How to increase and retain APP users?

How can APP do a good job in user operation? Can user portraits better help user growth and retention? What are the specific methods at different stages? This article analyzes and summarizes these issues. Let’s take a look~

Today, the cost of an APP acquiring target users and converting them into users of its own products has become extremely expensive. In such a fiercely competitive environment, leveraging data intelligence to manage new users has become a recognized trend in the industry.

However, how to use data intelligence technology to maximize the growth effect is a science. Today, let’s talk about how to solve the “increase” and “growth” of new APP users through user portraits?

The growth of new users includes three stages: the new user acquisition stage, the new user conversion stage, and the continuous maintenance stage. The role of user portraits in different stages is also different.

1. Customer acquisition stage: Acquire high-value users at the lowest cost

In the customer acquisition stage, APP mainly relies on precise delivery. The precision here has two meanings: one is the precise reach of users, and the other meaning is how to acquire the most high-value users at the lowest cost.

In terms of specific implementation, target users, channels, and content are all indispensable, and user portraits can focus the energy of the three and maximize their effectiveness.

Three steps to precise APP delivery:

1. Select high-value target user groups;
2. Target channels that match your target users;
3. Choose creative materials that match the characteristics of the target users;

Before precise delivery, the APP needs to identify its target users and understand the characteristics of the target user group.

APP can use the user portrait tool to conduct data analysis and big data modeling on the user groups that are most active, most interactive, and have consumption habits on their APP, convert abstract user definitions into data labels and features, and set clear goals for customer acquisition strategies.

In terms of channel screening, the APP can not only use the target user portrait for channel matching, but also retrospectively verify the previous delivery data, and ultimately select the media from multiple aspects such as quantity, TA concentration, price, and integrity.

In terms of content creativity, APP can group the target audience creatively through portraits and gain the favor of high-value users through customized content, rather than blindly pursuing low-cost users.

For example, information apps like to use curious materials as their content, because such materials can easily generate hits and generate 100,000+ views, and can easily attract users at a very low cost. However, some news apps that have attitude, depth and insight will not use curious materials, because such contents cannot bring them high-value users who have requirements for content depth and are willing to pay for good content.

At present, major companies have begun to pay attention to their own portrait systems and have derived more complex growth models based on them, and the effects are also very significant.

A well-known news APP uses model algorithms to select channels with a high concentration of high-value users and chooses in-depth reporting content or exclusive news to acquire target users. According to statistics, this growth strategy increased the contribution of the APP's overall DAU by more than 10%, achieving the goal of acquiring more effective users with the same amount of money.

2. Conversion stage: Do a good job of cold start to retain new users

Users' preference for using APPs has a "first impression effect", just like when people order takeout. If a restaurant has good taste, exquisite packaging, and excellent service the first time they order takeout, then the probability of ordering takeout from this restaurant in the future will be higher.

Therefore, how to leave a good impression on new users when they first come to the APP and make them like your product is the main battlefield for competition between APP operators.

First of all, to understand every new user who enters the APP, the APP needs to understand the new user portrait as accurately as possible.

• In the first step, the APP guides new users through the opening screen, allowing them to fill in or check basic information, interests and hobbies, and other characteristics. However, this information cannot represent the user. First, it is not objective and accurate enough. Second, the amount of information is far from enough to generate a user portrait.
• In the second step, the APP can leverage the capabilities of third-party data companies to quickly complete a new user portrait. The data dimensions of third-party data companies are more comprehensive, and they have a deeper understanding of users' online behavior preferences. They may be new users to the APP, but they may be very familiar users in the third-party database.
• In the third step, the APP can trace back the customer acquisition channels and the characteristics of the delivered materials, and combine these characteristics to make the new user portrait as complete as possible.

After completing these three steps, the APP can fully understand the interests and preferences of new users and recommend content, services or products that they need and like.

Secondly, after understanding the user portrait, the APP can also use algorithms to label the characteristics of new users, associate them with people with similar characteristics in the existing user group, and recommend content to new users through synergy.

For example, if a new user has label information such as Beijing, male, and Huawei user, he or she can collaborate with the same type of male users among old users and recommend content that the old users are interested in to the new user.

During cold start, the APP will also use third-party user tags to make supplements, but how to integrate the two sets of tag systems?

Through practice, we have come up with a good solution, which is to use the topic model algorithm to use third-party tags to group new users into groups, and then carry out group operations.

When a certain information APP cooperates with a third party, it groups new users and makes personalized recommendations based on the calculation results. For example, a new user group has a 60% chance of being fashion and entertainment experts, a 25% chance of being mothers and babies, and a 15% chance of being online shoppers. In this case, you can recommend fashion and entertainment content to this group, followed by mothers and babies content and consumer content, and pay attention to their reading habits and make dynamic adjustments.

At the same time, the clustering characteristics of this group of people can be coordinated with the existing user data to recommend the content with the best clicks among similar users. Ultimately, the app’s new user retention rate on the next day increased by 18%.

Finally, when it comes to content recommendation, the APP cannot rely solely on data; manual operations are also very important. Manual operations can provide users with emotional and warm operation measures.

3. Live stage: Continuous user portrait insights to find the best solution to user needs

The cold start of new users is an ongoing process, and it needs to change with the situation and the times in order to truly attract and retain users.

During the cold start process, the APP also needs to use various operational methods and product designs to keep users returning to the product frequently. For example, in operation, setting a corner mark to indicate unread content can stimulate the user's obsessive-compulsive disorder and prompt them to click to activate.

In terms of product functions, the APP needs to design user incentives such as novice guidance, points mall, and incentive activities to allow users to gain enough sense of accomplishment.

During the user cold start process, the iteration of user portraits needs to be continuously promoted. The APP can continuously conduct portrait insights on users with increasing activity, coordinate new user data with existing user data, and thus better understand users, continuously make growth hypotheses and growth strategies, and continue to explore users' interests through operating content, products, and activities, and ultimately find the optimal solution to meet user needs.

The increase and growth of users is the "spring growth" stage of the entire user life cycle. Only by doing a good job of user operation at this stage can we lay the foundation for the subsequent "harvest".

Therefore, the role of data intelligence operation tools at this stage is increasingly valued by APPs. How to better understand users, capture their hearts, and retain them through user portraits has also become an essential skill for APP operations.

Author: Personal recommendation

Source: Personal recommendation