How to analyze user activity data?

In this article, I will analyze user activity data in depth from the following four aspects:

1. Definition of User Activity

2. Factors Affecting User Activity

3. Ways to increase user activity

4. The value of user activity

1. Definition of User Activity

• Active users: Users who open the app are considered active users, regardless of their usage. A device opened multiple times per day will be counted as an active user.
• Daily active users: The number of users who use the app every day. Each device is recorded only once.
• Monthly active users: The number of users who use the app each month. Each device is recorded only once.

Active users of different products:

• Products without an account: Users who open the APP within the period are active users. For example, tool apps.
• Products with accounts: Users who log in to the APP during the period are active users. For example, online games .

2. Factors Affecting User Activity

Product attribute determination

Products with different attributes have different usage frequencies, which directly affects the product's activity data. As an operator , you must be clearly aware of this. Under the same number of new users, for example, social apps like WeChat are used every few minutes, so their daily active data is often higher; cleaning apps are used once every 2 to 3 days, so their daily active data is slightly lower; taxi and navigation software are used once a week, so their daily active data is even lower; products like 12306, which are used once a year, have no daily active data at all. Given the attributes of different products, you’ll find that some products are more valuable in terms of weekly activity than daily activity.

I was previously responsible for the operations of two apps, one was a cleaning software and the other was an application market . Without any operational intervention, the daily activity of the cleaning software was far greater than that of the market. Through research, I found that users have the habit of cleaning up mobile phone junk every day, but the demand for app downloads is very weak. Over a long period of time, users clean up their phones more and more frequently, and their daily active users are on an upward trend. However, the number of times users download software is decreasing, and their daily active users are decreasing, showing a downward trend. Different products require different operational ideas. For example, the operation of the "Tiger Talks about Operations" public account focuses on spreading practical information , which is a one-sided output. The operation idea is completely different from that of APP.

Note: The above data is only for illustration and is not real data.

Product experience determines

For products with the same attributes, the daily active data that influences competing products is mainly based on product experience. Let me take games as an example. Both are MOBA games, Honor of Kings and King of Glory released by Tencent. The two games have the same resources and were launched at the same time. Due to the different product experiences, Honor of Kings has been revised again and again, and the product experience has been optimized more thoroughly, which ultimately made Honor of Kings win.

When we make new products, we always think of features that competitors don’t have, to increase our own selling points, and also to improve user experience and keep users on our platform. That sentence: I have what others don’t, and I am better than others. It is from the perspective of improving user experience to increase user activity of the product, because the pie is so big, if there are more users with you, there will be fewer users in other companies. For example, the rise of Honor of Kings led to the decline of Onmyoji .

User attribute determination

Users of a product can be divided into three groups: core users, ordinary users and general users, which are divided according to the users' recognition and dependence on the product. For example, strategy games attract users who prefer strategy games, so the activity data will be very impressive. The more users come, the more active they will be. If there are many users who like to play AR PG games, the kind that can play them or not, the activity level will be much weaker. This requires us to try our best to identify the target users when bringing in users. Otherwise, if there are a lot of new users but the activity level does not increase, we will come to the wrong conclusion that the product is unreliable, although most products are indeed unreliable.

3. Ways to increase user activity

Optimize your product

Optimizing products is the lowest-cost way to increase activity. Users will not use the app when they encounter the following problems, such as slow loading, unsuccessful login, crash, etc. To give a simple example, reducing one step in the registration process may increase the conversion rate by more than 5%. Product optimization is a long and meticulous task, and it must be combined with data analysis . Otherwise, you won’t know which link causes user loss , and the results of product optimization based on your feelings are often not very good.

The most important thing in optimizing a product is to optimize its core functions. The first version of the cleaning app I previously operated only had the function of cleaning. As I operate it in a Buddhist way, I will not say much about the experience of the function. Whether it is the length of time for cleaning, the prompt when cleaning is completed, or the UI display, it is simply not good enough. If users have a second choice, they will definitely leave us.

The first version was introduced for user testing, and the data was indeed very low, with neither retention nor daily active users increasing. Later, the cleaning function was optimized, the cleaning speed was increased, and the UI was beautified, and the daily active data gradually increased. When the core functions are almost optimized, you can optimize the auxiliary functions, such as personal information, customer service section, etc.

Function development

If it is a low-frequency software, there is a good way to integrate high-frequency functions into your own products, which can also increase daily active data. Here are two examples, one of which is the most popular knowledge quiz at the moment. Another one is to add an information section to the application market.

The user activity of video software is related to product attributes and user living habits. Take myself as an example. I am busy on weekdays and have no time to log in to the video software. I only open it to browse when I relax on weekends. The usage period is concentrated on weekends. Now the software has added a question-answering function, which is available every day, with one session every few hours. I usually don’t watch videos at noon, but when I have free time I will log in to the video software to answer questions. A single example can tell the whole story. We can infer that their daily active data must have skyrocketed. The product itself has not changed much. The only difference is the addition of the question-answering function.

The inspiration for adding the information section to the App Store comes from 360 Mobile Assistant and Cheetah Clean Master. When I first saw the information content added to these two software, I felt that it was completely irrelevant to their own functions. Why did they add this function? Later I saw that 360 Mobile Assistant would push news. When I saw something interesting, I would click on it to read it. At the same time, it would open the app market and browse the latest recommended games. The software that I used 2-3 times a month before is now only opened once every 2-3 days. Therefore, we have also added a news section to our app store. Unlike before when we only pushed apps and games, we can now push news. The user click rate has increased a lot, and the daily active data of the APP has also steadily increased.

Activity stimulation

Activity stimulation is the most immediate way to increase user activity. Humans are profit-seeking animals. Under the inducement of interests, users will do anything. The most common activities in the game are to receive a five-star hero for logging in the next day and a cute girl (Diao Chan) for logging in after seven days. If users want to get these rare items, they have to log in every day. The game itself has many rare props, and it is very convenient to use these as bait to stimulate users to log in. Relatively speaking, applications do not have such advantages.

When I was making an application before, I thought of two activities to increase activity. One was to log in for seven days to get traffic , and the other was to develop a points redemption system, where you could log in to get points and use the points to redeem physical items. The essence is to use benefits to motivate users to log in. The demand was simple, but it was not easy to implement. A traffic package costs at least 5 yuan, and a user is worth less than 5 yuan. It would definitely be a loss-making deal to log in for seven days. After weighing the pros and cons, we changed it to logging in for 15 days to get traffic. In the end, not much was received because some people missed signing in. Points redemption is easier to understand, and it is also a common method used by many APPs. When redeeming points, just control the number of redeemed items.

Add tips

When users are not actively using the APP, adding reminders can effectively activate users. Here I will talk about a few methods that are more effective. First one, push . When we first started operating the app, there was a push function, but we never used it. Because I wanted to post an advertisement, I tried the push function. At that time, the number of ad opens was not large, but when I looked at the product data the next day, the daily active data increased by 15%. At first, I didn’t associate it with the push, and I looked for many reasons but couldn’t find them. I suspected that the data growth was caused by the push. A week later, we tried pushing the app again, and the daily active data showed the same trend. Later, we used the push function as an important means of operation, and the click-through rate of the copy increased from 1% to 5%. The activity data of the APP has remained high.

Another method is to add a red dot or send an icon to the desktop. It is not recommended to use these two methods frequently, and the best effect is achieved when used occasionally. Because users are used to seeing red dots, they will click on icons when they see them on the desktop. Through these two methods, the daily active data has also been significantly improved. If used frequently, users will become disgusted and this will lead to loss of users, which will not be worth the cost.

4. The value of user activity

Active users are a company’s storytelling capital. It is aimed at small companies with financing needs. If a company has more than one million daily active users, it will still be favored by investors even if it is not profitable. Monetization is a small matter, and whether there are users using it is a big matter. The current product monetization methods are very mature, and products with millions of users can also generate some revenue by connecting to advertising. If it is a game, with active users, revenue will come naturally. Take Knives Out for example, there was no paid entrance in the early days, but now there are all kinds of virtual props, which ensures that you can spend money if you want to.

When the value of a single active user is fixed, the greater the number of active users, the higher the revenue. When we calculate revenue, we calculate the value of a single active user. For example, for a stand-alone game, a daily active user who brings in 2 yuan is considered a high-quality product. Then 10,000 daily active users can bring in 20,000 yuan in revenue. If the number of active users is increased to 20,000 through operational means, the revenue will become 40,000 yuan. From this perspective, leaders will require operations to increase user activity while keeping the number of new users unchanged, in order to increase revenue.

The more active the users are, the more helpful it is to the product ecosystem. What I'm talking about here is a bit abstract, after all, there is no data to support it. When our game was launched, the revenue per user on a server with 1,000 people was much lower than that on a server with 3,000 people. Theoretically speaking, users’ attributes are similar and there shouldn’t be such a difference. The only explanation is the server ecology. The more people use a product, the greater the value each person contributes.

The above is a simple analysis of the core indicator of the product, activity. In our work, after the active data stabilizes, our attention to this area will be slightly reduced. In the early stage of product operation , active data is the most important. Once the active data stabilizes, the operation begins to focus on monetization - user recharge data.

This article was compiled and published by @老虎 (Qinggua Media). Please indicate the author information and source when reprinting!

Product promotion services: APP promotion services Advertising platform Longyou Century