NetEase Cloud Music product analysis!

This article is quite long. It comprehensively organizes and analyzes NetEase Cloud Music from the aspects of market and industry, product background and positioning, product functions, operations, business model and development direction. Based on the principle of "giving priority to serving mid- to high-end users, ensuring high-quality content and a good community atmosphere, thereby forming a natural attraction for mass users", it focuses on analyzing the excitatory functions, reverse functions and business model of Cloud Music and puts forward improvement suggestions.

It includes thoughts on how Cloud Music can deepen its recommendation function, how to socialize in an environment dominated by mid- to high-end users, its practices in segmented fields, how to maintain high-quality UGC, how to build a high-quality live broadcast platform, which customer segment to choose, how to expand revenue and reduce cost pressure.

Table of contents

1. Market and Industry Analysis

1.1 Market Analysis

1.2 Industry Analysis

1.3 Current Status of NetEase Cloud Music

1.4 Summary

2. Product Background and Positioning

3. Product Features

3.1 Product structure map

3.2 Functional analysis

IV. Operation

5. Business model analysis and improvement

5.1 Current Business Model Canvas

5.2 Analysis and Improvement

VI. Development Direction

7. Summary and reflection

1. Market and Industry Analysis

1.1 Market Analysis

1.1.1 Market size

The music industry has grown steadily in recent years.

Figure 1 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Compared with the mature American music market, it can be seen that there is still a lot of room for development in the Chinese music market.

Figure 2 iResearch: 2019 China Digital Music Industry Research Report

Figure 3 iResearch: 2019 China Digital Music Industry Research Report

Digital music accounts for about three-quarters of the music industry, followed by music performances.

Figure 4 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

iResearch predicts that the digital music market will continue to grow at a considerable rate in the future.

Figure 5 iResearch: 2019 China Digital Music Industry Research Report

1.1.2 User situation

With the advent of the mobile era, the scenarios in which users use applications have become more mobile and fragmented. Mobile music has a high usage frequency.

Figure 6 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

The scale of mobile music users has been growing steadily.

Figure 7 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

1.2 Industry Analysis

1.2.1 PEST environment

There is hope for copyright sharing; increased demand and technological development create more opportunities for mobile music.

Figure 8 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

1.2.2 External environment and development trends

(1) Upstream and downstream industry chain: The upstream is mainly music creation and music copyright, while the downstream is mainly copyright operation and distribution. There is still a lot to be discovered about more music gameplay and music derivatives.

Figure 9 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Figure 10 iResearch: 2019 China Digital Music Industry Research Report

Figure 11 iResearch: 2019 China Digital Music Industry Research Report

(2) Competitors

Trends of industry leaders in 2017: Trend towards forming cooperative alliances.

Figure 12 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Figure 13 iResearch: 2019 China Digital Music Industry Research Report

Tencent is actively building its own music ecosystem and establishing Tencent Music Entertainment Group.

Figure 14 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Tencent has the largest number of monthly active users, followed by Cloud Music. But Cloud Music’s users are more active.

Figure 15 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Figure 16 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Figure 17 2019.6.11 New Trends in QuestMobile Pan-Entertainment User Behavior

(3) Trends of potential competitors

Karaoke products are actively building an online and offline closed-loop ecosystem, and the future development trend is promising.

Figure 18 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Figure 19 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Figure 20 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

(4) Development trend

The platform connects the upstream and downstream of the industrial chain and builds a closed-loop ecosystem by combining online and offline.

Figure 21 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Cross-field cooperation will expand the platform's brand influence by exploring more music scenes and creating more music derivatives.

Figure 22 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

1.3 Current Status of NetEase Cloud Music

1.3.1 User scale

Judging from the data from December 2017 and the first quarter of 2019, the user scale of Cloud Music is second only to Tencent's music products, but it is far behind the first-tier QQ Music and Kugou Music.

Figure 23 Analysys: 2018 Annual Comprehensive Analysis of China’s Mobile Music Market

Figure 24 Analysys: In the first quarter of 2019, the user usage time of the mobile music market increased significantly, and music payment continued to be promoted

1.3.2 User situation

(1) Young users are the main body

Figure 25 iResearch APP Index data for July 2019

Judging from TGI, users aged 29 and below are the main user group of Cloud Music.

Figure 26 Baidu Index data for July 2019

(2) Mid- to high-end user groups pursuing high quality

Figure 27 ThinkingData Mobile Observatory data in November 2018

Figure 28 Baidu Index 19.8.12-19.9.10 data

(3) Evaluation of Cloud Music

According to the user reviews in APP Annie, they can be roughly divided into:

• Advantages: 1) Like the recommendation function; 2) Like UGC; 3) Good community atmosphere; 4) Beautiful interface.
• Disadvantages: 1) Lack of copyright; 2) Serious decline in the quality of reviews; 3) Dissatisfaction with membership and paid services; 4) Too many functions, questioning Cloud Music’s original intention for music; 5) Too many advertisements.

1.4 Summary

Compared with the mature American music market, the Chinese music market still has a lot of room for development. As part of the Chinese music market, the digital music market is currently in a state of steady progress, but there is still more room for development. The room for expanding the scale of revenue mainly lies in continuing to improve users' payment awareness and copyright operation and distribution.

Digital music platforms are based on the Internet. In the future, they will actively connect the upstream and downstream of the industrial chain, promote online and offline interactions, and form their own closed-loop ecology.

The user scale of NetEase Cloud Music is second only to Tencent's music products, but it is far behind the first-tier QQ Music and Kugou Music. Although the user base is not the largest, Cloud Music’s user activity often ranks first. Although Cloud Music is at a disadvantage in terms of copyright, its precise recommendation algorithm and high-quality UGC are still popular among users and have high user loyalty.

At the same time, we should also see that with the entry of a large number of mass users, the high-quality UGC of Cloud Music is being diluted. Some measures in the commercialization process and the continued lack of copyright have also caused a lot of dissatisfaction among users.

2. Product Background and Positioning

2.1 Company resources: other businesses provide financial support

NetEase’s revenue from games supports its investment in other sentimental products. Its tolerance for KPIs also enables Cloud Music to focus more on meeting user needs in product innovation and function design. However, apart from funds, Cloud Music has not integrated the company's other resources well.

2.2 Professional product team

Mr. Ding Lei said: The team members are all true music lovers. The product manager of Cloud Music once said, "When NetEase Cloud Music was cold-starting, the operation team invested a lot of energy in inviting senior experts in the industry, independent musicians, music critics, etc. to establish a professional tone and atmosphere for the product. Through the way of operators recruiting people, we also won over many music enthusiasts, who have contributed a lot of valuable content to NetEase Cloud Music."

The fact that Cloud Music was initially able to attract a large number of users with niche music tastes was inseparable from the team’s professional understanding of music.

2.3 The team’s product positioning for Cloud Music: helping users discover and share music

NetEase Cloud Music CEO Zhu Yiwen said that people have social needs when listening to music. Based on this user pain point, "helping users discover and share music" has become the focus of NetEase Cloud Music's product polishing.

It can be seen that the two major reputations of Cloud Music: recommendation function and UGC correspond exactly to the positioning of "discovery" and "sharing". Online music products before QQ Music did not pay much attention to user social functions and recommendation functions that actively help users discover other music. QQ Music's positioning successfully found unmet user needs and captured a considerable market share in the red ocean of online music.

2.4 Strategic Position

NetEase Cloud Music has accumulated a lot of user data, which will provide data support for NetEase's future development in other fields. The success of NetEase Cloud Music can also bring traffic to other NetEase products. At the same time, Cloud Music has also cultivated user loyalty and expanded NetEase's brand influence.

3. Product Features

3.1 Product structure map

3.2 Functional analysis

3.2.1 Basic functions

• It has basic functions such as account management, music player, music management, search, personalized dressing, rankings, etc., which can meet users' basic needs for listening to songs, searching songs, and managing music.
• In terms of content, in addition to online music, it also provides radio stations with other types of rich content, including cover songs, original songs, audiobooks, radio dramas, DJs, crosstalk and folk art, etc., as a supplement to the single music content. "Music Classroom" is another step in Cloud Music's exploration of online education.
• Music Cloud Disk: mainly cloud storage and automatic matching functions. Automatic matching can integrate local songs with the online resources of Cloud Music, providing a better viewing experience when listening to local music.

3.2.2 Expected Function

(1) Refine the search

It meets the user's demand for "searching under a certain module". Although it is a small scene, it has a high frequency of occurrence. This detailed processing improves the user's search experience.

(2) Identify songs by listening to them

Taking into account more song identification scenarios, two new functions have been added: identifying songs played in other apps and identifying songs by humming, to meet users' wider song identification needs.

(3) Whale Cloud Sound Effect

Providing users with simple music adjustment tools allows them to "re-create" music according to their preferences.

The simple and easy-to-use scene sound effect selection and equalizer adjustment lower the user's usage threshold. The modification of music satisfies the user's desire for creation and control, turning passivity into activeness, and improving the user's sense of self-realization. At the same time, the "motion effect" can bring users a dynamic visual experience.

Figure 29 Whale Cloud Sound Effect

3.2.3 Exciting features (recommendations and UGC)

3.2.3.1 Recommended features

The recommendation function is one of the word-of-mouth functions of Cloud Music. Online music products before Cloud Music did not pay much attention to the recommendation function. Songs were mainly searched by users, which resulted in users being able to listen to only songs they knew. The recommendation function of Cloud Music allows users to listen to more songs they don’t know. This differentiated feature was successful.

Why do users like the recommendation feature and are willing to recommend it to others? According to the functions of traditional online music products, users have very limited ways to discover new songs that suit their tastes. Cloud Music recommends songs that suit users' tastes based on their long-term listening data, which reduces the cost of finding songs for users . In particular, Cloud Music users are mainly middle- and high-end young groups who attach more importance to time costs.

It also meets users' personalized listening needs , allowing users to have the experience of "Cloud Music understands me", and this "understanding me" is actually an emotional need , and users can feel a sense of belonging from this experience. As users develop a deeper attachment to the product, it is only natural for them to promote it.

The recommendation functions of Cloud Music mainly include: daily recommendations, private FM, scenario-based recommendations (running FM, driving mode, parent-child channel, sati space), and professional recommendations (latest electronic music, jazz radio, and classical area). Although they are all recommended functions, they each have their own characteristics, target different users, and meet different scenario requirements, which can also help to identify some areas for improvement.

(1) Daily recommendations and music friends

Daily Recommendation is abbreviated as Daily Push. Its main function is to recommend songs based on the user’s recent listening preferences, and recommend a limited number of songs on time every day. This feature has been well received by users. The "Music Friends" function added to Ritui determines the similarity of music tastes based on the songs in common collections, and then recommends users accordingly.

Although the "Music Friend" function was created based on the user's demand for "wanting to communicate with users who have the same music taste", it does not make users particularly satisfied.

1) Why do users like daily push notifications? Accurate recommendations and "daily limited quantity" bring more expectations

The scenario in which users use daily recommendations is "click to see if there are any good songs recommended to me today's daily recommendations", and the demand is "to be able to listen to favorite songs", and based on the "limited" feature of daily recommendations, users are full of expectations in this scenario.

Thanks to Cloud Music's precise recommendation algorithm, users' needs in this scenario can be better met, giving users a sense of surprise. Today's satisfactory experience will make users more excited about tomorrow's daily recommendations.

2) Why is “Music Friend” not so effective? The recommended user is an abstract symbol, and it is difficult for users to chat

The demand raised by users, such as "wanting to communicate with users with the same music taste", does exist after using daily push, but simply recommending users cannot meet this demand well .

Think about the scenario of using "Music Friend" from the user's perspective. For the recommended users, will the recommended users really follow them or click on the chat window with them? Would you be willing to chat? What else should we talk about? What else can we talk about?

Or, would users be willing to click on the recommended user’s personal homepage and spend a certain amount of energy and time to find content of interest and get to know the user? Will users who receive private messages reply to the chat? Is the conversion rate from recommended users to friends significant?

Judging from the current user situation, users think that the algorithm for recommending users is not reasonable enough (because although "Music Friends" are displayed on the daily recommendation page, the "Music Taste Similarity" displayed is judged based on the two users' common collection of songs, which does not seem to have anything to do with the recommendations of the day, but encourages users to discuss the daily recommended songs), and they have no intention of chatting.

I think that for the recommending user and the recommended user, each other is just an empty abstract symbol, and there is not enough basis for chatting with each other. Unless users have already established a personal impression of each other, or the two can chat and discuss certain content (users are more willing to talk about content they are interested in), it is difficult for users to start chatting.

Therefore, we need to figure out what kind of social interaction users want in the scenario of "listening to daily recommendations" and how to make users willing to socialize .

Figure 30 Music Friend

3) Improvements to "Music Friend": Share your daily thoughts

If you want users to socialize through Ritui, you must first understand whether users have the need to "have further social activities" in scenarios such as "hearing a favorite song on Ritui". This does exist.

However, the social needs of users in the scenario of "hearing a favorite song" are the identification and thoughts about the song. The usual practice is to click on the comment area, read the wonderful comments in the comment area, like and reply to the comments they agree with, or even leave a comment based on their feelings, so the characteristics of daily push should be considered more.

The daily push notifications are updated at 6:00 a.m. every day, and they are limited edition. The feeling of users opening the daily push notifications is actually more like "opening a gift", and the gift is not considered "opened" before the user has finished listening to the daily push notifications. Therefore, I am more concerned about the scenarios after the user has listened to the songs in the daily push notifications rather than the scenarios during the process of listening to the daily push notifications , that is, the possible actions of users after the gift is opened .

Users will inevitably have some feelings after listening to the daily push, which may be surprise, disappointment, etc. Users have the need to share their thoughts on the daily push .

I think we can set up a discussion area for daily recommendations , where user A can post his or her own daily recommendations list and his or her favorite songs, express his or her thoughts on today's daily recommendations, and then recommend the content posted by user A to users BCD who also like these daily recommendations.

This feature not only allows users to express their feelings, but also explores a social scenario. This function can also be passively invoked (i.e., popped up actively) after user A has listened to the daily recommended songs, displaying content created by users BCD who also like these daily recommended songs, and encouraging the recommended user A to participate in discussions and sharing.

(2) Private FM

This function recommends songs based on the user's recent and long-term music preferences. It does not have a limited number of songs like the daily recommendation. Users can control the recommendation algorithm to a certain extent through operations such as "like, dislike, next song", etc.

1) Enhance user control: adjust recommended songs through operations

The good reputation of Private FM is also due to the precise recommendation algorithm. Unlike daily recommendations which only have a fixed number of songs per day and are based on your recent listening style, Private FM can recommend unlimited songs. The recommendations are not only based on your recent listening style, but also based on your long-term listening style.

Due to the unlimited nature, users feel more relaxed using Private FM, without the burden of "careful selection" like Japanese push.

When users use private FM, they usually listen in the background while doing other things. Depending on what they are doing, the style of songs they want to listen to is also different. Recommendations based on long-term tastes cannot satisfy them well, so users have the need to adjust the currently recommended songs .

Therefore, users have more control over Private FM than daily recommendations. Users can tell Cloud Music about their own tastes in music through these operations, and Private FM will also adjust subsequent recommendations based on these operations. These operations not only meet users' needs for scenario-based listening to music to a certain extent, but also satisfy users' desire for control.

There is a famous joke that goes, "Every time you click the red heart, it's like the mentor in The Voice of China turns around," which well reflects the user's need to control the recommended songs.

2) There is a lag in adjustments. How can we allow users to have more flexibility in controlling recommended songs?

According to my physical test, the recommendation algorithm is based on recent and previously accumulated music preferences. Although the songs recommended at the beginning are based on the style of the songs listened to recently, other types of songs will occasionally appear. These types are basically based on the types of songs that the user liked before. Once you click on the like button, songs of that type will continue to be recommended.

Especially if you change songs frequently, songs of the same style that you liked before will be frequently recommended, and such recommendations are very likely to be correct .

But I think that as users use private FM in more scenarios, this more robust recommendation algorithm will have a lag: Are there scenarios in which users have completely different ideas about listening to music than before? As mentioned above, users usually do other things while listening to private FM, so their tastes in music will be more diverse.

If the user's music taste today or at the moment is completely different from before, but the user has not accumulated song data of the taste he wants today, it will basically not be recommended.

This lag will result in a not-so-good user experience today, with the user feeling that “today’s private FM doesn’t really understand me.”

In this scenario, users will frequently switch songs. I think this scenario occurs quite frequently, and since private FM is controlled by users, it would be better to deepen the control and let users directly control and adjust;

For example, let users input keywords to adjust the recommended taste. When users frequently change songs, a prompt will pop up: "You are a little different today? (The prompt is very important. You cannot smear yourself by saying that your recommendation ability is not good . Instead, you should show that the user's taste today is special, which leads to inaccurate recommendations.) Try to enter keywords to get better recommendations."

Based on the keywords entered by the user, such as "high, rock, lyrical, Japanese", etc. (or you can choose to delete certain keywords), the user can independently adjust the recommended songs to obtain songs that are more in line with the user's current taste .

It can also be seen from this that as users have longer listening time on Cloud Music, there are more scenarios for users to use Cloud Music . Daily recommendations and private FM can no longer meet users' needs for more scenario-based listening. In addition to being able to adjust recommendations more flexibly, Cloud Music also needs to further explore users' different listening scenarios to meet users' more diverse listening needs. Apparently, Cloud Music has already started doing this.

(3) Scenario-based recommendations

Compared with private FM, scenario-based recommendations explore scenarios more deeply and provide more relevant recommendations for some user scenarios.

1) "Running" and "Driving" are typical scenes, "Parent-Child Channel" has rich content, and "Sati Space" is intended to help sleep and relieve stress of modern people

"Running FM" and "Driving Mode" are Cloud Music's detailed exploration of user scenarios. Running and driving are two typical scenarios, and the interface is designed to fit actual usage.

In addition to listening to music, the driving mode also adds a "driving radio", which is an extension of the scene. However, the content is not closely related to "driving". It is just random and ordinary radio content. It would be more perfect if it could filter out content such as "road congestion" or users could choose the radio content they want to listen to. It is recommended to add a classified navigation for radio stations.

The "Parent-Child Channel" is designed for parenting scenarios and has FM, radio and other functions. The content includes nursery rhymes, children's stories, English learning, Chinese culture enlightenment, etc.

Sati space has three modes: "sleep, focus, and decompression", corresponding to three user scenarios. The "Sleep" mode adds a white noise effect, which has a certain sleep-inducing effect and is closer to the scene than ordinary sleep playlists.

The "Decompression" mode has the classic "pop the bubble" function, which is intended to relieve user stress.

2) Use playlists to cover more scenarios

There are still more scenarios to be explored, such as doing homework, taking a bath, on a long journey, etc. In these scenarios, users will obviously have the need to listen to music.

If we were to set up a specific function page for each scenario like running and driving, it would be a huge project and not very necessary, because in fact many playlists created by users sometimes already serve as scenario-based recommendations , and users can even discover more scenarios that we can't think of by creating their own playlists. So can we use playlists more effectively?

For example, when the user selects a certain scene, the system searches for the playlist for the corresponding scene and plays it, or selects songs from several playlists of the same scene and plays them randomly.

What we need to do is to set up an entrance that users can think of when they want scenario-based recommendations . When the user enters this entrance and selects a scenario (the user can even enter the scenario himself), the system searches the playlist library for the playlist to play.

Figure 31 Song list coverage scene

3) Combining scenario-based and voice AI can tap into more scenarios and needs

The previous AI Xiaoice’s functions were not very strong, and there was no special experience brought by AI. When thinking about scenario-based recommendations, we can find that there are many scenarios where it is "inconvenient for users to operate the phone". These scenarios also have a high frequency of occurrence, such as running, taking a bath, doing housework, tidying up things, etc. The user's demand in these scenarios is to be able to operate cloud music without using both hands.

If we can allow users to communicate with AI in these scenarios and operate cloud music through voice like Siri, we can not only meet some simple operation needs of users, but also explore the scenarios more deeply , better meet the needs of users in these scenarios, and allow users to feel the intelligence of AI more truly .

I think we might as well start with some simple voice control commands, such as searching for a song to play, switching songs, clicking favorites, etc., to see the user's reactions and feedback, and then further iterate and optimize the AI ​​function.

If the technology can do it, AI can also make more personalized recommendations in interactions with users . For example, when asking Xiaoice to recommend songs, users can put forward their own ideas in the conversation with Xiaoice, such as wanting to listen to songs like xxx, wanting to listen to songs in a certain language, etc.

From this we can see that there are still many scenarios for the interaction between users and AI, and by digging deeper into these scenarios, we can discover more scenarios and needs .

The integration of music and scenes is a major development direction in the future, and more scenes and deeper integration need to be explored.

(4) Professional recommendation

Professional recommendations have set up three special areas for electronic music, jazz, and classical music, providing more professional content. This is Cloud Music's exploration of niche fields and provides better satisfaction for the listening needs of niche users.

In the electronic music and jazz modules, we cooperate with the highly professional DI.FM to provide a pure FM listening mode. Users can only choose a channel to listen to, switch songs, and collect channels. Compared with the usual listening page, private FM, etc., there are many fewer operations.

Figure 32 DI.FM

1) Classification of users for specialized recommendations: expert users and novice users

Since it is a niche field, a considerable number of users who use this function are expert users , and another part are novice users who are interested in this niche field (the "expert users" and "novice users" here are divided according to whether they understand the niche field or not).

When expert users hear some touching songs, do they have the need to leave their own thoughts and discuss with other expert users?

I think it exists. However, considering that songs are listened to in real time in FM mode , it is impossible to leave a message under a song like in traditional comment areas.

However, the channel always exists. If there are songs in the same niche under the same channel, you can leave a message under the channel and communicate through the message area of ​​the channel. This not only taps into a social scene, but also helps to gather expert users in the same niche and form a better discussion atmosphere.

However, I suggest that when users leave messages, they can attach the songs they are listening to (but it is not clear whether Cloud Music has the copyright to these songs) or display the song names, because users are more likely to generate ideas and then leave messages when they "hear a certain song". The songs are the carriers of the messages , which will also help people who read the messages to better understand the ideas of the users who left the messages and join in the discussion.

A feature of the message discussion area where expert users gather is that the content will be more professional, which will attract novice users who are interested in this niche field and become a knowledge acquisition area for novice users.

The message area can not only meet the needs of expert users for communication and discussion, but also meet the needs of novice users for acquiring knowledge. As for expert users and novice users, do expert users have scientific ideas when facing questions from novice users? (This is also a deep need that expert users may have to gain respect or even self-actualization )

If we think more deeply, once the atmosphere of the message area has been established, the message area will not be able to meet the needs of expert users to output knowledge and novice users to acquire knowledge. A broader platform is needed .

But in fact, Cloud Music already has such a platform, namely column articles and live broadcasts. Guidance can be provided in the message area, reminding users that they can express their ideas through these two channels. This will not only meet the further needs of expert and novice users, but also direct expert users to the column articles and live broadcast platforms, enriching and improving the content quality of the platforms.

Thinking more deeply, expert users also need to spend energy and time to popularize knowledge. In addition to spiritual satisfaction, if there are some real value returns, it can also motivate expert users to continue to output (but we must also consider that a considerable number of expert users pay more attention to spiritual satisfaction and are unwilling to let the content be tainted by money, so this function should allow users to choose whether to turn it on or not). That is, you can set a reward function for column articles, and live broadcasts also have a gift function, so are novice users willing to reward and give gifts after acquiring knowledge?

The main users of Cloud Music are middle- to high-end young groups who are more accepting of paying for knowledge. After they obtain high-quality content, they may have the need to pay for knowledge. Cloud Music also needs to educate and encourage users to pay for high-quality content, such as adding a reminder at the end of the article or during the live broadcast, adding a thank-you message after the payment is completed, etc.

At the same time, it should be noted that no matter which field, expert users are always a minority. How to better attract and gather expert users is the key , because only when expert users gather together can the above-mentioned "knowledge sharing platform" have room for development.

In addition to setting up a message area for expert users to communicate and discuss, are there any other approaches? Can users be activated through operations? For example, professional topics can be opened to encourage users to discuss, or operations staff can publish more professional columns to mobilize the enthusiasm of expert users to participate .

2) Improvements to the Classical Zone

The classical music section has richer content than jazz and electronic music. In addition to the classified music library and classical FM (the listening format is not a simple FM like DI.FM, but has more song information and operation functions like private FM), it also recommends classical works such as classical albums and playlists, as well as documentaries, performance videos, column articles, etc. You can also purchase classical records and classical musical instruments.

As analyzed above, the classical area also has expert users and novice users, and the module can be improved according to the above ideas.

However, I still don’t quite agree with setting up a module like a mall in the music application. From the user’s perspective, when users want to buy these products, why not go to an e-commerce platform that provides more professional services?

E-commerce platforms can provide better purchasing and after-sales services such as comparing prices, checking buyer reviews, and asking merchants questions to meet users' purchasing needs. Why choose to buy in a music app?

I think it would be better to provide more high-quality paid content instead of a mall. When users appreciate classical works, there is a demand for high-quality content, but the psychological burden of paying for online content is obviously smaller than paying for physical items ;

When paying for online content, users can enjoy the content they need faster and more directly; but when paying for offline entities, users need to consider more: Is the entity exquisite? How is the logistics? How to solve after-sales problems? etc. Moreover, paying for online content is more in line with users' positioning of online music applications .

3) Some problems found

There are some minor issues in usage: when listening to DI.FM, after exiting the corresponding interface and returning to the main interface, you cannot find the music playback bar at the bottom as usual, and the pull-down display still shows the paused songs on the main interface. You need to click in one by one to find which channel is playing.

This may be because there is a compatibility issue with DI.FM. Since users are likely to listen to DI.FM in the background, it is recommended to make improvements so that users can switch songs in the music playback bar and the drop-down notification bar at the bottom of the main interface as usual when listening to DI.FM.

3.2.3.2 UGC

(1) Comments section

1) The source of high-quality comment areas: a good closed loop

Comments are a successful attempt by Cloud Music to connect to music in a social way. High-quality reviews can further enrich the image of a song , making users feel more touched when listening to it.

In the "touching" scenario, users have the desire to express themselves and the need to express their feelings (and because the main users of Cloud Music are young users with more individuality, they have a greater desire to express, share and socialize). At the same time, users are also seeking a sense of identity and have the need to communicate with users who are listening to the same song, which will prompt users to leave comments.

If a comment can become a hot comment, the user's need to be respected (and possibly their vanity) can be better met, and in order to become a hot comment, users will be motivated to leave high-quality comments. This is a good closed loop .

2) Reasonable division of areas to make up for the shortcomings of traditional comment areas

Cloud Music divides the comment area into "wonderful comments, recent hot comments, and latest comments". Quoting a passage, it explains the setting of the comment area very clearly:

"The comment section of NetEase Cloud Music is actually composed of three small parts, from top to bottom: recent hot comments, wonderful comments, and latest comments. This arrangement is very clever.

Recent hot discussions: hot comments that were posted not too long ago. Generally, products have both the wonderful reviews and the latest reviews areas, but there are not many products with recent hot reviews. The comments displayed in recent hot comments are, on the one hand, newer than the popular comments and will not be less fresh due to the long time. On the other hand, compared with the latest comments, they are obtained after screening and can ensure excellent quality.

Excellent comments: Excellent comments are those with the most likes. These comments are generally relatively long, have stood the test of time, and are almost all of good quality. However, the excellent comments may not be novel enough because they are too old, so they can be used in conjunction with recent hot comments.

Latest comments: The latest comments are the comments section where the latecomers are at the top. This design allows everyone's comments to get enough exposure, allowing users to send comments they are satisfied with at any time without worrying that their carefully written comments will not be seen by anyone.

Source: How to create a high-quality comment section from NetEase Cloud Music https://www.niaogebiji.com/article-17845-1.html

The functional area division of the comment section of NetEase Cloud Music has effectively reduced the impact of comment sending time on comment exposure . It is a comment section with a sophisticated and reasonable design. ”

3) Areas that need improvement to maintain quality

The most criticized thing about the comment section at present is the decline in quality . In order to grab hot comments, low-quality comments such as making up stories, posting meaningless and irrelevant jokes, etc. have begun to appear in large numbers, disappointing native users.

This is actually the only way for a large number of mass users to flock in after the user scale expanded. Bilibili, Zhihu, etc. have also experienced this situation.

How to maintain high-quality UGC has always been a major issue for applications that target mid-to-high-end users after the expansion of user scale. We may try to find ideas to inspire cloud music from such applications.

①Retention rules are necessary

Like B station and Zhihu, in order to maintain a good community environment, Yun Music has also established a series of rules for the comment area, but it can still see many dissatisfactions of users:

When a product develops to such a large user scale, due to the diversity of users , there will naturally be an offended party when satisfying one party. This cannot be achieved at the same time. One party must be chosen to satisfy it first. This requires a trade-off based on the product's positioning and values . If it were me, I would choose to focus on the mid-to-high-end user market for Cloud Music, give priority to meeting mid-to-high-end users, and use high-quality content as the core attraction to create a natural attraction to the masses.

Therefore, although establishing rules will dampen users' enthusiasm and creativity to a certain extent, this is an effective way to ensure the quality of comments. I think the rules should be retained and the rules should be updated according to the comment situation.

② Improve the hot evaluation screening mechanism

Yun Music’s current hot reviews only look at the number of likes. This sorting indicator is slightly single, while Zhihu’s answer sorting algorithm (see the source article for details) is more rigorous:

"Zhihu answer sorting algorithm:

1. All users see the sorting is the same

2. Obtaining approval will increase the sort of answers, and getting opposition will decrease

3. Good answers in a certain field (differed according to the topics added to the question) will increase the user's voting weight in that field.

4. Voting of high-weight users in the field has a greater impact on sorting, and their answers are also sorted higher.

5. When using anonymous identity to vote or answer questions, the user's weight will not be calculated.

How does Zhihu calculate weights? The votes received by users in the past reflect the degree of recognition of other users' abilities in related fields. Voting by high-weight users will have a greater impact on sorting, and their answers will also have certain sorting advantages. ”

Source: How to improve the sorting of Zhihu’s answers? Introduction to the sorting rules and algorithms of Zhihu answers https://www.jianshu.com/p/c113b46247f2

Cloud Music’s comments may be referenced by Zhihu, a method that gives different weights based on multiple indicators . In addition to likes and reports, you can also increase the “step” button, refer to the user’s past comment behavior, etc. to set the weight for comment users . When sorting comments, it is not only referenced to the weight of the user who posted comments, but also referenced to the weight of the user who voted likes, reports, and strokes.

Working hard on the screening mechanism is easier to accept than directly deleting comments or banning users. Users still have the freedom to speak, but the exposure rate is different , and users will also work hard to create high-quality content in order to increase their weight.

4) The comment area has decreased attraction, and new high-quality UGC needs to be explored

Now, in addition to the decline in the quality of the comment area, users are becoming more and more familiar with the "routines" of comments . The comment area is not as attractive to users as before, and such an environment has also led to users not being as willing to share their feelings in the comment area as before.

In addition to taking measures to maintain the quality of the comment area, Cloud Music needs to explore new high-quality UGC to better attract users and promote it, and it remains to be seen whether the recently launched mlog function can effectively make up for the appeal of the comment area.

(2) Song list

This feature turns traditional users’ local playlists into playlists that can be shared online. The playlist also reflects Yun Music's consistent pursuit of high quality: high-definition and beautiful cover, careful title introduction, and rich and diverse playlist content.

The playlist is a "sharing music" education for users by cloud music, aiming to encourage users to share their listening list with others. In the early days, in order to educate users, Yun Music relied on its operators to create many high-quality playlists to guide and encourage users to create high-quality playlists, which successfully cultivated users' awareness of creating high-quality playlists.

Now when creating a playlist, considering that users have more diverse needs than sharing, it is more user-friendly and personalized , and there are options to set as a private playlist.

1) High-quality song lists satisfy users and encourage users to create high-quality song lists

The playlist is both UGC and a supplement to the system's recommended functions . When users have clear goals for listening to songs , they will often actively search for the song list of the corresponding songs to get recommendations for the corresponding songs. Compared with system recommendations, users can find songs that meet their current listening needs faster.

When users enjoy the good experience brought by high-quality playlists and see that high-quality playlists can be loved by many people, they will be more motivated to create high-quality playlists with encouragement from these scenes. Users will hope that their playlists will be loved and affirmed by more people, which is a need to be respected and self-realized .

2) Can it be improved? Recommended unpopular high-quality playlists by users: not only encourages playlist creators, but also meets some unpopular and high-quality enthusiasts

Nowadays, the playlist is extremely rich, and it is becoming increasingly difficult to create another high-profile playlist. If users work hard to create high-quality playlists but are unable to get good feedback for a long time, their self-realization needs will not be effectively met, and their motivation will decline .

When searching for a playlist, the popular playlist will be displayed first, and users often only listen to the popular playlist, which will cause the Matthew effect .

Can you find promotional functions for these unpopular and high-quality playlists? For example, there can be a "unpopular high-quality playlist recommendation area" on the song list interface, and users can recommend the song list that they consider to be high-quality.

This function mainly faces three types of users:

1. Users who create playlists;
2. Recommended playlist users;
3. Looking for users with unpopular and high-quality playlists.

The first and second categories of users may have a large proportion of overlap, that is, users can promote their playlists.

If you are different users, the second type of users will share and communicate after enjoying high-quality but unpopular playlists, seek a sense of identity, and then be willing to spread the demand for high-quality content. At the same time, users who are looking for unpopular and high-quality playlists can just find the playlists they need through these recommendations, and they may also join the second type of users.

When users who create playlists find that someone is willing to recommend their own playlists, even if the playlist is not popular, they can feel the satisfaction of their respected and self-realization needs and have the motivation to continue to create playlists.

This is more open compared to the current recommendation method of sharing playlists to dynamics, because there are not as many people in the dynamics as there are in the square and there is not much spread. Secondly, users who are looking for unpopular and high-quality playlists can find playlists with higher costs and lower rewards by following users' dynamic sharing of playlists.

More open recommendations are also conducive to the flow of playlists and will not be concentrated on popular playlists. At the same time, on the search playlist results page, you can also use three areas like the comment area to better display high-quality content.

Currently, Yun Music has launched a new AI personalized customized dynamic playlist, which is better satisfying for users who listen to playlists, but it is a more severe challenge for users who create playlists .

Since there is a more personalized AI customized playlist, is it unnecessary for users to build their own playlists?

I think that no matter how powerful AI personalized and customized dynamic playlists are, they cannot completely replace the user's own playlists, because users will always have different needs.

But it seems that AI customized playlists will cover more content and user-built playlists will be more difficult to become popular . As for how users’ self-built playlists can better exist by then, it remains to be seen and explored.

(3) mlog

mlog can publish pictures + (voice +) music or short video + music. The effect after release can be previewed during creation. The creative process is convenient and easy to operate, lowering the user's creative threshold.

mlog is quite high in terms of user participation popularity. It includes Amway stars, Amway Film and Television, expressing one's own feelings, two-dimensional correlation, recording one's own life, etc., which are very rich.

1) mlog as a supplement to the recommendation function, songs and comment sections

There are two main scenarios when users watch mlog:

First, "Look at it when you are bored". At this time, mlog can supplement the recommendation function. The diverse mlog content plus the multi-dimensional display method of "picture/video + text" can better meet users' needs to solve boredom and discover songs.

The second is to click on the mlog created by using this song from the play page of a song. At this time, the function of mlog is similar to the function of the comment area, enriching the image of the song and allowing users to have more feelings about the song. Similar to leaving a comment, in such a scenario, users have a higher probability of using this song to create an mlog.

From this we can also see that if the entrance of mlog is suspended in the comment area, it would be a better choice . mlog and the comment area together as a supplement to the song, and it is also a supplement to the comment area to meet users' needs for recognition and socialization.

2) From mlog to "dynamic": Deepening of social relationships

Yuncun has two main modules: mlog and "dynamic". mlog occupies the field of vision with high-definition large images. Compared with "dynamic", it weakens "people" and emphasizes "content" .

When following the user, the user's dynamics appear through the "dynamic" module. It can be inferred that the two are a progressive relationship. mlog helps users discover more bloggers that can be followed through exciting content, and then maintains them through "dynamics" after forming a social relationship.

The addition of mlog also helps to increase the activity of "dynamics", from "favorite content" to "focusing on creators" is a deepening relationship, which is conducive to improving user stickiness (user mentality ranges from "see what interesting content is there" to "see also what bloggers I like posted today".).

So how to deepen the relationship and make users willing to pay attention?

In addition to relying on users to choose and follow, I think we can also recommend other contents of creators of content that users have loved. If the user likes it many times, it means that the creator's creative style is more in line with the user's taste, you can actively remind the user, "You like TA xx content, do you follow TA?".

Figure 33mlog and dynamic

3) Improve functions through two types of mlog users

From the above analysis, we can see that the positioning of mlog is more clear, that is, it assumes the function of "discovery". Users can also be roughly divided into two types: content creators and viewers .

Content creators need to output and continuously output more exciting content, while viewers want to see more exciting content.

From the perspective of the creator, what needs to be satisfied is the need to be recognized, respected and shared communication, which requires more participation and interaction among viewers; from the perspective of the viewer, what needs to be satisfied is curiosity, and they are eager to discover more fresh and interesting things.

To better meet the needs of both, it is essentially to improve the matching between content and people . If the content that the viewer likes can be pushed to the viewer more accurately, the viewer will naturally be more willing to participate in the exchange and discussion, which can just meet the needs of the creators, so that a good closed loop can be formed.

One of the keys to improving the matching degree is the recommendation algorithm, which requires the recommendation algorithm to more accurately identify user preferences and push content.

In order to better improve the accuracy of the recommendation algorithm, I think it can also allow users to make reasons for content they like and hate, and adjust the recommendation algorithm more flexibly by increasing users' independent regulation .

However, I think in addition to pushing content that users like, we can also appropriately push content from some areas that users have not been exposed to to help users discover more areas that they may like.

In addition to recommendations, it is recommended to add search functions and classify content from the viewer's perspective. Because mlog is only recommended by recommendation algorithms, it is suitable for users to wander around aimlessly in the early stage of the function, but it cannot better meet the needs of users in scenarios where they clearly want to see the content type and have clear likes and dislikes. This demand will be more rigid when there are more mlog content and more obvious user tastes in the future.

From the creator's perspective, there is also a little improvement: when I first tried to create an mlog, when I finished writing the content and was about to add a topic, due to the habit of posting Weibo, I naturally entered "#" and waited for the topic to pop up automatically, but there was no option to add a location. After thinking about the topic for a while, I clicked "Complete" and found that there were options for "participating in the topic" and "My Location" after this, and I couldn't help but feel sad.

It is recommended that you follow the user's old habits here, and you will have the options of "participate in topics" and "My Location" before clicking "Finish", and change "Finish" to "Publish" to reduce the steps for users to publish content .

And, the concert stops when viewing the comments, this suggestion fixes.

3.2.4 Reverse function

3.2.4.1 Live broadcast

Live streaming can be found in two forms: video live streaming and language live streaming. Video live streaming is the same as traditional live streaming, and mostly relies on high-value anchors to attract users.

Language live broadcast is more implicit, with only a picture in the background and the anchor's voice. Although there are many categories of live broadcasts, the content is similar, mostly singing and chatting.

Judging from the requirements of Cloud Music to collect anchors, we also want to take the route of high-quality live broadcasts and also want to give musicians a space through live broadcasts. However, it seems that the quality of live broadcast content is not significantly higher than that of other live broadcast platforms, so the current live broadcast audience is mostly popular users rather than mid-to-high-end users expected by Cloud Music.

Figure 34 Live

(1) How to make mid-to-high-end users accept live broadcasts: Establish a new positioning for live broadcasts

I think the key to allowing mid-to-high-end users to accept live broadcasts is to change the long-term vulgar impression of live broadcasts by mid-to-high-end users. At least, users must have the positioning of "live broadcasts in cloud music are different from live broadcasts on other platforms."

If possible, don’t let ordinary users start broadcasting first, but try to start with live performing arts broadcasts , live music festivals, concerts and other high-quality content, and you can even join the "Music Classroom" in the classical area (of course, paid content should be paid first and then entered the live broadcast room), attract users to pay attention to the live broadcast function and re-establish the live broadcast image , so as to avoid most users avoiding it when they see the live broadcast.

Afterwards, try to introduce high-quality anchors , such as introducing musicians to create live music and sing, giving musicians more opportunities to show off. This is also one of the intentions of live broadcasts of cloud music; you can also introduce expert users (if the atmosphere of communication and discussion among expert users has been established in the "Professional Recommendation", you can try to divert these expert users to the live broadcast module, and this will also attract novice users who are interested in a certain sub-field to come to the live broadcast module) to popularize music content.

These high-quality anchors are also an education for future potential anchors and audiences . After users have a new positioning for "live broadcast in cloud music", they will slowly add user-created content to encourage users to cover, play, create, etc. With the previous education, I believe that users who start broadcasting later will strive to do high-quality live broadcasts, and viewers can also screen out high-quality live broadcasts to buy into high-quality content .

At the same time, the audience should be properly given regulatory power , such as investigating problems during the live broadcast, allowing the audience to write comments for the live broadcast and feedback to the backend, etc., to better manage the content of the live broadcast platform.

(2) Improve the live broadcast format

At present, the live broadcast forms in Yun Music mostly rely on their appearance to attract the audience's attention, but since you want to build a high-quality live broadcast platform, you should focus more on the live broadcast content;

Considering that mid-to-high-end anchors may have a psychological burden for showing their faces , we may try to add a "cartoon character blocking face" live broadcast mode to the live broadcast function for users to choose, reducing the psychological burden of users when starting broadcasts, and allowing viewers to pay more attention to the content of the live broadcast rather than the appearance of the anchor.

(3) Select high-quality musicians through live broadcast

Further measures will be taken for anchors with high-quality content, such as contracted musicians, columnists, etc., to give back to the upstream.

3.2.4.2 Short video

Yun Music has specially set up a partition for short videos. In addition to traditional MVs and live scenes, it also adds categories such as dance, ACG music, games, covers, etc.

In an interview with the vice president of Cloud Music, it was mentioned that "enhancing the short video function can increase the frequency of users using cloud music in other time periods, and it is also a way to increase daily active users."

It is true that this will improve daily active users to a certain extent, but I think it will also damage the user experience and users’ impression and positioning of cloud music to a certain extent .

From the user's perspective, why don't users watch these videos on B station, Youku and other video websites, but watch them in cloud music? Is the content of these video websites not as rich as cloud music? Or is the user atmosphere not good enough? Can the user atmosphere of Cloud Music be better than the barrage of B station?

I think that under the existing video content and user atmosphere of Cloud Music, when users have the need to watch videos, the first choice in their minds will be these video websites rather than Cloud Music.

We should think about "In what scenario will the user choose to watch videos in cloud music?" I think it was after listening to the song and found that the songs had related videos. In such a scenario, the user would choose to watch videos in cloud music.

Starting from this scene and demand, videos are derivatives of songs in cloud music, and their function is more to supplement songs. The recommendation of videos should be based on song-related, that is, videos related to songs are recommended in song-related recommendations. "Using videos as derivatives of songs" is also more in line with users' positioning of cloud music as a music application, and everything should be centered on music.

3.2.4.3 Make friends in Le

The overly straightforward social function does not match the positioning of cloud music. In cloud music, content-based social interaction is obviously more appropriate.

IV. Operation

4.1 Wonderful and interesting PGC: Communicate with users

In addition to encouraging users to create rich UGC, the early NetEase Cloud Music also invested in many PGCs, including the exclusive self-made columns "Post Music Live", "Super Face to Face", "NetEase Celebrity Meeting Room", the exclusive self-made music program "Ala Ding" and the official radio program, as well as the official editor's columns "Weekly Hot Reviews" and "Explore My Dooms" as well as "topics" that encourage users to participate in discussions.

These operations have created a pleasant atmosphere for Yun Music to communicate with each other, and have made important contributions to improving user stickiness and loyalty. (But most of these contents have disappeared now)

4.2 Promoting hot reviews, brand videos, cross-field cooperation: enhancing brand influence and building a fan circle

Wonderful comments were once a major driving force for cloud music users to spread independently. Cloud Music also attaches great importance to hot reviews. In many places and various operating activities of the product, it has been noted that hot reviews are used for promotion : Yuncun hot review wall, music signature, comment sharing functions, and promotion of hot review blogs on the official Weibo account, offline music review trains, music review aircraft, etc.

In addition, there are brand videos, cross-field cooperation, including opening a "Sleeping Music" theme hotel with Yaduo Light Home, the "Qingdao" music theme coffee shop jointly launched with Luckin Coffee, and cross-border linkage with Incisives Fengyin.

It is not difficult to see that these operational activities focus on leveraging the advantages of cloud music outputting high-quality content, "moving with emotion", allowing users to find a sense of belonging, cultivating user loyalty, and expanding cloud music's brand influence.

V. Business model analysis and improvement

5.1 Current business model canvas

5.2 Analysis and improvement

5.2.1 Customer segmentation: Priority for serving mid-to-high-end users

Due to the gathering of early mid-to-high-end users, Cloud Music has formed a unique high-quality UGC and a harmonious community atmosphere different from traditional music applications. High-quality UGC and accurate recommendation algorithms have attracted a large number of mass users in the later stage.

Like early applications such as B station and Zhihu that attracted public users with high-quality content, Cloud Music also faced a choice between two obviously contradictory user groups.

I believe that priorities should be given to serving mid-to-high-end users, ensuring high-quality content and a good community atmosphere, so as to form a natural attraction to the masses. There are two reasons why I choose mid-to-high-end users:

(1) There are currently no other music platforms that gather so many mid-to-high-end users like Cloud Music. It can be said that Cloud Music is the leader in this segment in the industry, while the entertainment of mass users has many products to satisfy, and they should not give up their unique advantages.

Moreover, most applications originally planned to be located in the commercialization process of mid-to-high-end users choose mass users, while mid-to-high-end users are always "exited". It is difficult for these users to have a pure land, with higher loyalty and stronger awareness of the community;

(2) Many applications that choose to "embrace the masses" attribute the problem to the choice between "high-quality content and commercialization", but is it really incompatible between "high-quality content and commercialization"?

I think "commercialization = embracing the masses" is not true. Mid- and high-end users can also create commercial value. Mid- and high-end users have the desire to pay for high-quality content as much as the public users' desire to pay for mass entertainment content (and mass users also have the desire to pay for high-quality content). However, most applications have not yet figured out a clear path, so Cloud Music might as well continue this exploration.

Currently, cloud music can support some portals to close. This function can alleviate the conflict between mid-to-high-end users and mass users to a certain extent. However, the places you will see in typical user scenarios such as homepage recommendations, song playback pages, and comment sections have live portals, which will continue to impact users' positioning of cloud music in their minds (unless they can successfully create a high-quality live broadcast platform).

At present, these users have no better place to go. If applications that focus more on mid-to-high-end users appear later, it is very likely that they will seize these users from Cloud Music, and they should be kept so vigilant.

So I think it would be a better choice to prioritize serving mid-to-high-end users. If these users continue to lose, then the core attractiveness of cloud music to mass users will also decrease.

5.2.2 Value Proposition: Provide high-quality content

Since it is decided to prioritize serving mid-to-high-end users, the content should also be provided with high-quality content.

How to maintain an environment that creates high-quality content, how to introduce high-quality content into other modules such as live broadcasts, etc., the above analysis of comment sections, professional recommendations, and live broadcast functions.

5.2.3 Source of income: Find more paid points

5.2.3.1 Develop different payment plans based on user segmentation to improve the differentiation of paid services

User payment is not only one of the important sources of income, but according to QuestMobile survey results, online music APP paid users have stronger stickiness than ordinary users.

Figure 35QuestMobile Pan Entertainment Paid User Insight Report

Therefore, it is of great significance to increase the number of paid users. The key to promoting user payment is to find places where users are willing to pay, and you can design payment plans by analyzing user segmentation.

1) Young fan group: learn from the Korean idol economy operation strategy and make full use of the fan economy

With the development of the domestic idol industry and the influence of Korean pop idol culture, young fans have a high willingness to pay for idol peripheral products. In addition to the payment plans for digital music and digital albums, you can also refer to the operation strategies of Korean idol economy to mobilize fans' enthusiasm for payment.

For example, establishing an authoritative list to stimulate fans to make a list and pay; selling idol peripherals (currently, Yun Music has members unlocking idol highlights, postcards and other songs, and selling performance tickets. If you can perform live performances through live broadcast, you can also try to meet the needs of fans who cannot get to the scene through online ticketing ).

Figure 36 iResearch Consulting: 2019 Research Report on the Development of China Digital Music Content Payment in 2019

Considering that most of the young fan groups are students, there is no independent source of income, but there is more disposable free time , we can learn from Spotify's method of setting up a student paid account: students' paid account can enjoy discounts (although it reduces the value of a single paid user, it helps to increase the scale of paid users).

And try to launch a model similar to "doing tasks for gold coins", allowing users to exchange benefits by completing tasks, such as obtaining a certain level of rankings, a certain number of gold coins can be exchanged for digital music, etc., so that users who are more sensitive to money can obtain the same effect through non-monetary methods .

2) Users with niche music flavors: listen first and then reward, encourage communication

For works of popular singers with high popularity, the payment first and then listening mode can be adopted (that is, selling digital music); for works of niche musicians, the form of listening first and then reward (the platform and musicians share) should be adopted, and users should be encouraged to pay for high-quality music .

Because mass singers have a fixed fan base, while niche musicians have few fixed fans and lack widespread dissemination of their works, if they pay first and then listen, not only will they not be able to obtain too much income, but may even hinder further dissemination.

In addition, for niche music, the method of " dissemination and reward " can also be adopted: if the user shares the song through the sharing code, when other users listen to the songs with the sharing code, a small amount of money will be rewarded to the musician.

Differentiated services can be further reflected here, such as the share code of member users will be rewarded with double the amount, etc.

3) Explore more segmented services and freely combine packages

By exploring more user segments, more personalized paid services can be proposed, but if all paid services are concentrated on members, the membership fee will be higher, and for segmented users who only need a small number of services, purchasing a membership is not that cost-effective.

If users can freely choose various paid service combinations (such as advertising-free + music packages, music packages + song peripherals, etc., of course, this will put higher requirements on the determination of paid service prices), it will effectively attract potential paid users to pay .

5.2.3.2 Encourage users to pay for high-quality content

As mentioned above, if Cloud Music can successfully introduce high-quality content into live broadcasts, column articles and other platforms, users can be encouraged to pay for high-quality content and obtain a certain amount of income through live broadcast gifts and article rewards.

5.2.4 Cost structure: Go deep into the upstream and downstream, increase copyright operation and distribution, and cultivate musicians

Copyright fees are one of the main costs of Cloud Music. If copyright operation and distribution can be increased, the burden of copyright will be reduced to a certain extent. If we can go deep into the upstream and cultivate more musicians, it will also reduce the pressure to fight for copyright. Yun Music currently has its own plan to train musicians.

How to increase copyright operation and distribution? As an online music platform, Cloud Music also has K songs and live broadcast platforms downstream . At present, Yun Music has made some arrangements in the live broadcast field, and you can also try to layout the K-song platform in the future.

However, there are karaoke and live broadcast products in the industry, which already occupy a position in the hearts of mass users. If Cloud Music intends to join the market, I think we should find our own positioning and make our own differences , such as the "high-quality live broadcast platform" analyzed above. Otherwise, why should users give up their original platform?

In my opinion, since NetEase Cloud Music has become synonymous with "high-end" in the minds of users, it might as well continue this impression of users and continue to create a "high-end live broadcast and high-end karaoke" platform.

And I think there are not only live broadcasts and karaokes downstream, but also more music derivatives . Cloud Music can provide users with copyright, encourage users to create secondary creations, and produce other high-quality music derivatives.

Moreover, through the operation of downstream products, users can select high-quality musicians to give back to the upstream.

5.2.5 Customer Relationship: Maintain interaction with users

Early cloud music had many communication and interactions with users through operational activities, which not only created a good community atmosphere, but also improved user loyalty. Now these interactions are much less.

It is recommended to strengthen interaction with users from the operational aspect, and at the same time, expand brand influence and improve user loyalty through offline activities and cross-field cooperation.

6. Development direction

The future development directions of cloud music are roughly:

6.1 Try to obtain copyright and make up for copyright shortcomings

As an online music application, there is no doubt about the struggle for copyright. With the improvement of national policies, I believe that the copyright issue will be better solved in the future.

6.2 Deepen recommendation functions and continue to create a high-quality UGC community environment and build your own moat

Recommended functions and high-quality UGC are the reputation of Yun Music. No other competitor can match Yun Music in these two aspects. Yun Music obviously realizes this and pays great attention to iterative updates of related functions in these two aspects in the update of the version.

6.3 Explore the community model where mid-to-high-end users and mass users coexist

Although Yun Music has not clearly stated whether it intends to "embrace mass users", from the update of Yun Music, Yun Music has not clearly chosen one of these two users. It is speculated that Yun Music also has a "embrace" attitude towards mass users, and it may explore a community model that meets both mid- and high-end users and mass users in the future. Currently, the "Collection Entrance" function may be an attempt to satisfy both users at the same time.

6.4 Go deep into the upstream and downstream of the industrial chain and create your own music ecosystem

We will go deep into the upstream to train musicians and increase copyright operation and distribution in the downstream. Major online music platforms are actively building their own music ecosystems. Yun Music currently has a musician plan and has entered the live broadcast. I believe that it will continue to develop in these aspects in the future.

6.5 Explore a profit model that can give full play to its advantages and improve the business model

In addition to traditional advertising revenue and value-added services, Cloud Music's current profit methods have also tried ticketing services, live broadcast sharing, mall revenue, and cooperation traffic packages with operators, but none of these seem to make good use of the core advantages of Cloud Music's mid-to-high-end users.

Is the way Cloud Music can play its core advantages in the future? Is it to guide mid-to-high-end users to produce high-quality content and then try to monetize high-quality content? (But monetization of high-quality content is a long-troubled issue in the industry, and there is no good method yet).

A more perfect business model is an effective guarantee for the sustainable development of products. Yun Music has not yet had a clear answer to some of the problems existing in the business model. Gradually improving the business model will be one of the future development directions of Yun Music.

7. Summary and reflection

This article has comprehensively sorted, analyzed and thought about improvement ideas for NetEase Cloud Music from the external environment and the product itself.

In today's world where product homogeneity is serious , deepening your own advantage segmentation may be a solution to combat product homogeneity. This article also focuses on analyzing and thinking about improving the existing functions and business models of Cloud Music on the basis of choosing mid-to-high-end users.

Since the author still has many shortcomings in theoretical knowledge and practical experience, the collection and interpretation of facts and data in this article may not be accurate enough and not close to the actual situation; the analysis in the article may still be shallow, not in-depth and essential enough; the proposed improvement plans may also have problems such as being too idealized, so I hope I will give you advice.

Author: yoonri

Source: yoonri