AARRR traffic funnel model｜Himalaya FM analysis report!

This article uses the AARRR traffic funnel model to conduct a detailed analysis of Himalaya FM's user acquisition, retention , revenue generation, self-propagation products, and a series of operational actions, and ultimately provides suggestions for further optimization.

Background and research purpose:

With the rise of knowledge payment , audio payment has become another hot form of recharging and relaxing in fragmented time. In 2017, Himalaya achieved sales of 196 million yuan on the same day under the wave of knowledge payment. In 2018, it is at a critical stage of product business development, and user demand is growing and becoming more diversified.

The purpose of this survey is to improve the user growth of the product as the main line, and use the adjusted AARRR model to sort out the product and operational actions of Himalaya APP user acquisition, user retention , revenue generation and self-propagation, point out the weak links and provide optimization suggestions.

Research ideas:

1. Product functional architecture and industry architecture;
2. Sorting out the timeline of key product operation periods;
3. Combined with A AR RR model for sorting;
4. The process can be optimized.

1. Product Functional Architecture and Industry Architecture

1. Product Description

Slogan: Listen to what I want to listen to, anytime, anywhere.

As a professional audio sharing platform, Himalaya FM brings together hundreds of millions of audio contents including novels, stories, courses, interviews, etc. It is considered to be one of the best FMs for recharging and relaxing in fragmented time.

Based on the following (in the past year) Baidu Index keyword searches, the target users are preliminarily determined to be office workers, car users, students, housewives, etc. It mainly adopts PGC and UGC operation models, and is committed to opening up the Internet and Internet of Things audio markets. It is jokingly called the paid version of Taobao for knowledge audio.

1. Product logic and functional architecture

1. Industry architecture and product life cycle

From the time when the Himalaya FM version was officially launched on the market to the present, the Android data collected by Kuchuan (20141101-20180820) shows that the product is currently in the growth stage of its product life cycle, which can be divided into three stages:

1. Introduction to the Exploration Phase (2014/11/01-2015/10/16): During this phase, the product completed the basic functions of Internet FM and considered in-vehicle functions, laying the basic foundation for the stable growth phase.
2. Stable growth stage (2015/10/16-2017/01/07): During this period, the product underwent multiple version iterations, the audio function was improved, and the awareness of Weibo and WeChat rewards and knowledge payment was gradually formed.
3. High-speed growth stage (2017/01/07-2018/08/22): According to the industry report of iResearch Consulting , this stage coincided with the structural transformation of China's cultural and creative industry and the 2017 knowledge payment boom. According to official data from Himalaya Research Institute in 2018, Himalaya currently holds a 73% industry share and has 470 million activated users, making it China's first paid audio knowledge unicorn and operating as a paid knowledge platform.

Summary: From the growth of the three stages of the product, we can see that while Himalaya is mainly operating with the "PGC and UGC" content model, it is constantly deepening its platform thinking and expanding its business into the hardware IoT market.

2. Product key operations

1. Key operational events and version review

Combining the data and information from Himalaya Research Institute, IT Orange, Kuchuan iteration records, and Himalaya official Weibo, we have compiled the following key operational events and version iteration highlights from December 12, 2014 to August 17, 2018:

1. Product milestones summary:

According to 1.1 and 1.2, it can be seen that Himalaya’s operational objectives and means for different product stages are also different.

1. In the product exploration stage, the main operation purpose is to attract new users and promote products . From the operation events, we can see that the product mainly focuses on offline scenarios and hardware market in this stage, such as brand car launches/exhibitions, Nut pro mobile phone launches, hot topic celebrity concerts, Didi ’s designated driver free Walkman, and as of October 2015, it has up to two million hardware users. From the iteration key points record, we can see that the key points of product updates in this stage are mainly concentrated on the optimization of basic functions such as registration, login, playback, and discovery page, and preliminary attempts are made to add rewards, social circles and commercial needs, sowing the seeds of commercial realization while consolidating the PGC foundation for the next steady growth stage.
2. During the stable growth stage of the product, the main operational purpose is to continuously attract new users and promote activation. From operational events, it can be seen that the product at this stage mainly focuses on online scene promotion and the copyright of high-quality audio content from big names. For example, it owns the audio copyrights of 70% of best-selling books, and works with a unique team to create the exclusive online broadcast of the boutique course "Speak Well", and the 123 Knowledge Carnival . From the record of key points of iteration, it can be seen that the key points of product updates in this stage are mainly concentrated on optimizing payment, optimizing personal homepages, building a membership system , expanding social circles such as listeners and discoveries, in preparation for the next step of rapid growth in user retention and monetization.
3. During the rapid growth stage of the product, the main operational objectives are activation, retention and monetization. From operational events, it can be found that the product at this stage mainly uses cross-border scenarios, membership system platformization, and UGC as entry points while maintaining high-quality content. For example, it launched listening cards, public welfare IP, talking mooncakes, and the Ten Thousand People and Ten Billion New Voices Incubation Plan with China Telecom . At the same time, from the analysis of milestone events, it is noted that after the rise of the 123 Knowledge Carnival, Himalaya has consciously cultivated users' awareness of understanding, consuming and using knowledge on the 423 Listening Day. From the iterative recording classes, it can be seen that iterations are extremely frequent in the stage of rapid growth, mainly focusing on fine operations and creation functions such as live broadcast , anchor workstation, personal backstage, content classification and recommendation, and planning for future personalized UGC, platformization and user systematization.

3. AARRR Model Review

Before conducting an AARRR model analysis, it is important to first differentiate between product, operations, and marketing actions.

According to the definition in the book "The Light of Operations", operations are responsible for creating short-term user value and assisting products in improving long-term value, mainly including user conversion , usage and payment. Products are responsible for defining and delivering long-term user value. According to Wikipedia , marketing refers to a series of activities undertaken by an organization to satisfy customers, including the continuous creation, delivery and management of intangible value of products, with the ultimate mission of maximizing profits.

In short, operations focus on short-term value, products focus on long-term value, and marketing focuses on tangible and intangible product value.

Through the above key operations, product life cycle, and the total download volume of KuChuan, we can easily see that the main growth stages of users of this product are mainly in the two periods of stable growth and rapid growth, that is, from the end of 2015 to the middle of 2018. The total growth during this period reached 1.2 billion downloads, doubling in 15 months, doubling in 9 months, and doubling in 6 months. The growth rate (20 million/month, 330 million/month, 500 million/month) has increased year by year.

The above figure is obtained by combining the growth rate and time rate. It can be seen that the product experienced different stages of growth in the 30 months, and serves as the time standard for the AARRR traffic funnel model.

1. Acquisition

The methods of user acquisition are mainly divided into product (content) acquisition, operation acquisition and marketing acquisition.

Various promotion methods are used, such as offline promotion , advertising , KOL promotion, APP application store distribution , etc., with the aim of acquiring new users, which is also known as "attracting new users" in operational jargon.

Product and content acquisition:

Lay the foundation for user acquisition from product content production and content recommendation:

1. Himalaya has been intensively cultivating its product content since 2013, and has collaborated with industry leaders to create high-quality courses to ensure the quality and richness of the content. Most of the big names in the high-quality courses bring their own traffic and directly gain fans of the speakers to a certain extent, such as Guo Degang and Cai Kangyong's emotional intelligence courses and talk shows. At the same time, the content covers entertainment, broadcasting, social sciences, humanities, etc., based on long and short tail user groups, laying the foundation for their diversified vertical demand acquisition.

• Free content: audiobooks, music, crosstalk, radio dramas, anchor radio stations, and headlines (partial);
• Paid content : master classes, boutique mini-classes, live micro-classes, Xima lectures, audio books (paid version);

2. The product content is clearly categorized, and the recommendation system is highly complete, covering classic must-listen, paid premium, daily must-listen, and one-click listening, which facilitates the selection and recommendation of content for new users. The search and classification on the homepage and discovery to a certain extent avoids the situation where users are at a loss when consuming content.

Operational actions:

Operational actions from KOL recruitment and UCG anchor promotion:

KOL attracts new users: Most KOLs have their own traffic, which directly brings new traffic and fans to the platform.

For example, at the end of 2017, Mi Meng’s team launched a paid audio course on Himalaya FM called “Mimeng Teaches You How to Make a Monthly Salary of 50,000” and promised that “if the salary increase does not exceed 50% after three years, the course fee will be doubled and refunded.”

This promise was a gimmick to attract traffic , and it successfully helped Himalaya bring in a group of paying customers (of above average quality).

UGC+PGC attracts new users: Himalaya recruits anchors from the B-end and C-end, builds Himalaya University, and improves the anchor training and upgrade system. It not only provides training, display, and monetization opportunities for professional/non-professional anchors, but also connects the resources of multiple live broadcast families for the platform.

The exposure has been greatly increased through anchor training, the Ten Thousand People and Ten Billion New Voices Plan, and the annual anchor voting and sharing, which has played a significant promotional effect and indirectly driven the growth of product users.

Marketing actions:

Offline promotion (city agent promotion + offline advertising): City service provider recruitment mainly sinks the product promotion channels , brings user traffic through local personal resources and event organization, and supplements online promotion.

At the same time, it is promoted through subway advertisements in core cities. The picture below shows the ground promotion of the 123 Knowledge Carnival.

Online promotion (sponsorship of large-scale events/programs + cross-platform cooperation): During the 2018 New Year's Eve, Himalaya and Zhejiang Satellite TV cooperated to organize the "Thought New Year's Eve Event" to allow major KOLs such as Ma Dong and Gao Xiaosong to conduct live talk show discussions on the seven hot keywords that young people are concerned about.

At the same time, in early 2018, it sponsored Hunan Satellite TV's high-rated variety show "Sound into the Scene", which to a certain extent enhanced national awareness, obtained TV user traffic, and drove product growth.

Through the Knowledge Carnival in 2016 and 2017, the revenue exceeded 200 million yuan. By distributing coupons and launching large-screen advertising in subways and major platform advertising pages (as shown in the figure below), UC Browser has largely formed the impression of "the knowledge payment version of Taobao" in the minds of users, thereby achieving the goal of attracting new growth.

At the same time, from the analysis of key operations, it can be seen that Himalaya distributes product promotion channels through the hardware market, and as many as 2,000 brands of hardware facilities such as audio, car, and mobile phones have access to Himalaya's content.

It also promoted the use of in-car applications in the in-car market through the 2016 Shanghai International Auto Show. At the end of 2015, it reached a cooperation agreement with Didi Designated Driver. All car owners who use Didi Designated Driver have the opportunity to receive free "In-Car Listening" and exclusive designated driver vouchers provided by Himalaya, which indirectly brought in more traffic.

1. ActivationUser activity:

To help users experience the AHA moment, Himalaya has put a lot of effort into customized product recommendation interfaces, operations (membership benefits and new member discounts), and marketing (hardware products).

Active products:

Product activity is mainly reflected in the two columns of personalized recommendation customization and daily customized recommendation.

Personalized recommendation customization:

For example, after a new user completes registration, in order to better push columns and audio content that the user is interested in on the recommendation page, personalized topics, hot words, gender and age will be selected. At the same time, as an old user, you can also enter the listening preference settings from the settings in the account center, but the entrance design is relatively hidden, which makes it inconvenient for old users to re-personalize the settings (as shown in the figure below).

Daily custom recommendations:

From the Daily Must-Listen and One-Click Listening on the homepage, you can sort out the classified recommendations and must-listen displays based on the user's daily listening habits and preferences, and indicate on the Daily Must-Listen that "generated based on your taste, the more you listen, the more accurate the recommendations" and other words. It is worth noting that in addition to the homepage entrance, you can also enter the daily customized recommendation interface on "I Listen". This shows that Himalaya's strategy of investing heavily in big data personalized needs since 2014 is being improved and realized step by step in product iteration.

Active operations:

The active operations are mainly reflected in the upgrade of membership benefits, new member coupons, and the points redemption gift mechanism.

In terms of membership benefits, users can enjoy a 7-day free trial and can cancel at any time during the free period. Users can first experience the membership benefits and directly feel the difference before and after becoming a member. There is also a sufficient buffer period. Seven days can be used to cultivate a new habit, allowing users to experience the quality of the platform content and the discounts on course purchases. To a certain extent, users will have the urge to continue to renew (formally purchase) membership after trying it out.

The new user coupons satisfy the user's need to purchase courses after successful registration. As shown in the figure below, there are new user/festival exclusive coupons, which make users feel that the product is "not bad".

The design of the points redemption mechanism gives users a feeling of "gaining knowledge, having fun and receiving gifts/vouchers" which is a bit unexpected. It allows users to feel a small sense of reward and accomplishment when they see their points increase.

At the same time, as the usage time increases, users pay a higher opportunity cost and are more likely to be reluctant to leave the platform. As shown in the figure below, it can be noticed that users can make money by participating in topic discussions and sending red envelopes in the points mall interface, allowing users to have the satisfaction and sense of accomplishment of being rewarded for their knowledge output. However, the entrance is not obvious enough and is difficult for users to find. It is possible that most users have not experienced this function.

Active Marketing:

By planning a series of stories about Xiaoya AI speaker, children of different ages can have different experiences with Xiaoya's audio content while being satisfied at the same time, thus solving children's cultural and entertainment needs.

At the same time, there are smart applications for adults' work and rest scenarios, making users excited about the access to this hardware product, and indirectly increasing the number of Himalaya users from the hardware-derived products.

1. Retention

Retention is mainly sorted out from the aspects of product retention (live broadcast, hardware smart device connection, etc.), operation retention (level and points mechanism), and marketing (66 membership days):

Product retention:

From the perspective of hardware smart devices, the product is connected to hardware devices, providing another way for users to retain users. Instead of just listening to audio on mobile devices, users can also use smart devices to connect and use them. This will increase users' dependence on the product to a certain extent, making it impossible for them to leave the platform.

From the perspective of the live broadcast function and the host's backend management function: live broadcast interaction is a function that allows the host to establish a close relationship with the listeners, and instant comments, feedback, questions and rewards will make users feel real and more pleasant during the interaction than receiving comments.

From the live broadcast footage, we can see that Himalaya mainly positions live broadcasts under themes with strong interactivity and emotions, which can greatly improve user retention rate.

Combining the host's backend, we can find that the data trend of the host's backend changes significantly, and the group management functions are rich. It can be inferred that the live broadcast and host's backend management are to enhance the intimacy between users and anchors, and to improve the user retention on the platform.

In comparison, the circle function of the course is more formal. Its core focus is on the discussion of high-quality content and the pinning of essential content . At the same time, it is noted that the dynamic function of discovery tends to update the content of interest that one is interested in through personal recommendations.

In short, it is speculated that live streaming and circles are two social functions with different levels of intimacy, which are different from PGC and UGC audio content. They can be used selectively for different audio content to promote user retention.

Operational retention:

The user level growth system allows users to visualize their time and upgrade records, which is very direct. At the same time, the ancient imperial examination system is used to classify each level and divide the redeemable gifts. The risk loss aversion of human nature (unwillingness to give up the opportunity cost of listening time and records) is used to increase user stickiness to the product.

On the other hand, old users can obtain corresponding coupons or commissions by inviting new users to register/purchase membership, which makes old users more willing to share audio content. Not only can they get benefits themselves, and then use consumption discounts on the platform to continue to use the product, but also let others see it on social channels, using weak and strong relationships to expand new users and achieve the goal of user growth.

Marketing retention:

For the 66th Membership Day 50% off unlocking guide, we used the three-step method of sharing to get free membership, and carried out large-scale promotion and marketing. We used the slogan "Limited time 6666, 0 yuan flash sale, share to get xx month membership + xx coupons + xx gift draw" to create a sense of scarcity among users, and encourage them to participate quickly to receive benefits.

The campaign was quickly promoted across the entire network, not only encouraging old users to continue using the product, but also directly promoting new users and increasing user growth.

1. Revenue:

The revenue is sorted out from three aspects: product (membership fees and course fees, hardware), operation (social and community), and marketing (advertising):

Product revenue:

It mainly comes from paid courses (master classes, boutique small classes, live micro classes), 1-year (monthly or annual) membership listening packages with automatic renewal function, and some smart hardware products within the mall, which drive user consumption.

Operating Income:

The celebrity courses that Himalaya has introduced since 2012 are all high-quality courses. Before joining the celebrity course customer discussion group, you must first pay a fee and then the host will accompany the fans.

Judging from the sales figures, the first session in 2016 achieved around 5 million eliminations.

The first sales volume of 50 million was almost the same as Taobao’s Double Eleven turnover in its first year. The sales volume quadrupled in the second year. The sales volume of the first 66 Member Days in 2016 also reached 61 million. Therefore, intangible knowledge products were monetized on a large scale through creating festivals.

Marketing income:

Plan and analyze marketing activities through location advertising, audio advertising, radio, etc.

1. Referral self-propagation:

Self-propagation is mainly divided into three aspects: products, operations and sales. In order to generate revenue through marketing, Himalaya held two 123 Knowledge Carnivals, and through the creation of festivals, it built an excellent way for itself to quickly monetize.

product:

You can earn commissions by sharing courses, and you can get coupons ranging from 20 yuan by sharing course check-ins and attracting new users. In addition, you can also receive benefits for free by giving audiobooks to your friends.

operations:

You can share the course with your friends for free, and you can get coupons ranging from 20 yuan by sharing the course check-in and attracting new users. At the same time, we use the Himalaya Listening mini program to deliver courses for free.

The total welfare content drives us to take action immediately, and to a certain extent promotes mutual dissemination among users, but the income of the product is still limited.

Marketing actions:

The 123 Knowledge Carnival, 66 Membership Day, and 423 Listening Book Day cultivate users' awareness and habits of creating festivals, allowing users to form an addictive cognition of shopping during festivals, forming an atmosphere of self-propagation. At the same time, the number of downloads increased by 2-3 times during this period, which shows that the purpose of user growth has been achieved and commercial monetization has been carried out.

Summary: Through the above key operations and product life cycle analysis, it can be simply seen that the product involves various elements in the AARRR model in the above three stages, but does not follow the traditional order of user growth and screening. Instead, retention, conversion, revenue, and self-propagation process appear in parallel or simultaneously during the product growth process.

Optimization process

From the above simple analysis and summary, we can conclude that: in the process of sorting out, we can see the presence of three major players, namely product, operation and marketing, in every link. It can be seen that the development of Himalaya itself is in line with the AARRR model, but the direction under the traffic funnel is generally vertically downward at different stages, and the funnel models at different stages are different.

After inventory and analysis, considering the total product downloads and future trends, I personally tend to think that there is still some shortcomings in user activity and user retention.

Recommendation 1: Refine user preferences and make recommendations to help users get started with the product faster and continue using it

Most registered users only set their preferences when they first register.

"If you need to adjust your preferences as you grow older and gain more experience, otherwise it will be like living in an information cocoon for a long time, and you can only see what you prefer," which will affect user growth and improve the recommendation algorithm on this basis. You can try to develop how to comprehensively integrate the combination of "daily preferences + topic-related selections", reduce the repetitiveness of excessive content, make the content concise and original, and try to integrate preferences with novel content interfaces, indirectly promote user acquisition and retention, and ultimately achieve user growth.

Suggestion 2: Encourage more UGC content creation, and drive high-quality users to gradually convert into anchors in the future.

First of all, following the trend, Himalaya's C-end and B-end content production, anchor training system, etc. are currently on the rise, covering the Ten Thousand Billion New Voices Plan and students of Himalaya University.

Combining Maslow's self-actualization needs with the needs of self-actualization, it is a particularly wonderful thing to encourage users to create more and realize their spiritual pursuits. It indirectly creates an invisible stickiness between users and the platform, achieves better user retention, and ultimately realizes user growth.

Source: Meinnnn