How to improve product user retention?

Rethinking retention

Retention is an often mentioned indicator. We often hear people say that the retention of a certain product is very good, reaching x, or that the retention of a certain product is less than 30%, and many other such descriptions. In fact, strictly speaking, these descriptions are vague, inaccurate, and even problematic.

Based on the calculation method of retention, we can actually give a calculation definition for retention:

The retention rate is the proportion of a group of users who repeat certain behaviors within a certain time interval.

As we can see from the definition, when we talk about the retention of a product, we must first clarify three factors:

1. Which group of users? (Is it a new user, an active user, a registered user, or an unregistered device)
2. What is the time interval? (Is it five minutes, a day, a week, or a year?)
3. Which behaviors? (whether the user is active in the background or foreground, whether the user has used the main functions, or whether the user has stayed there for more than 1 minute)

Therefore, when we say that the front-end activity of new users of a product on the next day is 50%. What we are saying is that 50% of the newly added active front-end users on a certain day will still be active on the next day.

The retention methods commonly used in our daily work are generally:

In addition to the common retention rates mentioned above, there are also some data for calculating weekly retention and monthly retention. In essence, the only difference is that the time interval for calculation has changed. Retention mainly measures the user stickiness of a product, or the proportion of users who continue to use the product. In addition to retention, there are also some other dimensions of data such as average monthly active days, DAU/MAU, etc. used to measure product stickiness, which will not be elaborated here.

Why do users stay?

Before discussing how to improve user retention , we must first understand why users stay after using a product?

Many people have given some models, for example: Nir Eyal analyzed the user addiction model in "hooked", from trigger (Trigger), action (Action), variable rewards (Variable Rewards) and investment (Investment), four factors constitute a cycle that makes users addicted. There is also the BJ fogg behavior model, which also tells us to a certain extent how to change and guide user behavior so that users can retain better.

After reading these "models", I found them a bit complicated. Combining them with my previous retention practices, I found that there is a principle that is the essential constraint of retention.

The services provided by the product need to exceed user expectations.

The first constraint is user expectations. All new users will have certain expectations after downloading the product and opening it.

All of our channels will more or less convey to users what our products do and what problems they can solve. If users with expectations do not quickly feel that their expectations are met after using the product for the first time, they will most likely abandon it. Only by meeting or even exceeding users' expectations can users be retained well.

We once developed a live streaming app overseas. When we first started in Vietnam, we opened the report one morning and found that there were a lot more new users, several times the usual number. However, our colleagues in the marketing did not pay for promotion, which means that our natural new users suddenly skyrocketed.

We were very happy at the beginning, but the second-day retention and subsequent retention were terrible. Later, by monitoring social media, I found out that some local media in Vietnam had reported it. The main content of the report was that we were involved in pornography, there were many pornographic live broadcasts on the terminal, and a collection of live broadcasts of pornographic videos that were borderline appeared on YouTube.

It is precisely because of these exaggerated reports that many users have an expectation of our platform: "There are sexy anchors and pornographic content here", but when they actually download and open it, they can hardly find such content. This means that users’ expectations are not clearly met, so they naturally won’t stay.

There is also a recently popular app called "FaceApp", which topped the App Store charts in all countries around the world at the same time. However, I guess that their retention in many countries must not be very good, because in their user feedback, many complained that after downloading, they could not open it, or could not access the server to download specific face-changing effects. It is obvious that the sudden explosion of new users has caused this small team in Russia to be unable to meet the simultaneous requests of so many users at the same time. This will inevitably lead to users with extremely high expectations being unable to be satisfied, and the number of uninstalls will also soar.

As discussed earlier, the essence of retention is to exceed user expectations. However, users’ expectations actually change during the use process. Expectations may become higher and higher, or even change completely. For example, when we first use a music app, our expectation is to find the songs we want to listen to. Later, the expectation may become that the app automatically recommends songs I like.

For example: When we first used Momo, our expectation might be to get to know some people nearby, but later, some people's expectation changed to how to quickly complete a "date". (PM is working hard to satisfy everyone)

Therefore, when it comes to meeting user expectations, different stages are different. This stage is generally divided into short-term, medium-term and long-term, which corresponds to short-term retention (1-3 days), medium-term retention (3-30), and long-term retention (more than 30 days). In these three stages, there are different emphases on improving retention. Let’s start with short-term retention.

How to improve short-term retention?

We can sort out the process of retaining new users as shown in the following figure:

User expectations satisfaction process

It can be seen from the figure that user retention goes through three main processes, and all our methods to improve user short-term retention are mainly concentrated in these three processes. The specific focus of these three processes is analyzed below.

Process 1: Forming reasonable expectations of users

1) Define what reasonable expectations are?

From the product's perspective, users' reasonable expectations are actually the problems that the product solves for users and the needs it meets. Therefore, before conveying reasonable demands to users, it is necessary to clearly define what reasonable expectations the product can provide? That is, it meets the needs of certain users in certain scenarios.

There are many methods for demand analysis, which I will not elaborate on here. I will only talk about a few more practical ones.

1. Interview active users and ask them why and what motivates them to use the product. A more effective question is to ask them what words they would use to recommend the product to others. For more information about the interview, please refer to the previous article "How to Conduct User Interviews".
2. Break down the main behavioral paths within the product, look at the user share of the main paths and the corresponding retention rates, and from this you can determine some of the most useful points for users.

2) Communicate expectations to users

After having reasonable expectations, the next step is how to accurately convey these expectations to users.

The most common mistake in this step is that there is poor information or communication between students who are responsible for obtaining users and students who are responsible for product operations. When I was working on a short video community, my colleagues in charge of promotion used ASO to acquire more new users and gained a lot of keyword volume. It is true that the number of new users has increased a lot, but the retention rate is very unsatisfactory. It is found that many of the keywords used are completely inconsistent with our product types or the expectations of users.

Therefore, in order to convey expectations to users, colleagues from promotion to product and operation need to discuss together the product features that best meet product expectations, and based on this, create advertising materials or optimize keywords.

Common areas that need attention include:

• Do the advertising materials and ASO/SEO keywords match the product features?
• Are the store descriptions and screenshots exaggerated?
• Is the product name/logo ambiguous?
• Do the materials disseminated by users themselves (copywriting, slogans, etc. shared on social media) conform to the product features?

Step 2: Meeting expectations

The two most important things in the process of meeting expectations are:

1) Carrying user expectations across different channels.

Users from different channels are very likely to have different user expectations. For example, for content-based community products, some users are attracted by funny materials, while others are attracted by curious information. When users enter the product, they should be treated by showing content that corresponds to their expectations.

What can be used here is mainly deeplink. Users with different materials can tell the client through deeplink, and then provide matching content.

2) Let users perceive core functions as quickly as possible and meet their expectations.

When a new user enters a product category, we need to make him aware of the features that are most likely to meet his expectations within a very short period of time.

A frequently used method is the funnel analysis of the core path - breaking down the intermediate process from when a new user enters the product to when they use the core functions, and then counting all the funnel data; solving problems from the funnel, deleting unnecessary steps and optimizing conversions of necessary steps.

"musical.ly" creatively removes the video feed, allowing users to directly face the content after opening it and experience the video content more quickly. This is a typical example of deleting unnecessary steps. There are also many live broadcast apps that directly lead new users to a high-quality live broadcast room, which is also to allow users to skip the screening step.

In terms of optimizing the necessary steps for conversion, a commonly used method is onboarding. For many games, after a new user enters for the first time, the onboarding will take you through the game quickly, allowing you to understand the core gameplay of the game as quickly as possible. There are also some voice chat room applications that have relatively complex gameplay, so they have many rooms where old users can guide new users. After new users enter, old users will guide them step by step to integrate into the entire application.

Process 3: Achieving/Exceeding Expectations

After going through the first two steps, users basically already know the features of the product, which may match their expectations. At this time, users need to experience the core value points repeatedly. Or give users another experience that exceeds their expectations.

For example: When it comes to funny community products, after a user has completed their first experience with funny content, more funny content should be recommended to him. After he leaves the product, we push new users back with appropriate frequency to allow him to experience as much funny content as possible.

Also, for funny community products, in addition to the expectation that the content itself is good, if users can post their own content in the community and get enough attention and discussion from other users, this will be an experience that exceeds expectations. Therefore, timely encouraging users to actively post and comment will bring them an experience beyond their expectations. And thus stay in the product longer.

Short-term retention of new users of major product forms in the industry

Here is a brief list of the more successful short-term (next-day) retention rates of new users for major product forms. There may be errors and this is for reference only.

postscript

This article mainly summarizes the definition of retention and how to improve short-term retention.

Related reading:

1. User growth step 1: How to retain users?

2. Product operation: new user retention and conversion strategy!

3. 3 ways for APP operation to improve user retention!

4. How can product operations increase the number of new users and retain them?

5. User operation: how to retain users after fission?

6. User operation: How to improve user retention/repurchase rate?

Author: Nancun Xiaofu

Source: Nancun Xiaofu