What is the viral user growth model? How to build a user growth model?

The concept of growth hacking is very popular now, and many companies are recruiting talents in the field of user growth . So how to achieve growth and how to build a user growth model? In response to the problem of user growth, Zhuge Jun drew a chart to tell you what the viral growth model is, and told you the problem of user growth, how to draw this chart, and taught you to build your own viral user growth model.

1. What is the viral user growth model?

First of all, do you know what a user growth model is? Simply put, the user growth and user loss curves brought by each channel are drawn on a graph to reflect the changes in the number of users.

So what is the viral user growth model? It is to add a viral growth curve to the user growth model to affect the overall user growth.

Don’t understand, right? I drew a picture, which is the following:

The blue, red and green curves represent new users from different channels, the light blue curve represents user loss, the horizontal axis is different months, and the vertical axis is the user level. If there is no purple curve, it is a user growth model, but with the purple curve, it is a viral growth model. The purple curve represents the number of virally growing users. The viral curve is a growth curve that relies on user dissemination, and other new curves are growth curves that rely on operational behavior.

This chart reflects the changes in new users and lost users over a period of time. The channels for new users are divided into stable new channels, such as application market and website diversion, and active new channels, like the unstable blue curve. For example, we have conducted a ground promotion activity or issued a press release, which cannot be used as a regular channel operation method.

The viral curve refers to the new users brought in by user self-propagation , not the new users brought in through the operator's promotional activities, so it is called the viral curve. In the viral user growth model, we pay more attention to the new users brought by user self-propagation.

2. How to design a virus growth curve?

The increase in users brought by operational behavior is the basic curve, while the viral curve has more factors to consider. There is a key word in the viral growth curve: K value, which is the viral coefficient.

Let’s take a look at how to design a viral growth curve through a minimalist viral model:

Viral coefficient K = (number of shares/number of sharers) * (number of successful conversions /number of shares) = number of successful conversions/number of sharers.

In simple terms, the k value is the number of new users that each user can bring.

The description of the k factor in the growth hacker book is k factor = infection rate * conversion rate . The infection rate refers to the extent to which users spread the product to other people, and the conversion rate refers to the number of infected users converted into new users.

The standard formula is k=i conv, and the specific formula is k=sharing rate conversion rate.

For example, in an event, 1,000 users invited other users to participate, and 1,000 invitations were issued, so i=1. 200 users were converted on the first day, so conv is 0.2, and the final K value is 0.2. Assuming that the K value is fixed, the 200 new users on the second day will convert 40 users, 8 users on the third day, and 1.6 users on the fourth day, which means we get a total of 1,250 users.

This excludes the situation where new users do not forward or delay forwarding. So what will be the effect if the K value is 0.5? Then 500, 250, 125, 62.5, and 31.25 users will be converted successively. What if the K value is 1? The overall number of users will continue to grow at a rate of 1,000.

To increase the k value, you can increase the number of shares for a single user, or you can increase the number of people who convert shares.

From this, we have derived a viral growth curve. The higher the K value, the more users participate in the fission, and the steeper the growth curve will be. We can see that only when the K value is equal to 1, the viral curve will continue to grow, otherwise it will gradually return to 0. However, in reality, we also have to consider the situation where the converted users do not forward, and at the same time, a user may also invite multiple users to participate, so we need to make the K value greater than 1.

For the relatively successful NetEase Red Poster fission activity, someone calculated that its K value was 6.85, which was required to achieve the screen-sweeping effect. Therefore, in the viral growth curve, we must increase the K value as much as possible. The K value reflects the attractiveness of the activity, including prizes, platforms, processes , copywriting, posters, characteristics of the initial user group, characteristics of the fission user group and other factors. Therefore, truly successful viral growth is rare.

3. How to use the viral user growth model?

The application of viral growth models is more reflected in the planning stage of user growth goals. For example, if we want to achieve a net increase of 100,000 users in 6 months (so 100,000 retained users), the viral curve can help us make a plan.

First, list the common user growth channels. We will bring in regular user growth through channels such as e-marketplaces. Regular channels will bring us a stable amount of new users. Suppose we can bring in 30,000 new users in total through five e-marketplace channels in six months. This estimate is based on observations of conversion data from conventional promotion channels .

Next, list the activities by which you can operate to promote growth. We can summarize the user growth effects through historical operational activities. For example, every 10 yuan spent on advertising in the electronic market can convert one download. Based on our budget, assuming we invest 100,000 yuan, we can convert 10,000 new users. For example, some news publicity can bring about user growth, so we find information channels to release press releases. Suppose we invest 100,000 yuan and convert 10,000 new users. The delivery plan is made based on the conversion effect and budget of the channel promotion method, which is a curve that actively promotes growth.

Third, consider the situation of user churn . After developing a plan for adding new users, we also need to consider user churn. Based on historical operational churn data, we assume that the user churn rate is 20% (I know it’s unreasonable, but if it’s any higher we won’t be able to achieve our goal). Then the above channels will add 50,000 new users in total, lose 10,000 users, and ultimately retain 40,000 users.

Fourth, design a viral growth curve. With 50,000 new users and 10,000 lost, there are still 60,000 short of the 100,000 goal. These 60,000 users need to be achieved through the viral growth curve, and the most important factor in growth is the number of people sharing.

Calculate the number of shares based on our goal, according to the formula:

K = number of conversions/number of shares

For example, if we have 40,000 users and our goal is 100,000 users, we need to add 60,000 new users. The K value is 1.5, which is the number of new users brought by each user. But in actual use, we do not have 40,000 users at the beginning, and not every user will share. Instead, we need to infer how many users are needed to share based on the k value reflected in the historical operation data, and then operate with the goal of increasing sharing.

Summary: Everyone can build their own viral user growth model to complete KPI. The purpose of establishing a model is to continuously correct their own operational behavior. For example, according to the planned channel promotion, 20,000 users should be brought in. What should you do if this is not achieved? Then it is necessary to revise the indicators based on actual data, further identify problems, increase investment or optimize activities.

Source: Zhuge io Data Coach