Useful information: In-depth analysis of 5 key indicators to build your own AARRR operating model

Every time I talk with my peers about operations , users, and products, we always come back to the AARRR model . This user life cycle model summarizes the five key links in the operation of Internet products . Customer acquisition is the foundation of operation. Promoting user activity can make the product viable. Improving retention and reducing churn can make the user scale larger and larger. Paid conversion is the key to the continuous operation of products and teams. If user self-propagation is done well, the product operating costs will be reduced a lot and growth will be fast.

However, now that I understand the principle, how can I apply it to my work? Attract new customers , promote activation, retain customers, pay, referrals. Everyone knows these words. Let’s talk about the details:

Do you know where your high-quality users come from? If we only use the data from the channel promotion platform to see how many conversions it brings, how can we analyze the sources of high-quality users? The customer acquisition process requires not only quantity but also quality.

Do you know how many of your 7-day active users are active for 7 consecutive days and how many are active on the 7th day? If you don’t know, how can you define active user groups and design targeted activities?

Speaking of retention, how do you design your retention indicators? Are there any effective channels to reach dormant users? Speaking of conversion, if you are an e-commerce user, can you find out how many users give up at the last moment of payment? Are there any means of reaching out to users to encourage them to buy?

If none of these are available, how can we construct an AARRR model?

1. Learn to build your own A AR RR model

As shown in the figure, a table is drawn for the AARRR model, matching the five key links with the corresponding operating methods , combining product characteristics and team capabilities, and designing specific executable plans for each step.

Acquiring users : There are many ways to acquire users, but the cost is very high. The promotion method should be selected based on the industry, product, company status, development stage and team capabilities;

Improve activity : To improve activity, the product needs to be designed according to its functions. Content-based products (news/community) highlight the interesting, useful, relevant, timely, and novel content. Commodity-based products (e-commerce) design activities based on hot topics and groups. Tool -based products need to find content or other active points to increase user usage frequency.

Improve retention : There are two ways to improve retention. One is to carry out sticky activities for all users, such as punching in, membership cards, and points. The other is to carry out targeted operations through user stratification. Simple stratification divides users into ordinary users, active users, and core users according to the 721 rule. Content-based products retain core users and expand active users through user honor induction and privilege induction. E-commerce products can establish an RFM user stratification model to improve retention.

Guiding payment : User payment is the key to a product’s self-survival ability and is also a key indicator for commodity products. Guiding payment requires the design of payment methods and product features. Commodities can be physical or virtual, such as knowledge payment . The key to guiding payment lies in analyzing user characteristics and grasping user psychology when recommending or designing activities;

Fission propagation : Fission propagation is an operation method that is now more valued. Fission propagation can increase natural traffic , reduce promotion costs, and also has strong user stickiness. The operation method of fission communication needs to find the core interest points in the minds of users, which is the internal driving force for users to refer others.

For the AARRR model, establishing your own operational logic based on the product and current situation can effectively guide actual work. This article will be a bit long. Below are some methods and ideas for the five links.

2. User acquisition: Design the focus of user acquisition based on the product life cycle

Due to differences in industries, products, and company resources, no two companies in the world use exactly the same new customer acquisition strategy. The same new customer acquisition strategy will have different effects on the same product at different stages, as shown in the figure below.

1. Start-up phase: exploring user needs and verifying product functions

The main goals of this stage are to identify user pain points, conduct functional analysis, quickly launch verification, and gain recognition from seed users .

Channel promotion suggestions: During the channel exploration period, when the product is just launched, it is not easy to acquire the first batch of seed users.

(1) Identify user groups and find promotion directions

User demands are becoming more and more diversified, and it is increasingly difficult to satisfy all user needs with one product. Even popular platforms like Didi and Ele.me do not target all users. In the early stages of the product, Didi mainly served users who had difficulty hailing a taxi, and Ele.me mainly provided food delivery services to campus users.

(2) Determine promotion channels based on target audience

After identifying the target user group, you need to find the right channel. The key to determining channels is to understand the attributes of the channels and the mainstream user groups of the channels.

Currently, marketing channels have expanded from offline to online and mobile. In addition to television, radio and other media, subways/focus media/outdoor advertising have also been popular channels. There are countless online navigation websites/ forums and other social networking sites, and the emergence of popular mobile apps, such as live broadcast /news information platforms, application stores , etc., have also rapidly expanded the number of marketing channels.

There are too many channels to try and error out on all of them, so it is necessary to understand the characteristics of the channel’s own user base. For example: As news and information platforms, Tencent News has more young users, while NetEase News , Phoenix News , and Sina News have more users who are highly educated and have high "three-highs". If your product users tend to be younger, it is recommended to use the resources of Tencent News. If you need to acquire high-quality consumer groups, it is recommended to use Sina News and Phoenix News. If the user group coverage is wide, it is recommended to use Toutiao . Most of the characteristics of a channel’s user base must be verified through exploration (refer to the above for channel evaluation) or initially understood through secondary information, and then ways to verify its authenticity are sought. Channel exploration needs to be carried out step by step, with gradual exploration and accumulation to form knowledge sedimentation, which will be beneficial to future channel matching.

2. Product growth stage: channel launch to quickly gain user growth

In the growth stage, the product has passed the seed user period and has also gained the recognition of seed users. At this time, it is necessary to quickly increase the product's traffic/sales and brand awareness through marketing means.

The main goals of this stage are to acquire users, convert them into cash, and establish brand reputation.

Channel promotion suggestions: During the period of large-scale channel launch, brand influence can be created through media advertising. This type of media advertising can be subway advertising or online advertising such as social networking sites. With the rise of mobile Internet , information flow advertising on super apps such as WeChat and Toutiao has also become popular. Wherever there is traffic, there are advertisements.

Brand Case 1: Social + Promotion

Audi discovered that people like to post photos of their beloved cars, so it launched an activity on Instagram: inviting users to share photos and content covering a certain topic, which may be reposted by Audi, taking advantage of the social psychology of users who like to share stories of having fun with their cars in bad winter weather.

Brand Case 2: Content + Promotion

Use the UCG (user-generated content) method, rely on users to impress users, and create a "chemical reaction" in online and offline traffic - each music review in the music review train is a story, using user UGC comments to express emotions, allowing subway office workers to see the power of music; use playlists to express emotions in a concentrated manner, allowing users on airplanes to feel the power of music during their journey; use music to connect graduation stories, giving graduates and social groups who have already graduated a sense of youthful courage and strength.

3. Mature stage: Optimize channel delivery effects and improve conversion

In the mature stage, the product has become stable and the main functional points of user use of the product have been clarified. The main goals of this stage are to activate and maintain old users, while maintaining the growth of new users and continuing to achieve stable revenue and profitability.

Channel promotion suggestions: During the channel iteration and optimization period, the promotion effect of each channel can be clearly seen through the user conversion rate , order conversion rate, and channel return on investment ROI of each channel. In the subsequent large-scale launch, focus on high-conversion and high-return channels.

During this stage, more new users are brought in through word-of-mouth marketing from highly loyal old users. After all, word-of-mouth traffic is the best quality among all traffic, and the conversion rate of word-of-mouth traffic is very, very high, higher than 60%. In this era where everyone is a self-media , the purchasing behavior of each of us will be influenced by our relatives, friends and colleagues. We find that users are not just users, but also our communicators and salesmen, helping companies promote, recommend and sell products. Therefore, a company’s biggest marketing resource is not advertising in any media, but the power of sharing by every consumer. User word of mouth is a very important communication asset.

4. Decline stage: retain key channels and reposition marketing direction

In the decline stage, the product is going downhill and has gradually lost its competitiveness. Its sales and profits continue to decline, and it cannot adapt to market demand. Better competitors have emerged, and its own user churn rate is also increasing.

The main goals of this stage are: on the one hand, to operate existing users well and reduce loss; on the other hand, to explore new user groups for product application scenarios and adjust product positioning.

Channel promotion suggestions: During the channel contraction period, reduce channel investment and retain key channels.

Brand Case: WeChat

WeChat, as one of the most successful products in the social field, has the social belonging brought by the entire chain of acquaintances. However, voice intercom has no curiosity for users. The "Shake Nearby" function was successfully whitewashed after bringing a surge in users, but basically no one uses it after completing the mission of user promotion and expansion. We are also becoming more and more accustomed to the endless Moments , WeChat groups and public accounts , and the little red dot no longer makes me anxious. At the same time, the number of likes decreases, and the increase in psychological thresholds makes the experience of superiority worse and worse. Finally, the number of friends continues to increase, the number of acquaintances decreases, the social pressure increases and the relationships are diluted, the status of friends in the Moments decreases, and irrelevant articles increase, all of which lead to a decrease in users' sense of belonging.

However, WeChat did not fail quickly due to the decrease in these psychological satisfactions. Why can WeChat still grow healthily? The core of a product is the value it provides, the essence of business is improving efficiency, and the core of Internet products is to improve the efficiency of production, dissemination and transformation of information. From psychological stimulation to efficiency dependence, from impressing users at first sight to making users unable to live without you, this is a process of love and also the core wisdom of business growth.

The above are some ideas for channel promotion when attracting new customers. With the intensification of market competition, the traffic dividend has disappeared, and data-driven refined operations are the future direction. The starting point of refined operations has extended to the customer acquisition stage before users enter the product, which has completely subverted the traditional promotion ideas, from "quantity-oriented" to "quality-oriented", and from pursuing new numbers to assessing retention.

Through the data analysis platform, we statistically analyze the subsequent behaviors of users from various channels, including retention, activity, contribution, etc., to judge the quality of attracting new users, adjust channel promotion strategies , and reduce conversion costs.

3. Improve activity: Segment users and design targeted activities

1. Define the “activity” metric

User activity is an important task after attracting new users. By maximizing the value of users through activity, the cost of promotion is worthwhile. However, many teams do not have a good definition of the indicator "activity" itself. Activation promotion is defined as "taking reasonable measures to encourage retained users to establish a high-viscosity interactive relationship with the platform." Just like we may open "Pesticide" to play a round every day, but it is difficult to persist in visiting Taobao and placing an order every day. Therefore, the definition of activation standards actually varies from product to product and from business to business characteristics. Of course, it is the next level of measurement before users complete the conversion.

To judge user activity, we can introduce the "stickiness" indicator. The essence of traffic-based products is actually high-intensity usage by users. Usually, the activation strategies adopted for different user groups are different. For example, for Internet financial products, invested users may be pushed the latest financial products to guide them to re-invest. Then, whether the users have viewed this product may be a criterion for evaluating the effectiveness of activation. For non-invested users, they may be guided to bind cards and authenticate. Under such a strategy, whether the users have entered the corresponding page may be a criterion for evaluating the effectiveness of activation.

2. Segment users and label them

Activation should start from the user introduction period. It is tantamount to mending the fold after the horse has bolted before remediating it when a large number of users have "cooled down/become silent" or left. Before promoting activation, you must first understand the users. For example, if you regard the target users as patients, the task of promoting activation is to prescribe the right medicine after understanding the patient's symptoms and medical history until the patient recovers. In terms of operational work, the patient's symptoms and medical history are equivalent to user labels and user behaviors . Only after a clear understanding can you adopt targeted and reasonable activation strategies.

Therefore, before new users enter the product, we need to first do a good job of data collection, establish a basic user behavior data monitoring mechanism, label users, and do a good job of user grouping. In order to better restore user characteristics and increase the possibility of conversion, we usually label users, such as risk preference, purchasing power, consumption preference, social preference, etc. The first- and second-level labels can usually be designed to reach dozens or hundreds of types. Evaluating users from multiple dimensions and angles is also a refined thinking for operating users in multiple scenarios. Many times these labels come from users’ own attributes, but more often they come from insights into behavioral data.

Each user has a label feature, and the next step is to operate the users in different groups. In order to ensure that the operation strategy is not disrupted or to avoid repeated contact with users, we usually look for a main line to design the operation strategy, such as: based on the user life cycle . How to promote activation based on different user life cycle users? Usually we have a graphic method:

Taking e-commerce products as an example, if we find that if a user places an order again within 7 days after completing his or her first purchase on the platform, the probability of the user becoming an active user of the platform in the next six months will increase significantly. Based on this, we can launch such an activity to ensure that users become active again within 7 days: "Any order purchased again within 7 days after completing the first order payment can enjoy the "free shipping/free shipping" benefit once.

3. Activation strategy based on “user life cycle”

(1) For users in the potential user stage , they usually occasionally enter the product with an "exploratory mentality". At this time, it is necessary to establish a user attraction mechanism to attract these users to convert into active users. For example: setting up gift certificates for novices (material incentives), such as giving cash red envelopes when registering for mutual finance products; setting entry thresholds (spiritual incentives), and only those who meet the conditions can obtain access qualifications; establishing an invitation mechanism (functional incentives), such as the invitation incentive mechanism adopted by social products when acquiring target users. You can get points as rewards for each friend you invite, and the number of points is related to the individual's operational permissions.

(2) For users in the mature user stage , they already have some understanding of the product content and functions, and have relatively stable user relationships and usage paths within the product. From a certain perspective, they are also high-quality users. At this time, it is necessary to design and improve the incentive system from the user's perspective to extend the user's active cycle. Material incentives are no longer what they value. They value more the spiritual rewards from the product. Therefore, user levels/points privileges can be bound to "user stay time" and "number of opens". Users with high points can redeem certain privileges and can view high-quality content/enjoy unique identity tags. In addition, based on user behavior data, "intimate care" is provided to the target user group, and even "one-to-one" precise operations are carried out to make users feel valued, but the information reached between groups must be differentiated, which can further motivate users to share.

(3) For users in the silent user period , this part of users is in the process of understanding some or all of the functions of the product but does not use it frequently. This part of users needs greater efforts to awaken and care for them in order to delay their silent period. For example: conduct a needs survey for this group of users, develop new functions as soon as possible to meet the needs of this group of users, and then activate them through marketing means.

4. Improve retention rate : find the cause of loss through data analysis

There are many reasons for user churn . The difficulty of investigating churned users lies in getting users to participate in the investigation. There was an e-commerce company that invested heavily in subsidies when it was first launched, but after the subsidies were cancelled, user churn was very serious. Taking this incident as an example, I would like to share some ideas on how to find the reasons for churn through existing users:

First, through the user grouping function, we filter out users who have made purchases, and then divide these users into two groups: one is users who have made purchases and stayed, and the other is users who have made purchases but churned. Based on user behaviors, such as successful payment, the data can be further refined and divided into two groups for comparison.

After separating the two groups, first check the user portraits to find the differences. Through observation, it is found that the number of times the two groups of users trigger the category navigation is different. At this time, the two user groups can be compared to determine the activity of triggering the category navigation events.

As shown in the figure, the blue columns are retained users, and the yellow columns are lost users. The blue column is much higher than the yellow column. At this point, we can put forward a hypothesis: users who have triggered the category navigation have a higher retention rate. Then, we need to verify the hypothesis next.

During verification, users were also divided into two groups to see if there were any differences in the category navigation events.

The first group includes users who have triggered the category navigation most times within three days;

The second group includes users who have triggered category navigation less frequently within three days.

Through funnel analysis, we found that the blue columns represent users who trigger the category navigation more times, and the yellow columns represent users who trigger it less often. Obviously, the conversion rate of blue users is much higher than that of yellow users.

Looking at retention, the blue curve represents users with more triggering times, while the yellow curve represents users with fewer triggering times. The blue users are nearly 20% more than the yellow users. This verifies the previous hypothesis: users who have triggered category navigation do have a higher retention rate than users who have not triggered category navigation.

From a product perspective, the role of category navigation itself is a form of content organization. It is a way to improve the search efficiency of individual users, bring a more comfortable experience to users, and allow users to find their favorite products faster and complete relative conversions. Based on the analysis results, the product was revised:

Before the revision, the category navigation was in a small corner at the top. After the revision, the category navigation was adjusted to the center of the page and divided into eight major entrances. After the revision, the retention rate has been greatly improved.

Through this case, we can see the data analysis method for analyzing the characteristics of retained users. The reasons for activity are different, but the product needs to find the core value for users, reduce churn through core functions, and increase activity.

5. Induce users to pay: guide users to repurchase and maintain user vitality

For products such as e-commerce, online courses, and Internet finance, when users download and register for products, they have certain demand expectations. On the basis of good product and process conversion, it is not difficult to satisfy users' expectations for the product and obtain their first purchase. However, users often leave without paying. You can use funnel analysis to look at the key links of user loss to find the reasons, or you can select a certain number of users to observe their behavioral characteristics and find out the reasons for payment interruption.

In addition to payment guidance, repurchase rate is also the key to product operation. Users who have made purchases represent their recognition of the product. Even if the purchase is induced by coupons, the platform has also provided services, which reduces the sense of unfamiliarity and makes it easier for old users to repurchase than to convert new users.

If it is based on a single transaction, no matter how high the average order value is, it will be very difficult to achieve growth in the long run. Therefore, more and more companies are exploring new user operation models. There are three ideas for trying to increase the repurchase rate:

1. Member privileges give users reasons to purchase for a long time

The membership card model has been used to the point of no innovation in traditional offline stores. However, it is indeed an effective means of gaining consumer trust in exchange for future cash flow. In the past year, more and more e-commerce platforms have begun to move this strategy online with the same purpose: to bind users. When seeing "Become a Black Card member, you can save 200+ on this order and all orders throughout the year will be free shipping", I believe many JMS will be tempted. After all, if you become a member, you will not consider the issue of group buying when buying anything. Even if I just want to buy a pack of makeup remover cotton, I will place an order directly.

At present, including Amazon 's Primeair , JD.com 's JD Plus, Xiaohongshu 's Black Card, etc. are all aimed at forming users' physiological dependence on merchants. The VIP88 card launched by Alibaba covers a variety of scenarios such as takeout, shopping, film and television.

2. Insight into crowd needs and design activities

If membership is directly bundled without any care, then crowd-based operations may require some thought. That is to say, insight into user needs. For example, through behavioral data, users who have viewed, added, and collected shoes on the list page, search bar, favorites page, and shopping cart in the past three days must have a need to buy shoes. If they bought a haircut package 20 days ago, they should have a need for a haircut again recently. If they only view but do not buy or have not bought yet, they must be missing a certain opportunity. A user care or a targeted and timed coupon may be able to improve conversion.

You can target users’ own attributes, as these users all have some group characteristics. For example, you can offer a half-price strategy to students and try to keep them hooked, because there will always be a time when they are not students. You can target pregnant mothers and send them various coupons that are valid for half a year or a year, so they will most likely continue to use the products after the baby is born, and so on.

3. Hardware entry

With the development of technology, people's attention is focused on and things to play with are increasing. Mobile phones and computers are not the only ones. Many e-commerce and content platforms have extended their carriers, such as Amazon's Kindle, Himalaya 's Xiaoya, Fire Tv, Fire phone, Echo, etc. As soon as the smart speaker Echo appeared, it is no exaggeration to say that Amazon has basically achieved its goal. As long as you ask the smart speaker to buy me a carton of milk or a tube of toothpaste, it will reply: "Is it the same brand as last time?" All purchasing behaviors are completed. This is the hardware, which helps users pre-embed the "ordering program" and thus generate continuous repurchase behavior.

6. Self-propagation: Improve K value and find the incentives for forwarding

In the process of user self-propagation, there is an indicator that cannot be ignored: the K value. K-factor virality metric = (number of invitations each user sends to his friends) × (conversion rate of people receiving invitations to new users)

Assuming that on average each user will invite 20 friends, and the average conversion rate is 10%, K = 20*10% = 2. When K>1, the user base will grow like a snowball; when K<1, the user base will stop growing through self-propagation when it reaches a certain size.

In order to induce users to self-propagate, that is, fission activity, we need to find a way to increase the K value. Obviously, the K value is related to whether the user is willing to forward, so what content are users willing to forward?

1. Share useful and relevant content

The content here refers to the specific content generated by the activity, such as a user participating in a baking experience and being willing to share his or her DIY work experience on WeChat Moments.

(1) Establishing a “personality”

The word "personality" has been very popular recently, especially with the collapse of personalities within a certain circle, but in fact, each of us has our own personality. The so-called personality is actually "how others see me", and how others see me is gradually built up by what I say, do, express, and share. How can I make the way others see me be the same as my self-expectation, or even better than the real me? Then I need to control the quality of the content I output.

Each of us has n label-like identity symbols, such as fitness expert, beauty weed puller, skin care guinea pig, store explorer and foodie, China's good travel buddy... and the content contained in our external information is also exuding such symbols or reinforcing them subtly. People have different personalities, but most people want their personalities to be positive (for example, we all want to be seen as beautiful and positive by others). If your content can help users enhance their personalities, they will naturally be willing to share.

Providing materials will link your content with the identity symbols that users want. The simplest and most direct way to connect with users is to share materials, such as photos, gift boxes, etc. Providing users with high-quality sharing materials is the first step. Of course, this material still needs to be further segmented according to the population. How different groups of people want to be presented to the outside world is linked to the early user research.

(2) Build scenarios to help users show off their skills

None of us wants to be a self-promoting character. For example, build a stylish scene to help the user express what she has not said through the environment and aspects.

(3) Seize the opportunity and provide users with forwarding materials

The desire to share is particularly time-sensitive. It will gradually weaken over time. Providing users with sharing materials at the peak of their expression immediately after the event ends will have a very high success rate.

(4) Share topics

Topics of conversation can be roughly extended in three directions: rare, popular, and subversive to existing cognition . For example:

Rare : attending a meet-and-greet with writer X that is only open to the public once a year (the value of this rarity needs to be universally recognized, not unilaterally emphasized)

Popular : "The most popular nowadays" "Story of Yanxi Palace" - fully integrate the hot spots, and don't hope to be left behind by the times;

Subvert existing cognition : a good-looking and delicious fondant cake (users often think that fondant cakes are not delicious), help users prove their uniqueness (personal uniqueness, unique experience), and easily promote sharing.

2. Share your views and speak for users

Helping users express themselves is the core of sharing opinions. There are two types of opinion dissemination:

(1) I can’t explain it well, please help me:

This is an era where everyone is a joke teller; and users are keen on forwarding and sharing mostly because the poster’s expression is consistent with their own views, but their personal expression ability is limited. If you happen to have some finished content, why not forward it?

(2) I can’t explain it, someone help me explain it:

This is particularly prominent for those born in the 1980s and 1990s. There are actually only a limited number of people who can truly express their emotions. Many people not only do not know how to express themselves, but also dare not express themselves.

3. Incentivize sharing and drive forwarding with value

In addition to the emotional and spiritual motivations mentioned above, the driving force of interests is also indispensable to drive users to voluntarily share. The motivation for sharing may be different for different users in different situations. The benefit-driven approach can be self-serving, altruistic, or both to encourage users to actively share. Common forms include cash and vouchers, of course, benefits can also be presented in other forms, such as free goods, movie tickets, etc.

Summary : The above shares some ideas and experiences based on the five key links of the AARRR model. In fact, each link contains a lot of content. For example, the activation link can also use the push of the APP. For example, the conversion and payment link can also push benefits to users at key nodes through data analysis. For example, a user churn early warning mechanism can be established. Therefore, the AARRR model basically covers the operation mode of the entire life cycle of products and users. A complete AARRR working method can guide the work of oneself or even a team.