For the online education industry, low-cost customer acquisition is one of the goals everyone pursues. There are many ways to acquire customers at a low cost. The author of this article introduces one of them - referral, that is, your customers help you introduce other customers or share your course content for promotion. So how should referrals be used? Let’s look at the author’s analysis.

After several years of rapid development, the online education market structure has gradually become clear. The era of relying on financing to enter the market in the early days has gradually become a thing of the past. For online education institutions, it has become very important to find more low-cost traffic and to let users who have become students bring us traffic.

Each company has also put a lot of effort into referrals. According to media reports, the referral data of these companies are simply enviable.

In May 2020, Xigua Programming received a B+ round of financing. According to the referral data, on average, each Xigua Maker old user recommends 3-4 new users to the platform on weekdays, and this data has increased significantly during the epidemic.

As of August 2019, Spark Education has nearly 60,000 full-time students, with a renewal rate of 80% and a referral rate of 75%. There are more than 1,800 full-time lecturers and tutors.

On March 20, 2018, "Baby Play English" officially announced the completion of RMB 150 million in Series B financing, with more than 20% of its students coming from 100% referrals.

The referral rates of the above three brands alone are enough to support a company on a large scale.

However, facing the referral operation suggestions on the market that tend to disassemble the process, many partners responsible for referral growth have found that the seemingly simple referral process is not effective in my company.

Through 356 days of practice, this article attempts to introduce to you from the perspective of the company's overall operations: What is the essence of referrals? What are the current mainstream referral designs and data of various institutions? What are the challenges and opportunities for referrals?

1. What is the essence of online education referrals?

Referral means that after you have provided a series of services to your clients or potential clients, the clients will be satisfied with your products or motivated by some kind of motivation to introduce you to their friends, relatives and other potential clients to experience your courses or services.

In the early days, offline institutions could only provide in-store experiences through face-to-face meetings, but now with the advent of online social tools such as WeChat, users can conveniently recommend courses to friends who are not offline geographically related through WeChat.

Online learning is a long process. Once online institutions streamline word-of-mouth referrals through Internet products, the referral mechanism will become a very important means of customer acquisition for online education institutions. Institutions can effectively manage data and control processes of the productized referral system, thereby achieving ultimate bottom-level customer acquisition.

In this way, the essence of online education referrals is to ensure the ultimate product learning experience, through the design of online education learning processes and the analysis of user behavior habits; to design appropriate incentives for different users and provide appropriate social sharing products, so as to realize the process-based, data-based, transparent and universal learning of referrals.

In the above financing cases, referrals seem to be very profitable and the execution process is simple to implement, but why do most of the companies that follow suit fail to succeed?

2. Build an online education system with an 80% referral rate from scratch

From early 2018 to early 2020, check-in sharing in the form of posters on WeChat Moments became an important means for many companies to acquire customers. Because of its simple and efficient way of acquiring customers, many early-stage operating companies have reaped a huge wave of dividends. Some companies have imitated all the processes, but they just can’t make the customer acquisition method of check-in and sharing work.

In offline education, user sharing is listed as referral, while online people call it sharing fission. Since user sharing fission is a referral, how big is the user referral in the social scenario of WeChat social scenario? Why do users share? It is a very important step in designing a referral campaign.

How to automate user referrals using online products is also a very important breakthrough.

2.1 How big are the referral opportunities for online education users?

In offline education, there are two common types of referral opportunities for an offline student: community parents and classmates’ parents.

The main reason is that a student's offline social radius is fixed, and the selection radius of offline students is also commonly within 3 kilometers of the campus - this is why the main forms of offline campus marketing activities are: school promotion, community promotion, and cross-industry cooperation among similar institutions.

Once referrals move online, the user communication and sharing channels of educational institutions will become online, and the students' referral capabilities will be amplified several times because of the online nature.

When calculating an organization’s online opportunities, an average user’s social relationships in the WeChat ecosystem can be divided into two categories:

1. Contacts: Family, friends, colleagues, parents of children’s classmates, and friends you get to know through learning communities related to self-improvement and children’s growth;
2. Groups: family, friends, colleagues, parents of children’s classmates, and groups of friends met through learning communities related to self-improvement and children’s growth.

Now that we know the social relationships a user can influence, what is the approximate proportion of users that a user can refer or influence to make referrals?

At the sub-forum of [Internet + WeChat] of the 2018 Tencent Global Partner Conference, WeChat officially released its first public class on WeChat user data: 60% of WeChat users are young people (15-29 years old), and young people have an average of 128 friends.

If the number of WeChat friends of users born in the 1980s and 1990s is increased by 30%, that would be around 200 people. Based on the 30% proportion of married people with children, a parent of a child of school-age can cover around 60 people.

60 people are just the maximum number of people that one person can cover. In addition to the strongly linked relationship chain, the social relationship chain of an ordinary user is also very large. Even if each user has 8 types of social relationship chains, and each community has 100 people, then the pan-social relationship chain has nearly 800 people; if the precise relationship chain community relationship is calculated at 40%, there are almost 400 people.

Assuming that an online education institution has 10,000 users, then its possible pool of accurate referral users is 600,000, and the possible overlapping weak-relationship users is as high as 4 million. If we calculate based on 10,000 new users per month for each institution, there is a potential pool of 600,000 accurate users behind each new student added each month waiting to be tapped by each institution.

Based on more than a year of distribution referral practice, the average number of referral users brought by an active referring mother is around 10. With such huge commercial value, it seems worth it for all companies to promote referrals as a strategic project.

From the perspective of opportunity, there is a lot of room for referrals; from the user's perspective, personal friends are my core trust assets, and the community is my important circle asset. If it is a recommendation, there must be a reason.

So let’s take a look at why users share?

2.2 Why do online education users share?

As an ordinary user, if I want to recommend the products of an institution: first, I basically trust the products; second, my learning experience is good; last but not least, my children have achieved good results in learning with you.

Recommendation is generally a difficult behavior for ordinary users, and it may come from their own rejection. People with different personalities have different execution of recommendation behavior, for example:

• Rational parents: No need to introduce too much, the key is to bring out the effect of children, and their recommendation behavior will not be too big;
• Enthusiastic parents: They pay attention to their children's performance and give timely feedback. They make many recommendations but the results are average.
• Kol type parents: focus on their children's learning situation, provide timely feedback, and their recommendations are effective and frequent, with the best results.

In offline education, campus referrals generally use:

• Sign up and get 3 free trial classes that can be given away;
• Successful referrals will receive additional lessons and prizes;
• Invite your friends or colleagues to attend the sharing session to recognize outstanding students for free.

From the perspective of mainstream design, it is generally about greed and showing off. The core point is still based on the benefits for participating students, and they can participate and show off their children's performance. The strategy is not difficult. When the sales results are certain, the key still depends on the execution of the offline referral person.

After users enter the online world, they can share more nodes in online scenarios, which generally revolve around the entire life cycle of users' online learning: registration experience period - experience period - registration period - formal student period - renewal period - hardcore student period.

The needs of users sharing are also different at each stage:

1. Registration trial period: inform users that I have tried a course and that I chose a good course and need someone to learn with me;
2. Experience period: my experience with a product, my child’s learning situation, and my child’s learning certificates;
3. Registration period: I have just selected a course, help me get a discount. I chose a class and my child had a good experience. I recommend it to you. My child’s learning certificate;
4. Formal student period: My child studies every day and has good results. I am participating in a PK activity. Participating in the PK will give you a prize, a certificate for your child’s study.
5. Renewal period: I just selected a course, which helped me get a discount and a certificate of learning for my child;
6. Hardcore student period: I have chosen a career that we can all come together.

From an understanding of the needs shared by parents at different stages, the above needs also confirm the three most common character defects in human nature:

1. Lazy: Package Offers
2. Greed: free trial, group purchase, flash sale, lucky draw
3. Vainglory: Certificate of Merit

Therefore, to understand why parents want to share, we should start from the character defects of human nature and find reasons for parents to share based on the quality of education. I believe that we can achieve twice the result with half the effort.

2.3 Looking through the entire life cycle of online education to find the golden node of referral design

In the previous article "How to establish a golden closed loop of low-cost user growth in online education when investment continues to suffer losses? In the article "Online Education User Flow Map", we introduced the user flow map of online education. In the user flow map, there are also different referral opportunities hidden. Let's take a look at them one by one:

User turnover is divided into four key periods: registration experience period, experience period, registration period, old student period, and hardcore period. The core points of user service in each core stage are different, as follows:

1. Trial period: Build confidence in online learning and experience the benefits of the course;
2. Registration period: emphasizes the urgency of learning;
3. Renewal period: Emphasize the benefits of continuous learning and provide feedback on the child’s learning progress;
4. Hardcore period: emphasize the future value of good products and good businesses.

Once you have found the service focus of each link, you will lay a good foundation for establishing the golden point of referral based on the learning process. After understanding the key points of the user turnover cycle, the key to implementing it in the learning link is how to stimulate users' desire for referrals.

Taking Spark Education’s math class as an example, the children are mainly between 6 and 12 years old. Let’s take a look at the closed loop of user learning: in-class learning - after-class practice - after-class training - after-class review - stage test feedback, which are five key learning links.

This leaves parents with a good golden point for referrals, as follows:

• In-class learning: A Spark class will have five parts: course introduction, new knowledge exploration, course review, course summary, and class comments. In each part, the teacher will occasionally invite students to answer questions, record the situation of each child, and give feedback to parents;
• Summary video: The teacher summarizes the course and then tells the students the key points of learning;
• Homework consolidation: students take offline teaching aids to take learning tests;
• Students explaining the questions: This stage is the most interesting and is also the best way to show parents the learning results. Students explaining the questions need to be recorded and uploaded to the teacher;
• Topic assessment: There will be test questions based on the learning situation at each stage to ensure that students can apply what they have learned to other situations;
• Fun Challenge: This Spark Thinking event is called the Thinking Games, a collective PK test. There will be offline medals, certificates and other physical objects to stimulate parents' sense of honor.

Spark Thinking not only ensures good learning results, but also closely focuses on the entire learning process and designs good sharing points, such as:

• Certificate of excellence in your child's course;
• Grading of children's homework after it is uploaded from offline workbooks;
• Video of the child explaining the questions to the parents after learning the course and the teacher's rating;
• Certificates for participating in special assessments and fun challenges or obtaining good results, etc.

In the five key periods of user conversion, each institution used different referral strategies in each period, as follows:

1. Registration trial period: There are rewards for registration invitations, discounts for group purchases, points for check-in sharing, points for shared views, and rewards for successful invitations, such as: Watermelon Programming, Baby Playing English, and Zebra AI Class;
2. Registration period: Sign up and share to get instant rewards, for example: Vipkid;
3. Old student period: points for check-in sharing, points for shared views, rewards for successful invitations, rewards for PK activities, points for renewal sharing, discounts for group purchases, for example: Vipkid, Spark Thinking, Zebra AI Class;
4. Hardcore student period: long-term rebate distributors, franchised regional partners, franchised regional dealers, such as Spark Education, Yunshuxie, and Fandeng Reading Club.

Based on the public data of the three surveyed companies, each company has different referral designs:

1. VIPKID: The main recommended product is still monthly sharing. Recently, we have launched activities such as free classes, invitations to receive physical objects, and photo awards. The core is still to carry out activities in a multi-dimensional way based on the form in which users are willing to participate, and there are long-term exchange coupons for energy stones.
2. Spark Thinking: The main referral forms are sharing posters and sharing screenshots every month. We also acquire customers through quiz PK invitations, and have long-term exchange of Spark Coins.
3. Zebra English: It mainly uses monthly sharing and poster sharing as referral activities, and innovatively designs recommendation ranking rewards and invitation visit lottery rewards to increase the willingness to recommend.

The three event designers are mainly based on students who have registered and experienced the event. Zebra has tried the free invitation link. Currently, there is not much difference between the companies in product innovation for referrals.

The sharing process and the motivation for sharing are roughly the same, but why is the effect better for you and worse for me? The key factor is the involvement of the executors.

2.4 Full-process benefit design for relevant departments of online education companies

In offline institutions, customers are generally acquired through marketing and then sales, who are the first to have contact with customers and are also the group with the most users in offline education institutions.

A small number of converted customers will be included in the contact list of class teachers and teacher services for long-term maintenance.

In offline institutions, sales referrals are generally given more emphasis; and the class teacher and teaching teacher ports also have certain referral indicators, but relatively speaking, the relevance of interests is mainly based on closing deals.

Suppose a salesperson is paid a certain amount for each successful referral, a class teacher is paid a certain amount for each successful referral, and a teaching teacher is paid a certain amount for each successful referral - benefit design is generally based on relevance rather than absoluteness, thus ensuring that everyone can promote each other and bring in referrals collectively.

For online agencies, it may seem as if the process is online and referrals are easy to execute. However, based on what we know, the data correlation is poor, and there are even situations where customer boundaries are not very clear, especially for small and medium-sized startups, where referral effects are not good.

This is what we mentioned earlier, the fundamental reason why the seemingly simple referral process does not work well after being planned by the operations growth department.

According to the investigation, VIPKID, Youdao and Spark Education have made relatively detailed system architectures for referrals, which are more instructive in terms of specific implementation. Let’s take VIPKID as an example to look at the design of referral benefits.

2.4.1 Course Sales Referral Promotion Strategy

Sales is the core link of educational enterprise revenue. In terms of referrals, some sales are active while others are not. This is because the expectations for referrals are good but the results are not very clear. To promote referrals through course sales, we need to pay attention to the following three points.

2.4.1.1 Related to the core to which the sales particles belong

After users pay to register for VIPKID, the salesperson will tell the user to share the poster to their friends circle immediately, and they can immediately get the extra class hours. The list of referral customers acquired through sales actions will flow through the system into the current sales pool to ensure that the referral promotion is related to their own interests.

Taking Youdao as an example, customers who have completed community sales will be promoted for referrals, and customers will return to their next experience camp, ensuring that sales can place orders relatively stably when there is a certain pool of old users.

2.4.1.2 Bringing referral users will receive particle rewards

Although it seems that referrals are closely related to the final sales performance, because there is a certain sales pressure in sales itself, there is also sales laziness.

The tool can be used to set up separate rewards for orders brought in by sales referrals, which can not only increase sales motivation but also save costs for the company overall. However, in terms of specific implementation, some companies believe that employees are employees of the company and there is no need to repeatedly issue rewards for non-core responsibilities.

2.4.1.3 Sales promotion methods

Some companies also set up a form when registering, which is that after registering, you can share it on your circle of friends and enjoy an exclusive discount and welfare package immediately - this can not only help promote sales, but also enhance the user's experience of participating in referrals.

2.4.2 Strategies to Promote Referrals from Learning Service Teachers

Sales are under pressure to achieve sales performance at the front end. If the income from referrals is not obvious, the motivation to participate is generally not very strong.

When users reach the back end, the learning service teachers may not be very willing to participate, because in the early stages of students' learning, the pressure on learning service teachers is more on promoting students to complete the course and give good reviews, and the core is to promote student renewal.

However, because the after-sales service teacher is a relatively long link in the entire student cycle, this stage is particularly important.

Therefore, at present, when users sign up, each institution uses a very simple strategy, which is to share rewards in the circle of friends. However, in order to better promote referrals by service teachers, attention still needs to be paid to the design of measures.

2.4.2.1 Bring in referral users and get particle rewards

This is the same as front-end sales. Few institutions set up referral rewards. This will cause service teachers to not pay attention to this work. Referral work is mainly based on rewards and is not a core job, but rewards are needed to ensure its implementation.

2.4.2.2 Rewards for bringing referral users to close deals

Because the reward for referral particles is relatively small, in order to ensure that the class teacher continues to make referrals, a relationship between the user transaction and the service teacher can be established.

2.4.3 Referral Promotion Strategy of the Referral Distribution Operations Department

In order to deepen the promotion of referrals, some companies have launched special referral distribution operations departments.

In the early days, many companies focused more on referrals based on non-paying high-intent users, and a few large companies would open up formal students to join as referrals for paying students. However, in the end, many companies' non-paying high-intent user referrals ultimately failed, so a separate promotion strategy must be developed.

2.4.3.1 There are rewards for referring user particles

The operations department in online education companies is generally rewarded based on performance-based accounting. The task indicators are relatively broad, which does not provide enough incentives for the executors, and they are paid only for what they can accomplish.

2.4.3.2 Need to continue to attract paying students

In referral operations, if the students are non-paying for a long time, their activity level will continue to decline and the quality of users will be relatively low, requiring sufficient support from the company.

Based on our referral practice over the past year, we have found that only when the execution and outcome benefits of each link are deeply bound together can we ultimately achieve an increase in the number of participants and an increase in transaction results more effectively.

Of course, there will definitely be the problem of multiple reward issuance in the performance accounting of user relevance, but compared with acquiring customers through a single market channel, it is still worth it for each education company to make in-depth attempts and practices.

2.5 Design of online education referral sharing products

As far as the referral products of various online education institutions are concerned, they are mainly in the following forms:

1. 【Poster】Learning course posters: Officially designed posters, or customizable learning posters;
2. 【Poster】Lecture content: lectures by big names or rare content produced by institutions;
3. 【Connection】Data report of learning courses: learning report based on the learning process;
4. 【Connection】Related learning articles: related to the course and stage content;
5. 【Connection type】Mini program group buying or coupons.

Judging from the application of referral products seen so far by various companies, the main force is still using products directly related to the courses to acquire customers. However, with the proliferation of posters from various companies, the effectiveness of the posters is gradually declining.

Although the customer acquisition effect is declining, the content format based on posters continues to innovate. For example, the poster is a lecture by a famous teacher rather than direct course content. Companies like New Oriental, GSX, and Yuanfudao have all launched courses on the qualities of famous teachers to acquire users on a larger scale.

Connection-type referral products are more common, such as learning-related articles, course learning reports, etc. Compared with direct course posters, content-based links are more likely to be more appealing to users, allowing them to understand the content in advance and achieve effective conversions, thereby increasing the effectiveness of customer acquisition.

In addition to the product form, the pricing of the product is also crucial to the effectiveness of customer acquisition. The current mainstream categories are as follows:

1. 0 yuan: The course is mainly a 5-day recorded course, mainly suitable for the early stage of user experience;
2. 9.9 yuan: Applicable to both recorded and live courses, and to both free and paid customers;
3. 29.9 yuan: The courses are mainly recorded courses, applicable to both free and paying customers.

Judging from the current referral product designs of various companies, most of them are free courses, while the referral products of Zebra AI courses are mainly paid courses at 49.9 yuan. For companies at different stages, different companies should design referral products based on their own company situations.

Assuming that your company currently has a lot of sales but insufficient particles, you can try a short-term free invitation campaign to cope with the company's short-term customer acquisition pressure; if your company currently has a large number of particles and needs higher-quality referral users, you can set up high-unit-price products for referrals.

Generally speaking, because referral activities are in the long-term promotion process, with the operational planner taking the lead and other departments cooperating, the specific leading party will be unable to control the effectiveness of the referral activities for a long time.

From a practical perspective, low-priced paid products and free products should coexist, which can continuously ensure user referrals and a certain amount of high-quality referred users.

Of course, in the early days, for example, Zebra's referral product 9.9 had factors such as low price and high overall expenditure cost, and there was also a long-term practice of being fleeced by users. This requires timely optimization of the form of activity rewards during the referral operation and execution process.

2.6 Design of online education referral rewards

With the guarantee of a good mechanism and the support of referral products, careful cost accounting and risk control are needed to ensure that referrals are finally implemented and costs are controllable.

Based on the reward form above, let’s talk about it in detail.

2.6.1 Direct rewards for sharing screenshots

This form of reward worked very well when WeChat sharing was not yet so common. Because of the design of this mechanism, many big Vs could get a lot of free classes by sharing and referring others. This was especially true in the early days of VIPKID, and VIPKID also gained a lot of high-quality customers.

Based on current practices, the cost of directly forwarding screenshots of Moments is relatively high, but the effect is unclear.

2.6.2 Rewards for successfully placing an order for a trial listening invitation

This form is generally implemented by small and medium-sized startups. Because the recommendation system is not related to the learning system and it is impossible to know the user's learning situation, they are forced to implement it. This form currently has a lot of room for fake orders, especially the free trial invitation rewards.

2.6.3 Rewards for successfully inviting a trial class and placing a trial class order and completing the class

This form generally involves a self-developed learning system. The referral process is related to the learning process, and rewards are given to referred users after they complete their learning. This will greatly improve the effectiveness of recommending users.

2.6.4 Rewards for successfully inviting a trial class to place an order and then purchasing the class after the class

This form is a way to focus on the recommendation effect. Generally, courses like live broadcasts will choose this form. Because the cost of live classes is relatively high, AI courses can implement full life cycle rewards in referral design before starting profit sharing and distribution.

Because the recommenders are sales or learning service teachers, there is no atmosphere for users to PK with each other. Generally, the referral activity design will design a tiered reward plan. In order to improve the overall PK effect, the overall monthly ranking rewards will also be designed to stimulate users from different stages and dimensions.

Referral rewards are generally the following:

• Virtual currency: a kind of points related to the learning process designed by the institution, which can be exchanged for completing the registration separately. It is generally applicable to products with free invitations for trial listening and orders, and can be exchanged for coupons, physical objects, and separate new courses.
• Exchange Coins: The calculation of successful recommendation rewards is carried out separately, which is not related to the points in the learning process and can be exchanged directly for physical objects;
• Lucky draw: The calculation of successful recommendation rewards is carried out separately, which is not related to the points in the learning process and is used for lucky draw incentives;
• Class hour rewards: generally applicable to live classes with real people, high value;
• Cash rewards: generally applicable to low-priced paid trial courses, directly entered into the user's cash account, and the redemption process is relatively complicated.

Relatively speaking, the combined operation of virtual currency and exchange currency is more common, and this is also the preferred form for institutions to implement referral systems in the future. However, there are certain development requirements, which is more difficult for start-ups.

Startups generally focus on sharing screenshots and successful orders, which can achieve certain results, but in the long run there are more risks in terms of effectiveness.

3. A brief discussion on the risks and near-future development of online education referrals

Referrals do play a crucial role in the current project. The customers brought by referrals each month account for more than 80% of the entire market’s customers, creating excellent conditions for the company’s steady development.

Therefore, all online education institutions must attach importance to the referral business and conduct strategic referrals based on the company's overall situation.

From current observations, referrals from educational institutions of various small and medium-sized enterprises are still in a relatively early stage. In the absence of suitable IT products to support them, referrals are in a chaotic stage, or even a stage where they appear to be effective, but the effect is unclear.

If your company is concerned about referrals and doesn't know where to start, I hope this article will be helpful to you and allow you to effectively promote referrals while having a good grasp of referral costs and internal interests.

In the long run, customer acquisition for educational institutions has shown a relatively obvious trend of fission growth at the front end, referrals in the middle end, and investment and franchising at the back end. Referrals are the most effective structure at present - it not only supplements the fission growth at the front end, but also lays a good foundation for investment and franchising for the brand in the later stage.

Finally, let’s summarize the core points of this article:

1. Referrals are based on user trust in the course service, not a simple operational action;
2. The emergence of WeChat social networking provides a good technical foundation for online referrals, doubling the efficiency and effectiveness;
3. When building a referral system from scratch, you need to pay attention to the user's willingness to share;
4. Referrals cannot be separated from the learning process. Finding the referral point in the learning process will achieve the best efficiency and effect.
5. Many small organizations are not good at referrals because they lack a holistic design of the company’s overall structure;
6. Although referrals seem simple, they have requirements for the design of referral products and rewards used by students.

Referrals are very simple in a small operational process, but due to the lack of upper-level design, the results are generally unsatisfactory. I hope that after reading this article you can avoid detours.

Author: Yifan

Source: Operation Hunter