xiaohongshu, their creative ecology should be better. However, it may not be appropriate to discuss creators in the community, and the goal of the community is not necessarily to keep creators active and earn money. The real value of the community is to build a “city-state” where everyone can obtain spiritual needs, and creator activity and commercialization are the means to help achieve this goal, not the only option or condition. There are more possibilities in the community, and it may not be necessary to give creators such generous rewards.
Su Qingyang: The core of the content-based community is to establish an upward channel for content and an upward channel for people, and as for monetization incentives, if the two natural upward channels are not built, it is basically in vain. The ascending channel has a great relationship with the algorithm. The interactive community can only rely on praise, attention and return visits, and the platform can’t help him, but the product can do something, such as the rightmost god comment, NetEase Cloud’s hot comments, to highlight the user’s interactive content and give exposure opportunities. Creators in forum-type communities want to rise, so they can only “kneel to the moderators”.
Song Jian: We must regard business behavior, transaction behavior, and advertising as a service, especially for users, and form a habit of promoting transactions. Many people say “never advertise”, implying that advertising is a burden, but good advertising is a value to users. To find ways to facilitate transactions and services for creators and users, the community should think about this from day one. Good videos have been tried a lot lately, don’t think it’s something embarrassing to talk about. It is also a good behavior to provide users with good advertisements and let users tip the author. Users can also be happy to spend money, which can bring dignity and benefits to their favorite authors. The platform should not regard this as a burden or harm, it is beneficial to both parties and can make the ecology stable for a long time.
Albert: The game example illustrates this very well. When doing MMO or SLG games, how to ensure the experience of Big R (the person who recharges the most)? The answer is: the only way to make big R more enjoyable is to make the server more active small and medium-sized R. In the same way, the core of solving the creator’s problem is to make his content consumed by more people, so that his creative motivation and desire to produce can be guaranteed. That’s why the community goes to content platforms, because only in this way can it help creators continue to gain more recognition.
Su Qingyang: There are different opinions, and the creator’s pursuit of identity value is not absolutely dependent on scale. That’s why outside of big content platforms, there is always an opportunity to exist. The value of identification is different, and the value of Zhihu’s approval, Douyin’s red heart, and Xiaohongshu’s like value is different. The creator is happy to get 10,000 likes on Xiaohongshu, and the Kochi blogger is satisfied with 1,000 likes on Zhihu. This is not about the absolute scale, but about the quality of the interaction identification. Social currencies are similar to real currencies in that they have different “exchange rates” for different interactions and recognition behaviors. If I am a two-dimensional person, the likes of three-dimensional users feel different from those of two-dimensional likes. So small communities always have opportunities.
Albert: I agree with Qingyang and add that the key is not absolute numbers, but perceived growth. If he is happy to get 20 likes today and hopes to try to get 50 next week, 200 a month later, but the community begins to shrink and no more than 200 likes are expressed, a fatal problem for the creator, he will seek where to get 400 likes. What is needed is to be able to perceive and gain more recognition.
10 Comments on the community’s success
Pan Lan: Some communities start by moving, but when the scale reaches a certain level, original content will naturally appear, which seems to be a common law.
Shen Zhenyu: This is not necessarily, many moving communities want to be original but have not done it.
Su Qingyang: The success of commentary communities such as Connotation Jokes, Far Right and NetEase Cloud Music does not depend solely on handling. Taking NetEase Cloud Music as an example, our community is essentially based on comments and interactions. Connotation jokes and the far right should consciously bless users who are better at commenting and actively interacting. This interactive content superimposed on the carrying content creates a chemical reaction that makes the content more interesting.
Originally, I might not have thought of a certain meme when I watched a video, but after the appearance of the divine commentary, it gave the content new value. The same goes for the music comment section. There are two categories of creators: content producers and interactive expressers. In most communities, interactive expressers are ignored, but these platforms amplify the presence of interactors – NetEase Cloud Music’s hot comments, right-hand comments, and connotation jokes are exposed, so that those who were originally small and transparent on the Internet can suddenly be noticed by many people, which is very attractive to them. This mechanism allows the community to have a large number of interactives, and also gives new value to the content and attracts users to it.
Shen Zhenyu: Today’s opportunity is rare, and the chat room gathers the product leaders of China’s three most successful comment-centered products – NetEase Cloud Music, the far right and the connotation joke. This is a very interesting phenomenon: users’ secondary processing and secondary creation of content have derived a very unique content ecology, which is different from the traditional post building or Xiaohongshu and Douyin’s people-centered feed release, but has produced its own community culture. How did this culture come about? I was curious as a bystander.
Su Qingyang: Taking NetEase Cloud Music as an example, before 2017, the music player did not have a comment function, or there was no setting to like comments. In 2017, NetEase Cloud made two functional iterations: the likes and hot comments functions for the comments themselves, which doubled the number of comments several times.
Interactive communities need resource support. For music, resources are music. Cloud Music has always insisted on distributing niche music, including music UGC playlists, so that niche boutique music such as European and American music and folk songs can be distributed. Its distribution strategy and resource allocation strategy allow specific groups of people to be screened out – cloud music was used to dominate the world with European and American music and folk songs in the early days, attracting a group of people who love these music and have high music taste. When they gather in the comment area, a specific cultural atmosphere is naturally formed, which is to gather people first and then have a community culture. The mechanism of hot comments and likes on comments amplifies the desire to express and forms a good community atmosphere.
11 The future situation and development direction of the community
Pan Lan: What might the future situation of the community be?
Su Qingyang: The future is trendy players!
Shen Zhenyu: I feel that the next few things are inevitable:
Higher information efficiency. Tieba is not developing well in the mobile Internet era, largely because the information density is not high enough and the acquisition efficiency is poor. Information efficiency is the goal that Internet products continue to pursue and will continue to approach the limit.
Stronger real-time interactivity and feedback. The biggest difference between a community and a content platform is that creators are extremely demanding feedback. I create in this community, not in that community, and the core is to see what kind of feedback I get. Community users have always sought stronger, more direct, and more resonant feedback.
The functional attributes of interest communities will become stronger and stronger. In the BBS era, a platform has to solve all problems – content release, information acquisition, comment interaction, and the burden is very heavy. Nowadays, there are various vertical special products (Taobao shopping, Xianyu second-hand, Douyin short videos) that divert many content consumption scenarios, making the burden of interest communities smaller and more focused on solving the problems of interest itself.
For example, Picture Worm solves the photographer’s need for photo hosting and compliments, we trendy players do second-hand transactions, and script killing is a scoring system. These are focused on specific vertical scenario needs. General needs have been solved, while special needs of interests and hobbies are often not solved by general platforms, which is an inevitable trend of vertical instrumentalization. Discord is a toolbox that provides open interfaces for developers to provide solutions for various vertical scenarios. More interest-based solutions will emerge in the future.
Su Qingyang: There are several trends in the future development of the community:
Device migration: Changes in content carriers may give birth to new communities, such as VR and Web3 applications.
Scarce resources: Communities that focus purely on content information value will be eaten up by the top few large communities, which is a black hole effect. Bilibili, Kuaishou, and Douyin occupy the video field, while Xiaohongshu and Zhihu occupy the graphic field. If your vertical communities are still based on content value, they will not be able to beat them because they have enough traffic and can also be vertical through distribution and search.
However, there is a type of community with exclusive scarce resources, such as the scarcity of trendy toys based on commodity attributes. Its commodity attributes, content attributes and cultural attributes are a combination of three things. Even if Bilibili or Xiaohongshu provide high information value, it will not lead to user migration because there are unique people and things here. The scarcity of the obtained object is based on the appraiser.
I say that the trendy toy is the future because it represents the model of “future commodities are culture, and culture is commodities”. In the future, the form of community will not necessarily be expressed by content, but by goods, which is very common among young people.
Pan Lan: How much does it cost to be a community? It’s so difficult.
Shen Zhenyu: We recently found that the business of burning money is unhealthy. Some people ask how to acquire customers, in fact, we do not spend money to attract users through products, tool values, and transaction services. Successful communities don’t burn money. Bilibili burned money after it was successful, and it was not burned before success. By 2014, after entering the mobile Internet era, it had formed its own spiritual and cultural symbols and a strong community atmosphere, and the barrage culture had taken shape.
Pan Lan: Give me an example of non-PC successful communities not burning money after 2015.
Shen Zhenyu: It still depends on the investor’s father, don’t think about it without the investor’s father. However, the connotation joke is half a joke, and since 2015, there has been no external investment for three consecutive years, and it is self-fending because the company feels that it is not worth burning money. But during that time, it gradually precipitated its own unique community culture.
Community culture is formed through long-term user screening in the process of organic growth and operation – unsuitable users are eliminated, suitable users are retained, and a unique culture is formed after long-term precipitation, and then the outward tension of this culture continues to expand.
Some people in the group mentioned that Xiaohongshu didn’t burn much money in the early days. When I did the Hong Kong shopping guide in 2013-2014, it started from a small community, and with the founder Wang Chao and his friends gradually expanded, I didn’t burn much money in the early days, and it wasn’t until 2018 that I started to invest in growth.
Pan Lan: There is a key point in the connotation joke that is the first to embrace shorter vertical screen short videos. Although I also realized in 2016 that visualization is the future, adding images, short videos, GIFs and other elements, it does not do autoplay directly on the homepage like it does, all vertical screens, and integrate it into the feed.
12 Thinking about the essence of community development
Shen Zhenyu: I would like to emphasize that the formation of community culture is not obtained by investment. Community culture is slowly formed through internal screening in the process of organic growth – unsuitable users will be eliminated, suitable users will stay, and a unique community culture will be formed over time. This culture then creates outward tension that helps the community continue to expand.
At the heart of community success lies finding its unique niche and value. Communities like Trendy Toys that combine product attributes, content attributes and cultural attributes can remain competitive in an environment dominated by large platforms because they provide value that cannot be simply replaced.
In the future, communities will be more vertical and instrumental, focusing on solving the unique needs of specific interest groups. From a business perspective, communities that can combine content consumption with commodity consumption are more likely to achieve a sustainable commercialization path. The charm of a community lies in its ability to provide users with a sense of belonging and identity, a core value that cannot be replaced by any recommendation algorithm.
Su Qingyang: I would like to add that a successful community needs to balance three levels of value: content information value, social companionship value and cultural value. Most platforms only focus on the first layer, and communities that can provide higher-level value can form true differentiated competitiveness.
Young people’s consumption is shifting from function-oriented to cultural identity-oriented, which provides new opportunities for community commercialization. When a community can cultivate its own unique culture and make it touchable and owned through physical goods, it can form a self-circulating ecosystem, which is the success of vertical communities such as hipsters.
The future of the community is not about simply copying existing models, but about identifying emerging needs and building differentiated value. No matter how technology evolves, the need for belonging and identity will always exist, and communities that can meet this need will remain viable for years to come.
