5 core steps to acquire app users

Many startups are interested in trying to discover the next tipping point through channel monitoring. However, before we test our user acquisition channels, we need to ask ourselves one most important question:

How much should you spend to acquire users?

No one wants to run meaningless ads on ineffective channels. Therefore, we all need some criteria for judgment: What kind of channel is of better quality? Is there any way to help us quickly understand users? What are some effective ways to acquire users? What tools and quantitative methods are worth learning from?

Step 0: Three basic questions

Before reaching conclusions on the above core issues, we need to clarify the following three concepts:

1. What is the total lifetime value of a paying customer?

The so-called lifetime value (LTV) is the total revenue that a user brings to your product from the first day they use the product to the last time they log in. It can be seen as a long-term accumulation of ARPU value. Traditional LTV calculations are relatively complex, so business intelligence startup RJ Metrics has developed a calculation tool specifically for this purpose.

But we can calculate the total value of a customer's lifetime with a simple formula:

LTV = (Average Monthly Spending of Paying Users – Gross Profit) / Churn Rate

Among them, gross profit means: the selling price of goods minus the basic cost of product sales, such as support and organizational expenses; monthly churn rate means: what proportion of users will leave, cancel their subscriptions, and stop paying, calculated on a monthly basis.

2. Proportion of paying users

Once the total lifetime value of paying customers is guaranteed to a certain extent, increasing the proportion of paying users will become an effective way to increase the company's receivables.

In the early stages of product promotion, many companies will attract customers by offering free services; then, through value-added services, they will convert some free users into paying users and achieve revenue growth. So, for us not every registered user is paying. Therefore, it is necessary to calculate and improve the proportion and conversion rate of paying users.

Proportion of paying users = Number of paying users / Number of registered users

3. Visitor registration success rate

Through a simple registration rate channel analysis, you can find out the registration conversion rate of users from each channel on the landing page (or the channel download conversion rate of the App):

Number of registered users/ Number of channel visitors

For marketing activities, the goal is usually to obtain some intermediate event data, such as test drive user information, sales lead data, etc. At this point, the calculation should be like this:

Number of leads/number of visitors or number of orders/number of leads

For e-commerce companies, since there are no free products, the calculation can be simple:

Transaction rate = number of orders / number of visitors

Now that we have calculated three values: user lifetime value, registration payment rate, and visitor registration rate, we can clearly know how much it costs to acquire a new user and how much it costs to acquire a paying user.

The next five steps will answer our initial question: What methods and tools can help us gain a deeper understanding of and acquire new users?

Step 1: Calculate the cost of user acquisition

In addition to CPS (cost per transaction) on e-commerce platforms, the other two core cost indicators are: CPA and CPC. CPA calculates the cost of getting a user to register, and CPC calculates the cost of getting a click to the site.

The calculation of CPA for companies that provide freemium products consists of two parts:

1. Cost of acquiring a paying user

CPPC: Cost of acquiring a single paying user

LTV: Customer Lifetime Value (Basic Question 1)

CPCC = LTV / 3

One-third of the customer lifetime value is the upper limit of your cost to acquire a paying customer.

This limitation does not apply when testing new channels. But when you find a decent channel, your optimization goal is to reduce the cost to one-fifth of the LTV.

2. How much should it cost to register a user?

CPA: Cost per user acquisition

CPPC: Cost of acquiring a single paying user

CR: paid conversion rate of registered users (Basic Question 2)

CPA = CPPC * CR

If a portion of your registrations are free or trial users, you need to take this additional factor into account when calculating how much you will earn per user registration.

Note: Some companies that provide enterprise services charge much higher fees than self-service services. In this case, the conversion rate from self-service to enterprise service needs to be taken into account. The cost formula for acquiring enterprise service will be:

Self-service CPA * Enterprise service conversion rate * (Enterprise service LTV/3)

Step 2: Calculate the cost of a single inbound flow

Usually advertising platforms will let you choose a maximum bid for a word click, rather than letting you pay for advertising based on the actual number of registrations you get. This is the value of the “visitor registration rate” in Basic Question 3. It will help you understand how much of these channel traffic can be converted into registered users.

CPC: Cost per click

CPA: Cost per user registration

CR: Registration rate (Basic question 3)

CPC = CPA * CR

This formula will tell you your cost per user acquisition. This way you can check the actual user acquisition cost at a glance and see whether you are making a profit or a loss.

Step 3: Optimize your channels

The above formula tells you how to spend your budget, and event tracking, analytics, and attribution tools will help you continuously optimize the user acquisition cost of the channel.

For example: the cost of bringing users through Baidu's SEM may be higher than through SNS promotion (Weibo, Zhihu, etc.). As a senior operator, the LTV value of registered users in each channel must be considered, which will determine the choice of channel.

With the above formula, you can easily evaluate each channel by just entering the relevant numbers. Please note that the above calculations require sufficient traffic as a basis to be statistically significant. If you find that the user LTV value of a particular platform is too low or the acquisition cost is too high, you can abandon that channel. Or if you find a cheap channel, you might as well give it a try.

Step 4: Related Tools

1. A/B Testing

A/B testing is essentially a split-group experiment. By creating a test page, this page may be different from the original page in terms of title, font, background color, wording, etc. Then these two pages are pushed to all browsing users at the same time in a random manner. Next, we count the user conversion rates of the two pages respectively to clearly understand the pros and cons of the two designs/creativity.

In the past, the technical and resource costs of A/B testing were relatively high, but now with the emergence of a series of visualization tools, such as Optimizely, Leanplum, Taplytics, etc., A/B testing has gradually become a common method for optimizing operational strategies.

2. Marketing Automation

That is, event-driven email and notification services. Tools such as Customer.io, Outbound, Vero and Kahuna can help developers optimize the user activation process and increase the purchase rate of registered users through personalized interactive content and real-time interactive channels.

3. Attribution Analysis

If you run a lot of campaigns, there are a lot of tricks to monitor conversion rates and spend. The best tool is the data tracking tool of mobile app. GA and domestic tools like Umeng, Zhugeio, Talkingdata, etc. can easily tell you which channel performs better.

Step 5: Integrate the platform

When the team size is limited, identify a convenient and easy-to-use platform tool to obtain the data required for all the above calculations, and then distribute this data to relevant analysis tools. The advantage of these data analysis platforms is that you only need to complete the embedded development once, and you can check the data required for operations at any time and switch to different statistical analysis methods at any time. Whether it is for optimizing internal team collaboration or improving the efficiency of operational strategies, this is the most time-saving way.