User growth and basic process!

However, today's article wants to do one thing: to be as close as possible to the majority of people and scenarios engaged in so-called growth work, and to tell you simply and clearly: what is user growth and how to do user growth.

01What is user growth?

Years of training in science and engineering thinking have allowed me to develop a habit: starting from concepts and looking for the most original meaning. So, we must know what exactly user growth is.

User growth can be understood from two aspects: one is "user" and the other is "growth".

First, what is a user. Most people understand it as “people who use the product”, which is actually a relatively perfect answer. We can clearly understand the logic behind it: the product is the premise, the user is the result, and there is no user without a product.

So, what kind of users are there? This means there are many latitudes for division. For example, according to the user life cycle, there are new users, active users, silent users, and lost users; according to the product usage process, there are follow users, registered users, paying users, repeat users, refunded users, etc.

The user attributes you target are different, and the user growth work you do is different.

Now let’s look at what growth is. This is actually a simpler concept, which is an increase in quantity. However, what does this growth state or growth curve look like? Is it a straight line or a wave? Is it a constant slope growth or exponential growth? It is important to know that growth curves of different shapes often reflect the effectiveness and problems of growth strategies.

So, returning to the concept of "user growth", we can clearly sort it out: through products, operations, channels and other means, drive the number of a certain type of users to achieve a certain state of growth curve. This curve contains expected values ​​such as quantity targets and speed targets.

Therefore, user growth is not simply about attracting new users, activating users, retaining users, converting users, and splitting users, but about ensuring the quantity targets and curve requirements. This is actually something that many growth books or articles do not mention. That’s why we need to make it clear.

02Basic process of user growth

By understanding the essential meaning of user growth, we can better practice and complete growth work, which is inseparable from the basic user growth workflow. It can be said that this is the most important thing for us to understand user growth.

In fact, compared to the growth process mentioned in many books or articles, real user growth work only requires three steps: determine the growth formula -> MVP testing -> scale promotion.

Let’s interpret them one by one.

1. Determine the growth formula

The so-called growth formula is the relationship between core growth indicators and basic growth factors. This relationship will present a data model, which is the main basis for user growth.

For example: Total number of paying users = number of old paying users ✖️repurchase rate + new channel traffic ✖️conversion rate + number of old paying users ✖️recommendation rate. Among them, the total number of paying users is the core growth indicator, also known as the North Star indicator, while the number of old paying users, repurchase rate, new traffic channels, conversion rate, recommendation rate, etc. are the detailed data dimensions that affect these growth indicators.

However, the above is still a relatively macro growth formula. When it comes to specific user growth, we will focus on a certain growth method or a certain operation stage, so the growth formula will be more microscopic and the data model will be more detailed.

For example, if the growth method you use is new media placement, then the growth formula may be: Number of low-priced product users = channel coverage ✖️ conversion rate = number of public account fans ✖️ open rate ✖️ group entry rate ✖️ conversion rate. Then, based on this formula, further split the data indicators that affect the basic factors, such as group entry, attendance rate, etc., to form a more detailed table:

Once you have a data model like this, you can design growth strategies, paths, content, etc. and move on to the next stage of practice - MVP testing.

2. MVP Testing

MVP testing is an essential method for Internet product development, and it is also applicable to user growth. Because both products and models need to go through high-frequency, small-scale demand verification, feedback, and optimization to achieve the desired evaluation data indicators. The data table extended from the growth formula above is actually used to evaluate whether the user growth model is reasonable and the optimization direction.

So, how to conduct MVP testing? Personally, I think two points need to be achieved.

First, you need to select a suitable group of user channels for testing. Note, it’s a batch. This means you need to choose a channel and segment your user base for your subsequent multiple tests. For example, for social channels, you may need to prepare 100 groups and test 20 groups each time, for a total of 5 tests.

Second, data must be recorded for each test to facilitate timely comparison and adjustment and optimization. The focus here is on comparative optimization, which is also the core work of MVP testing. Simply promoting and recording data is meaningless. You need to extract useful information from the data to guide how to improve your model.

For example, the selection of carrier sequence in the fission process can be solved by MVP testing.

If you want to know whether the churn rate of "WeChat group-personal account-public account" is lower or "WeChat group-public account-personal account" is lower, you can select two channels to promote fission activities, and then compare the data models of the two channels. If you find that the latter has a high retention rate, you can choose the latter as the regular path for fission.

In fact, most of the work of user growth is to conduct various tests, especially tests of many tiny details. Only with the increase of such tests can a growth model become a real growth method.

3. Scaled promotion

Once a growth model is successfully verified, it can be used to advance the business and undertake certain business goals. Of course, it also means that more resources can be used to achieve the goals.

The most obvious is channel resources. Compared with the cautious use of channels in the MVP stage, there must be sufficient channels for promotion in the scale-up stage to allow more users to participate in the new growth model. This can not only expose and solve more problems, but also discover many hidden highlights.

In fact, scaling is the final stage of verifying a growth model, because early tests are subject to fewer uncontrollable factors. However, with the participation of large-scale users, uncontrollable factors increase exponentially. For example, if your model requires frequent communication with users, it will easily lead to reduced efficiency if the volume is particularly large. Will this cause the scaling stage to be less effective than the MVP stage?

Therefore, once the data model is determined, it must be tested continuously during promotion. Even if it becomes a core business, it must be iterated, because growth is also another form of "product" that also has a life cycle.

03 Conclusion

This article briefly sorts out what user growth is and the basic process of user growth work. Although it is inconsistent with the definitions in some "textbooks", the core is the same. I believe that readers who have actually worked on growth can understand it.

Author: Wild Operation Community

Source: Wild Operation Community (dugu9bubai)