Zhihu Product Analysis Report

In 2021, Zhihu celebrates its tenth anniversary, and its slogan is officially updated to " Where there is a question, there is an answer ." Let’s use this sense of ritual to analyze Zhihu.

Table of contents:

1. Industry Analysis

1.1 Industry Background

1.2 Market size

1.3 Development Trend

2. Product Overview

2.1 Experience Environment

2.2 Product Positioning

2.3 Iteration Path

3. User and Demand Analysis

3.1 User Characteristics

3.2 User Profile

3.3 Demand Analysis

4. Product Analysis

4.1 Business Logic

4.2 Business Process

4.3 Functional structure

4.4 Typical functions

5. User Feedback and Improvement Suggestions

5.1 User ratings and reviews (App Store)

5.2 Improvement Suggestions

VI. Summary

1. Industry Analysis

Let’s first take a look at the industry background, market size and development trends of the knowledge payment industry in which Zhihu is located.

1.1 Industry Background

1.1.1 Development Stage

1.1.2 Driving factors

The rapid rise of the knowledge payment industry is mainly due to four factors:

1. Demand side - knowledge anxiety : The information explosion in the Internet era has led to redundant information and high screening costs. Users hope to obtain high-value knowledge (i.e. information that has been refined and summarized), social topics and class identity by paying.
2. Demand side - payment habits : Mobile payment has matured and become popular, and users have formed the habit of paying by scanning.
3. Supply side - development of the pan-entertainment industry : The carriers of knowledge content have been enriched with the development of the pan-entertainment industry, from text, images, audio to video, meeting the users' diversified and fragmented needs for acquiring knowledge.
4. Supply side - Big V effect : In the era of free knowledge sharing, outstanding creators become big Vs in their fields through high-frequency and high-quality output. Big Vs are in urgent need of monetization methods, while their fans trust and admire them. Paid knowledge products just meet both needs.

1.2 Market size

Since 2016, the knowledge payment industry has experienced rapid growth. In 2020, although the COVID-19 pandemic caused the real economy to shrink, social phenomena such as "involution" intensified users' knowledge anxiety, bringing benefits to the knowledge payment industry. According to iMedia Data Center, in 2020, 63.1% of Chinese users purchased paid knowledge products, and the market size is expected to reach 39.2 billion yuan.

On the other hand, leading pan-entertainment platforms such as Douyin, bilibili, and WeChat Video are also actively exploring knowledge payment.

Therefore, we have reason to speculate that the knowledge payment industry will continue to maintain rapid growth in the foreseeable future.

1.3 Development Trend

1. Video content : The short video trend exploded in 2018, cultivating Chinese users' behavioral habits of using videos as a medium to obtain information and the user mentality of using fragmented time to watch videos. Therefore, in response to the changing demand, the knowledge payment industry explored the medium-length video ^[1] content based on the fact that knowledge content is more serious and contains more information than entertainment content. Typical representatives include Zhihu's "video" module, knowledge-based up-loaders on Douyin/Bilibili/WeChat Video Account, etc.
2. Category verticalization : With the increase in the number of players entering the market, the content coverage of the knowledge payment industry is wide and highly homogenized. Small and medium-sized enterprises have begun to adopt a verticalization strategy, striving to gain a share of the market by deepening the knowledge content of vertical categories. Typical representatives include "Dingxiang Doctor" in the medical and health category, "Babytree" in the women's parenting category, "Logical Detective" in the life encyclopedia category, "Cloud Smart Agriculture" in the agricultural course category, "Everyone is a Product Manager" in the product knowledge category, "Bird Brother's Notes" in the marketing knowledge category, and so on.
3. User penetration : The fragmented and entertaining learning habits will make it possible for knowledge payment to penetrate into third-, fourth- and fifth-tier cities in the future. For these users, their motivation for learning is mainly to have fun or solve life problems, and their preferred content is mainly interesting, easy-to-absorb entertainment content or common sense content.

[1] Medium videos generally refer to videos with a length of 1-30 minutes, which is between long videos and short videos.

2. Product Overview

2.1 Experience Environment

• Product version: Zhihu V7.0.0
• Mobile version: iOS 14.1
• Phone model: iPhone 7 Plus
• Experience time: 2020/1/25

2.2 Product Positioning

Zhihu is a knowledge sharing platform launched in July 2011. Zhihu is a free question-and-answer community at its core, while also providing knowledge and information services such as columns and e-books. Since 2016, Zhihu has actively expanded its knowledge payment business, and gradually launched various paid business forms such as "Zhihu", "Zhihu Live", and "Private Class" paid columns (later changed to "Paid Consulting").

Zhihu’s logo is a blue background with white Chinese characters “知乎”. Blue is the color of rationality, wisdom, and technology, which fits in with Zhihu's original professional and high-quality Q&A content and elite circle atmosphere.

Due to the strategic needs of growth and monetization, Zhihu's positioning has changed several times. In 2021, based on its tenth anniversary brand renewal strategy, Zhihu launched version V7.0.0 on the App Store and introduced a new slogan "If you have a question, there will be an answer." This version focuses on interest/life-related Q&A and social hot topic discussions, and provides video and social modules. Next, we will mainly discuss and analyze the iOS V7.0.0 version.

2.3 Iteration Path

We divide Zhihu's version iteration path into five stages, and the specific changes in each stage are as follows.

Phase 1 (July 2011-February 2013): Feelings | A small and beautiful elite Q&A circle

“Discover a bigger world.”

In July 2011, Zhihu was launched. At that time, Zhihu was benchmarked against the American question-and-answer SNS website Quora, and was known for its high-level and professional answers. Compared with mainstream question-and-answer platforms such as Baidu Knows and Sina iAsk, Zhihu has the following characteristics:

• The problem categories are verticalized and segmented, and IT and Internet related problems are particularly abundant;
• The answers are of high quality and professionalism, thanks to Zhihu's excellent seed users and the extremely high user entry threshold brought about by the invitation and review system;
• We encourage an open and dialectical philosophical and scientific spirit of thinking. We do not highlight the correct answer through the subjective judgment of the questioner, but rather sort the answers by data such as the number of likes, comments, and reposts. There is no single best answer.
• Connecting knowledge and people, using knowledge as a starting point and subscription, following and other functions as a bridge, incubating and empowering outstanding creators (big V);
• Connect people and create a community atmosphere through functions such as likes, comments, reposts, follows, and private messages.

Phase 2 (March 2013-April 2016): Growth | An approachable Q&A sharing platform

“Share your knowledge, experience, and insights with the world.”

In 2013, Zhihu opened for registration . In 9 months, the number of registered users grew from 400,000 to 4 million.

At the same time, Zhihu's slogan has also changed from "Discover a bigger world" to " Share your knowledge, experience and insights with the world ", emphasizing the spiritual core of exploring and sharing knowledge, and encouraging users to actively participate in content production and discussion.

In terms of channel communication, Zhihu has begun to expand information circulation channels, transforming from closed content to open content. Users can reach Zhihu through search engines, social network sharing and other channels.

In terms of product functions, Zhihu focuses on polishing its core question-and-answer module, and gradually improves system stability and product usability while ensuring usability. In addition, based on the "popularization" route, Zhihu launched the " Random Look " function to lower the threshold for new users to obtain information (that is, users do not have to have specific questions/follow big Vs to search/be recommended relevant information).

Phase 3 (May 2016-April 2017): Monetization | Diversified knowledge sharing platform

“Share your knowledge, experience, and insights with the world.”

Since May 2016, Zhihu has begun to actively explore monetization models other than commercial advertising, and has successively launched Zhihu Live , Zhihu (later changed to paid consultation) and Zhihu Bookstore , trying to build a knowledge payment matrix.

The significance of doing this is:

• A healthier way to monetize : Commercial advertising is contrary to the nature of Zhihu as a knowledge sharing platform, while the paid knowledge business can achieve healthy and sustainable monetization without harming user value;
• More excellent creators : Attract excellent creators from other platforms to join, encourage them to continue to produce high-quality content, and increase the stickiness of new and old creators to the platform;
• More diversified ways to acquire knowledge : meet users' needs for acquiring knowledge in more scenarios, thereby increasing user stickiness to the platform.

Phase 4 (May 2017-November 2017): Entertainment to Death? | A Pan-Entertainment Knowledge Social Platform

"Live interactive Q&A, a social platform for sharing the hottest knowledge." >> "Interactive Q&A, a knowledge social platform for sharing ideas."

The core of content products is to seize user time. The number and time of sinking users are far greater than the elite users in first- and second-tier cities, making them a battleground for content products.

Therefore, in order to continue to expand market share and improve return on investment, Zhihu adjusted its positioning again:

• Pan-entertainment : Transformation from niche serious knowledge to more popular pan-entertainment content;
• Strong socialization : Enhance social attributes and try to get a share of the business of Weibo, Douban and Toutiao.

Since May 2017, Zhihu has successively launched functions such as " Topic Plaza " push, Zhihu News , Daily New Knowledge , and " Ideas ", and enriched the content of topics in the film and television variety show module. Some of these features have been successful in the market and have been retained to this day, such as "Topic Plaza" (later upgraded to "Hot List") and "Ideas" (similar to Weibo updates); some have failed in the market and disappeared, such as Zhihu Information and Daily New Knowledge.

Phase 5 (December 2017 to present): Going Downward | A Knowledge-Based Social Platform for Daily Life

"If you have any questions, go to Zhihu." >> "If you have any questions, there are answers."

Since the end of 2017, Zhihu has continued its strategic transformation towards pan-entertainment and strong socialization, while adhering to its core values ​​of knowledge exploration and sharing to consolidate its question-and-answer content barriers. There are three main changes during this period:

First, actively expand online literature business and seek incremental users in the sinking market.

Second, enter the field of video content , cater to the development trend of video, and expand the richness and fun of content.

Third, increase the breadth of Q&A content. As the "moat" of Zhihu, the Q&A content has become more and more down-to-earth. In the current version, it mainly focuses on the following two types:

• Interest life : including interest-based Q&A with circle divisions as the entrance, as well as life-related practical Q&A with emotions, workplace, and parenting as the core content;
• Hot Discussions : Real-time discussion and timeline tracking of hot events in various fields such as social events, film and television variety shows, technology and digital, games and e-sports, education and workplace.

3. User and Demand Analysis

3.1 User Characteristics

3.1.1 Creator characteristics

Zhihu has many creators with solid professional knowledge and excellent output capabilities.

In January 2021, according to Zhihu, there are currently 100 creators on Zhihu with a monthly income of more than 100,000 yuan, and 1,000 creators with a monthly income of more than 10,000 yuan.

We classify these 100 and 1000 respondents as top V and waist V respondents, respectively. They have the following characteristics:

• Top V : number of followers ≥ 1 million. Such as literary giant Zhang Jiawei, Internet giant Li Kaifu, pianist Lang Lang, sociologist Li Yinhe, and psychologist Li Songwei. Most of them are well-known elites in the industry and Zhihu natives. Not only do they output high-quality content for Zhihu, they also provide strong endorsement.
• Mid-level answerers : The number of fans is 500,000 to 1 million. Compared with the top Vs, mid-level answerers are mostly less well-known industry experts or self-media writers. They have relatively low loyalty to Zhihu and are more profit-oriented. The content output capabilities of waist-level big Vs are also excellent.

It is precisely because of these excellent creators on Zhihu that any professional or non-professional, even unconventional questions on Zhihu can get answers full of useful information. Although a large number of creators of varying quality have poured in in the past two years, Zhihu users still have a stronger sense of trust in the quality of content on the platform than on other platforms.

3.1.2 Characteristics of ordinary users

1) Gender distribution: Equal ratio of men to women

According to the China Industry Information Network, as of the end of 2019, the sex ratio of the total population in the country was 104.64 (with females as 100), which means that males accounted for 51.13% and females accounted for 48.87%.

According to iResearch Index's statistics in November 2020, male users account for 51.15% of Zhihu users and female users account for 48.85% . Therefore, the proportion of male and female users on Zhihu is almost equal . As knowledge payment platforms, Himalaya and Duoduo have significantly more male users.

The main reasons are speculated to be as follows: In the iOS 6.39.0 version on April 29, 2020, Zhihu began to promote Q&A in the field of "parenting" and consciously increased the proportion of female users.

2) Age distribution: 20-35 years old are the main force

Combining the iResearch Index and Baidu search, most Zhihu users are between 20 and 35 years old , and the average age is younger than that of Himalaya and Duoduo.

The age of 20-35 is the golden period for cognitive growth and career development, which means that most users tend to use Zhihu to meet their needs for acquiring knowledge and personal growth . But at the same time, the overall age of users is young, so it is speculated that some users also like to engage in social activities and participate in discussions on hot issues on Zhihu.

3) Education level: mainly high education

According to iResearch survey data, the average education level of Zhihu users is very high, with 80.1% of them having a bachelor's degree or above. During the same period, the average proportion of Chinese netizens with higher education was only 20.4%. This phenomenon is easy to explain: people with high education tend to be more curious, have strong learning ability, and have a stronger demand for knowledge; and the environment they live in is more "competitive" and they are more eager to achieve personal growth.

3.2 User Profile

Based on user characteristics, we have sorted out 4 typical user portraits:

*The following user portraits are only for ordinary users and do not include creators.

1. Sophisticated middle class : This group consists of white-collar workers aged 25-35. They have a good income, love life and pursue quality. They love money in a wise way and make this refinement sustainable through financial management. On Zhihu, they read questions about finance, travel, home, and fashion.
2. Future elites : This group consists of new employees aged 23-30. Their goal is to pursue a successful career and they believe that knowledge can help them realize their dreams. I usually have the habit of paying attention to industry information and in-depth reading. Zhihu is an important way for them to learn professional knowledge in depth.
3. Trendsetters : This group consists of college students, postgraduate students or new employees aged 20-28. They have a wide range of interests, like new things and love to follow gossip. I love to play and know how to play. I like to make friends through similar circles on social networks. Although his income is not high, he is keen on online shopping and likes to buy interesting and inexpensive small items. They will browse or participate in hot discussions on Zhihu, follow circles of interest and make friends with like-minded people, and read popular questions and answers (for example: "What are some small things that can improve the quality of life and comfort?").
4. Curious Babies : This group consists of college students or graduate students aged 18-25. They have not yet joined the workplace competition and have no pressure from family. They are often smart and curious, and like to get to the bottom of historical stories, the origin of the universe, natural laws, business laws, etc. On Zhihu, they pay attention to some bizarre questions (for example: "Since cancer cells can proliferate indefinitely, why can't they be used to make meat products?").

3.3 Demand Analysis

We use the classic " user-problem-solution " framework to analyze the common needs of all Zhihu users, as well as the special needs of the four typical user categories mentioned above (sophisticated middle class, future elites, trendsetters, and curious babies).

After summarizing, we can see that Zhihu users have three core needs:

1. The need to answer questions : Users have requirements for information matching, that is, they hope to see customized answers. Willingness to pay is related to the urgency and importance of the issue.
2. The need for high-quality information : Users have requirements for information quality but are not demanding. Low willingness to pay.
3. The need to obtain professional information : Users have strict requirements on the quality of information and hope that it is in-depth and structured. Higher willingness to pay.

There are four categories of non-core needs:

1. The need to communicate opinions : Users want their opinions to be seen so that they can gain a sense of existence and value.
2. Based on the needs of interest-based social networking : users hope to see information that they are familiar with and like, so as to gain the satisfaction of entertainment consumption.
3. The need to see hot topics and follow gossip : Users want to see real-time hot topics such as politics and entertainment that everyone is paying attention to, so as to satisfy their curiosity and desire to explore and gain social topics.
4. The need to explore the unknown : Users want to explore knowledge beyond common sense and gain a sense of satisfaction from entertainment and curiosity.

Interestingly, the above core and non-core demands are aimed at different user groups and have contradictory values. The former are represented by old Zhihu users who seek knowledge (loyal users before Zhihu was commercialized), while the latter are mostly new users who seek entertainment. This is also why old Zhihu users believe that "Zhihu is no longer the original Zhihu" and have criticized the entertainment-oriented content and various commercial advertisements on Zhihu in recent years.

For Zhihu, on the one hand, it is under pressure from investors, and profits are inseparable from traffic, and traffic is inseparable from entertaining content. On the other hand, high-quality and professional content is Zhihu's moat. If Zhihu excessively damages the experience of old users, causing a large number of old users to leave Zhihu, then high-quality and professional content will find it difficult to attract traffic, and excellent creators will leave one after another or start writing content that caters to new users. Finally, Zhihu's content barriers will no longer exist.

On one hand, there is pressure from investors, and on the other hand, there is word of mouth from old users. Zhihu seems not to want to offend either side. This entanglement and vacillation is more concretely reflected in its product functions.

4. Product Analysis

4.1 Business Logic

There are three types of roles on Zhihu. Their names, definitions, and core behaviors are:

1. Ordinary users : content consumers. They will browse the existing Q&A library or ask new questions.
2. Creator : Content producer. They will answer the questions through system recommendations or self-selection.
3. Platform mechanism & algorithm : content maintainer. Through the process mechanism, high-quality questions and answers are weighted and their exposure is increased, while low-quality questions and answers are downgraded and their exposure is reduced; through AI algorithms, questions with similar keywords are recommended to ordinary users, and questions that are suitable or that creators are good at are recommended to creators.

Among them, the role identities of ordinary users and creators can be converted into each other according to the scenario . For example, if ordinary users produce content in their professional or expert fields, they become creators; if creators consume content in fields that they are unfamiliar with but interested in, they become ordinary users.

Based on role categories, role behaviors, behavior sequences, and interactive relationships, we have sorted out the core business logic of the question-and-answer module as follows:

4.2 Business Process

Because Zhihu’s business model is complex, it has many functional modules. Therefore, we only analyze the business process diagrams of two typical usage scenarios of the question and answer module : the question process and the "take a look" process .

4.2.1 Questioning Process

The subject of this process is ordinary users, and the scenario is when the user has a specific question that needs to be answered.

4.2.2 “Just a look” process

The main body of this process is ordinary users, and the scenario is when users casually look at questions and answers that they are interested in or are interesting, and do not have a clear need to answer the questions.

4.3 Functional structure

Zhihu adopts the current mainstream rudder-style navigation structure, which includes five first-level functional modules: Home, Video, Membership, Message, and Mine.

According to the functional structure diagram, we can see the three characteristics of the current version of Zhihu:

1) Enrich and unify the knowledge payment business .

The "Membership" module integrates a variety of paid business forms, including knowledge payment services and e-commerce services (welfare club). Among them, paid knowledge services include e-books (reading clubs, novels), high-quality articles (Yanxuan column), and lectures (Zhihu Live). Most services are subscription-based , and users only need to subscribe to Yanxuan membership to enjoy them for free; a few services appear in the form of flash sales/discounts, and members still need to pay again.

This design approach has two main advantages: first, it clearly displays membership benefits to users, which can increase their willingness to pay; second, paying users are often more willing to pay again than free users, so the payment conversion rate can also be improved.

2) Powerful incentive mechanism for creators .

High-quality content is Zhihu’s “moat”, and high-quality content comes from excellent creators. After experiencing three "big V escapes", Zhihu has increasingly strengthened its support and incentives for creators in recent years. There are currently at least 14 creative activities underway, covering online shopping recommendations, science, social sciences and humanities, games, skill sharing, etc.

In addition, Zhihu also encourages users to produce content through a creator growth system , that is, creators with higher levels can obtain more rights and interests, such as content analysis, live broadcast, product recommendations, appreciation function, roundtable host, Zhihu Live, brand tasks, etc.

3) Active attempts to convert content into videos.

Since June 2019, Zhihu has begun to experiment with turning text and picture content into videos, but it has been lukewarm.

At that time, Douyin and Kuaishou became extremely popular and had entered the 100 million DAU club; in 2020, WeChat launched video accounts, marking that the videoization of content has become an inevitable trend. Although most of the content at that time fell into the category of entertainment consumption, as the red ocean of entertainment content deepened, platforms and creators began to target the blue ocean of the knowledge field. At the beginning of 2020, super knowledgeable Vs such as Luo Xiang and Banfoxianren emerged on Bilibili. In March 2020, Douyin opened 5-minute video permissions to knowledge creators; in October 2020, Xigua Video held the Xigua PLAY Curiosity Conference, and its ambition in the mid-length video track was obvious. The so-called "curiosity" is also aimed at the amount of information and depth of the content, which overlaps with Zhihu's positioning.

Therefore, in the second half of 2020, Zhihu began to work together on both the user side and the creator side to create high-quality video content . For Zhihu, this is more of a "defensive battle" than a "battle of breakthrough" that it has to deal with.

On the user side, Zhihu has opened an independent entrance for the video module to encourage users to browse video content; on the creator side, in October 2020, Zhihu launched the " Haiyan Plan ", investing 500 million in cash subsidies and 10 billion in traffic support to encourage video creation. Since most of Zhihu's original creators are good at text and picture creation, in order to guide them to switch to video creation, Zhihu even provides a creation tool that can convert text and pictures into videos in seconds. His determination is evident.

4.4 Typical functions

4.4.1 “Recommendation” page: pictures, texts, and videos, a hundred flowers bloom

The "Recommendation" page is the default page after entering the Zhihu App. It originally displayed a graphic and text information flow (Q&A/articles) recommended by the algorithm. The current version has added a video information flow, and the content format is richer.

The video information flow on the "Recommendation" page has the following characteristics:

(1) Structural layer

1) Most of the videos are medium-length.

In terms of video length, the vast majority are medium-length videos ranging from 1 minute to 30 minutes.

2) Independent algorithm mechanism.

From the perspective of video types, it has been observed that the recommendation algorithm mechanisms for video content and graphic content are different for the same user. That is, the algorithm will recommend similar pictures and texts based on the pictures and texts that the user has browsed, and recommend similar videos based on the videos that the user has browsed, but there is no cross-recommendation relationship.

Personally, I think this approach is debatable. After all, pictures, texts or videos are just different ways of presenting information, and the user's preference for information content is the core. Therefore, if video/picture and text recommendations are performed based on the same mechanism, the user experience may be smoother.

(2) Framework layer

1) High exposure

Judging from the location and number of videos, the first or second item on the “Recommendation” page will most likely present video content. The number of videos and pictures strictly follows the ratio of 1:6, that is, there must be 1 video after every 6 pictures.

2) Highlight the playback volume information

Zhihu’s traditional graphic and text information flow presents interactive information like this:

That is, the number of approvals and likes for the answer is displayed at the bottom of the preview answer.

The video information flow presents interactive information like this:

That is, the number of views and likes of the video is displayed at the bottom of the preview video. Most video content platforms do not disclose playback information. This information is generally only available to creators for evaluating video quality. Zhihu makes it public to ordinary users, presumably because Zhihu's video module has not yet gained popularity, so by displaying a larger amount of playback information, it can enhance users' interest in watching videos.

3) Nested structure

When we click on a video thumbnail on the "Recommendations" page, we will enter a new video information stream instead of the video playback page under normal logic. As shown below:

In the new video list page, the first video displayed corresponds to the video thumbnail that the user previously clicked. After pulling down, the system will recommend similar videos to users based on the keywords and other information of the video.

There are two problems here:

• The video sizes are not uniform, resulting in an unsmooth viewing experience . In Figures 3-1 and 3-3, the video aspect ratio is 16:9, which can display selected comments and interactive information normally; however, 3-3 has black edges because the original video aspect ratio does not conform to 16:9. In Figure 3-2, the video is in vertical format, and the selected comments and interactive information are not fully displayed, with only half of the text being displayed. The overall viewing experience is not beautiful, smooth, or complete.
• Too much text content . Generally speaking, there are two ways to display a single-column video feed: one piece of content per screen, or multiple pieces of content per screen. The former is suitable for providing users with personalized content and immersive experience when the platform has an accurate understanding of user preferences; the latter is suitable for giving users more choices when the platform is not sure about user preferences. However, Zhihu's approach is to display a video content on one screen, with the picture and text each taking up half. This results in neither providing an immersive experience nor giving users room for choice.

4.4.2 Video Module: Content Defense Fortress

The video module has an independent entrance and is located just after the homepage, which shows Zhihu's determination on video content. Let’s first take a look at the recommendation page, which is very similar to Station B in all aspects:

(1) Scope layer

1) Mainly for entertainment and transportation content .

The author counted the first 28 pieces of content on the video recommendation page. In terms of type, life-related content accounts for nearly half, while the overall ratio of knowledge content to entertainment content is 2:5. In terms of originality, most of the videos are directly copied from excellent creators of other content platforms (mainly Bilibili), and there are few original contents.

If Zhihu cannot differentiate itself from Bilibili in terms of video positioning, its latecomer disadvantage will be quite obvious . First of all, users have formed the habit of watching entertainment videos on Bilibili, and the platform's accurate recommendations provide a very good user experience. If Zhihu also enters the field of entertainment videos, even if it spends a lot of money, it will be difficult to make users' experience on Zhihu better than on Bilibili, not to mention the considerable migration costs. Secondly, Bilibili has formed a good ecosystem for video creators. On the one hand, excellent creators have already gained a large number of loyal fans. On the other hand, Bilibili has a complete creator incentive mechanism, such as short-term theme prize activities and annual top 100 up-master activities. Even if Zhihu invests a lot of money in incentives, it is difficult to buy these excellent creators who are extremely sticky to Bilibili.

2) Barrage function

The barrage function is the core social function of long video and medium video platforms. Its user value lies in:

• Provide a window for users to know each other's existence, reduce loneliness and create a sense of companionship;
• Encourage comments on specific video scenes/plot time points to increase interest and stimulate resonance;
• Provide supplementary or predictive information (such as portals, high-energy warnings, and progress bar warnings) to answer users' current confusion or make them anticipate the next content.

Therefore, regardless of whether Zhihu positions its video business as entertaining, informative, or somewhere in between, the barrage function is reasonable and necessary.

Currently, Zhihu supports the barrage function, the barrage switch button, and the quick barrage function. Among them, the quick barrage function will analyze the current video and recommend corresponding short sentences, which to a certain extent lowers the threshold for users to speak.

(2) Framework layer

1) Single column + double column information flow .

The information display method of the video recommendation page combines single column and double column. After every 3 rows of double-column videos, show 1 single-column video. The two-column information flow displays the video cover image, playback volume, duration, title and type. The single-column information flow replaces the cover image with a video preview, and adds creator information and a follow button.

Generally speaking, single-column information flow is suitable for situations where the platform wants to recommend accurate content to users, and has a low tolerance rate; double-column information flow is suitable for situations where the platform wants to present rich content to users, and has a high tolerance rate.

However, Zhihu actually prefers the application of double-column information flow. The author observed that there is not much difference in the number of views and likes between single-column and double-column videos, so the display method is not differentiated based on video quality. Therefore, the significance of a single-column information flow may be to help creators attract fans faster through video previews and external "follow" buttons, thereby increasing the platform's attractiveness to creators.

2) Nested structure .

Like the video information flow on the "Recommendation" page, the design method of embedding video stream within video stream is also adopted here. I won’t go into details here.

4.4.3 “Circles” module: a place for like-minded people to gather and communicate

The "Circle" module is a secondary functional module under the "Home Page". It is comparable to the Weibo Super Topic Community. It is a secret garden where people with similar interests and hobbies focus on communication. It has strong entertainment and social attributes.

The homepage of the circle module is as follows:

There are three functions from top to bottom:

1) The circles I frequent .

It is presented in the form of a horizontal navigation bar to help users quickly locate a specific circle.

2) Posting function .

The entrance is very simple, including a text box for directly entering text on the current page, as well as shortcut buttons such as "Album", "Photography", and "Vote". Encourage users to post more.

3) Recommended updates .

The content is updates about circles that the user follows or updates about circles recommended by the platform. The display format is a single-column information flow, including title + text content overview + pictures. The author believes that since circles are guided by users' personal interests, as long as users follow one or more circles, the platform should reduce recommendations and focus on displaying the dynamics of the circles that users follow, which will be more in line with users' usage habits.

It is speculated that the main purpose of Zhihu launching the circle module is to increase social activity. However, judging from the current level of activity, the number of active users and interactions in Zhihu circles are far less than those in Weibo super topic communities . Let’s take the Genshin Impact circle as an example. As of January 23, 2021, the Zhihu "Genshin Impact Circle" had 5,203 members and 900 posts; and the Weibo "Genshin Impact" super topic had 260,000 followers and 109,000 posts. The distance between the two is a hundred times. This is actually very easy to understand. Circles or super topic communities have strong entertainment and consumption attributes, which are naturally in line with Weibo's downward and entertainment positioning, and are inconsistent with the tone of Zhihu's knowledge sharing platform.

4.4.4 Zhihu Live: Delicious and Inexpensive “Fast Food of Knowledge”

Zhihu Live is one of the earliest knowledge payment services launched by Zhihu . The background at that time was the rise of various knowledge payment platforms, and Zhihu urgently needed to find ways to help creators monetize in order to attract and retain creators.

Simply put, Zhihu Live is a type of online voice + text live lecture , which also provides recorded playback. The initiator must be an excellent creator (creation level must reach lv7). The Live content covers a wide range of areas, including science, finance, literature, social sciences, parenting and other fields; the content is vertically positioned and has a strong goal orientation. Taking the science field as an example, the special topics include "How to self-study computer professional courses?", "[Innovation and Entrepreneurship Competition] How to find good ideas" and so on. The main pricing is 9.99/19/29 yuan. The affordable price itself is easily accepted by Zhihu's young and highly educated user groups. In addition, most of the content can be enjoyed for free by becoming a Yanxuan member (19 yuan/month). In this way, the role of Live pricing is more to set the user's psychological price anchor, thereby encouraging users to purchase memberships.

Most Live sessions last from 20 to 150 minutes, which can meet users' fragmented learning needs during commuting time, as well as their needs to set aside large blocks of time to focus on learning.

In addition, if users participate in lectures in real time, they can interact with the initiator in the Q&A session and get targeted answers; the initiator can not only earn cash, but also gain a fan base that is willing to pay. Smart initiators will even pull users into WeChat groups in the interaction column to obtain private domain traffic.

4.4.5 Yanxuan Column: Courses and Online Articles “Melting Pot”

In September 2017, Zhihu launched the “Private Lesson” feature. In August 2019, "Private Lessons" was upgraded to "Yanxuan Column", which is one of the core businesses of knowledge payment along with Zhihu Live.

The Salt Selection column mainly contains two parts:

(1) Graphic/audio/video course series.

This type of course is comparable to the GetApp and is known for its professional and systematic knowledge explanation. The target audience is mainly highly educated users who have demands for information density and structure. The course pricing is more expensive than Zhihu Live, including prices of 29/39/69/99/199/299 yuan. You can enjoy most of the content for free by becoming a Yanxuan member.

(2) A refreshing article that is easy to read and does not require much brainpower.

This type of content is mainly short and refreshing articles with compact plots, smooth writing, layered reversals, and a strong sense of routine, and has a clear entertainment and recreational nature. The target audience is mostly low-end users who demand information readability and entertainment. The pricing is basically in the range of 19.9 to 39.9 yuan, and you can enjoy the content for free by becoming a Yanxuan member.

There are two serious problems with the Yanxuan column:

(1) Confused positioning.

Serious courses and entertaining articles are very different in terms of content type and user groups. The overlap between the two types of user groups is very low, and there is even a chain of contempt. After integrating the confusingly positioned content into a unified "Yanxuan Column" module, both types of users will have positioning and cognitive confusion: "What exactly is this column used for?"

(2) Weakening the brand value of “Yanxuan”.

"Yanxuan" is a brand concept launched by Zhihu. Its core concept lies in "strictly selecting good books, strictly selecting good courses, and strictly selecting good content, so that knowledge can empower everyone." Its value lies in helping users screen the highest quality content. In terms of purpose, salt selection is the endorsement of official high-quality answers. For example, we will see answers from "Salt Selection Growth Plan", "Salt Selection Science" and "Salt Selection Life" under many questions. They are often long, logical, profound, and need to be paid. Therefore, there is a conflict between the core concepts and values ​​of online literature and "Salt Selection", which will seriously weaken the brand value behind the word "Salt Selection" and harm the willingness of some high-tech users.

5. User feedback and improvement suggestions

5.1 User Rating Comments (App Store)

5.1.1 Rating | The overall rating is acceptable, and there are many 1-star reviews

Comparing Zhihu with Himalaya, which is also a knowledge payment platform, and bilibili, which is also a video content platform, we can find that although Zhihu's overall score is 4.8 points , it is not excellent compared with similar products, and can only be considered as a normal person. Moreover, one-star bad reviews account for a large proportion.

5.1.2 Comments | User quality declines and community environment deteriorates

In his early years, Zhihu was known for its elite community atmosphere.

Nowadays, Zhihu is "out of the circle", and a large number of users with mixed traffic are coming: sinking users, young users... They all have their own opinions and are emotional and easy to be incited, resulting in the continuous deterioration of Zhihu's community environment.

For example, gender antagonism is already very serious on Zhihu. When Master Cicada browsed the latest comments, I found that among the nearly 60 one-star comments, at least 35 were related to " discrimination against women " and " hateful girls ", which contained a large number of words insulting men.

From the platform perspective, we need to think about how to improve the content supervision system . At present, Zhihu has established a creator salt value system, weighted and encouraged excellent and compliant creators, and downgraded creators who make inappropriate remarks. At the same time, two methods are adopted to fold, block, ban, and ban improper remarks and corresponding users.

5.1.3 Comments | About banned and banned: A review mechanism to be improved

Continuing with the above, Zhihu has adopted penalties such as folding, banning, and banning illegal remarks in order to supervise content compliance. However, it also raises many questions: What kind of content is illegal? Who will determine whether the violation is violated? How to make the punished feel fair?

A large number of users have expressed dissatisfaction with the content review mechanism on the App Store and Zhihu, believing that whether to ban the ban depends entirely on the emotions of the reviewer (little housekeeper), and there are many high-quality people among these users.

The author believes that the underlying logic of the audit mechanism should follow the "innocence assumption" in the law, that is, if there is no clear evidence, it is better to forgive the criminals than to catch the person. In terms of specific practices, first, the violation information and processes should be more transparent, such as disclosing the penalized information of other banned people (user ID, violation time, and reasons for violation) to those who have been banned, so that users can feel "procedural justice" and can accept punishment more; on the other hand, a cross-review mechanism can be adopted in manual review to reduce improper punishment due to personal factors of the auditor.

5.1.4 Comments | Content quality declines, Weibo and post bar

The decline in user quality and Zhihu's promotion of traffic-based content have led to a decline in the quality of Zhihu's content, which is no longer professional and profound as it was in the early years. Many users regard Zhihu as Weibo/title bar/hupu, reading novels, chasing gossip, or reading low-quality content that seems noble but is actually nothing.

Many insightful and logical answers are buried, while clever jokes and sensational and useless chicken soup are more popular.

Of course, it does not mean that Zhihu should insist on being small but beautiful, so it should not make money, but Zhihu can make money more elegantly.

For example, clarify and differentiate the positioning of free users and paid users . Provide free users with low information density knowledge, so that they feel comfortable learning, and supplemented by entertainment content, so that users feel that Zhihu is an interesting platform that combines education and entertainment. Provide paid users with high information density and quality assurance knowledge, so that users feel that Zhihu is still the pure land that allowed them to explore knowledge and wander in it.

5.2 Improvement suggestions

5.2.1 Browse Experience | Added "Last See" function

Many times, when we read hundreds/thousands of answers to a question on Zhihu, we just saw half of it, and we were interrupted and cut out the App. When I came back, I found that I had to go back to the homepage and go to "My"-"Recent Views" to find the half-read answer again. Moreover, if the answer is long, Zhihu will not position us where we read it, and we have to read it again.

What's worse is that at least there are traces to graphic content, while the video content cannot even be found. Zhihu is mostly medium-sized videos. Many users may cut out the app before they can finish watching the videos. When they come back, they cannot find the viewing record, and the experience is very poor.

Therefore, the author believes that the "Last See" floating window can be added to the Zhihu homepage, so that users can conveniently continue to browse articles or videos that have not been finished last time.

This approach not only improves the continuity and fluency of users' use of the app, but also helps them improve the efficiency of information acquisition. Compare the possibility of the following two situations: (1) User: I want to continue reading the articles I didn’t finish reading last time; (2) User: I read a good article a few days ago and forgot to collect it. I went to browse the browsing history. Obviously, the probability of the first situation is much greater than the second.

5.2.2 Video function | Added a new video type tag card in the navigation bar

The current video module navigation bar has only two option tags, "Recommended" and "Sea Salt Plan". There are two ways for users to watch videos: first, pull down the recommendation page and watch the videos recommended by the platform; second, find the videos of interest through keywords in the search bar.

There are two questions here: First, a large number of Zhihu users watch videos for purpose, and they have preferences for video types and hope to have more choices. Search is a time-consuming and laborious way to screen information; Second, if users are unwilling to search for videos, Zhihu can only try its best to make the recommendation algorithm more accurate and match, and reduce the cost of screening information for users. However, with Zhihu's current data volume and algorithm mechanism, most recommended video users do not like to watch it.

Therefore, it is very necessary to add a new video type tag card to the navigation bar. Since Zhihu's videos themselves are classified, such as "technology and digital", "workplace", "life", etc., you only need to add these types of tags to the navigation bar, and users can position the content they are interested in more quickly.

5.2.3 Member Module | Gather high-quality content and remove sinking content

As mentioned above, the author believes that Zhihu should distinguish between the services provided to free users and paid users: provide low-information density knowledge and entertainment content to free users, and provide high-quality and professional content to pay users.

Then, in the member module, most of the sinking content should be removed, and the online novel part and "novel" module of the "welfth club" module, "reading club" and "salt selection column" should be moved to the homepage, integrating it into two modules "welfth club" and "novel", and users can also allow users to customize the navigation bar label card of the homepage.

After removing the above module, the new member module is shown in the figure below:

6. Summary

In recent years, Zhihu has been doing addition for traffic and profitability, and more and more functional modules are being made. At first glance, traffic has increased and advertisements have been monetized better. But in the long run, Zhihu is actually overdrawing the capital that can be monetized in the future: the unique positioning of a "professional knowledge sharing community", an accurate user group and creator group, and a good reputation.

Video content is the general trend, and Zhihu has no need to resist and has to join. As for Zhihu's original positioning, many knowledge and exploration content knowledge are suitable for presentation in video format, such as book videos, and travel vlogs. Perhaps, Zhihu should explore the positioning of video content and be distinguished from other platforms instead of simply copying and pasting, so as to stimulate a healthy native video ecosystem.

In addition, even if Zhihu must develop towards pan-entertainment due to strategic needs, it should also be distinguished between entertainment users and knowledge users so that both types of users can clarify "What did I do when I come to Zhihu?" instead of being confused and dazed when I came to Zhihu and looking at the complex and cumbersome and contradictory functional modules.

Finally, in terms of user reputation, I hope Zhihu will improve its technical capabilities and ensure the availability of the basics; optimize the content review mechanism , do not punish users at will, and protect users' trust and favorable feelings for the platform.

Author: Product Usalata Rabbit

Source: Products Usalata Rabbit