Behavior analysis provides data about user behavior. It analyzes when, where, and what actions users perform in your app, through which channels, and for how long. This information helps you understand user behavior patterns, access paths, and characteristics.
To view behavior analysis data, follow these steps:
Log on to the console. Click Products and Services > mPaaS, and select an application.
In the navigation pane on the left, go to Mobile Analysis Service > Basic analysis > Behavior analysis.
Filter the analysis data. At the top of the page, select the platform, application version, and channel. Then, select a date to view the corresponding behavior analysis data.
The behavior analysis page uses charts to display the following behavior metrics:
Hover over the question mark (
) in the upper-right corner of a metric to view its description and calculation rule.Statistics are based on historical daily summaries. The database records these daily summaries for all statistical items. Data from the current day is not included in the results.
For more information about metric calculation rules, see Basic metric calculation rules.
User activity analysis
The time a user starts using the app is calculated by subtracting the background duration from the log reporting time in the system field. Starting from 00:00, the day is divided into two-hour intervals, such as 2:00 AM to 4:00 AM. A user's active time is assigned to the interval in which they start using the app.
Daily active duration: The average time a user spends on the app per day. This is calculated by dividing the total app usage duration by the number of unique device IDs (UV).
Times/day·person (times): The average number of times a user opens the app per day. This is calculated by dividing the total number of app launches (PV) by the number of unique users (UV).
Top 5 popular pages
This section lists the five most visited pages in the app. You can click a page name to view its detailed analysis page.
Channel distribution
Based on activation instrumentation, this section displays the top three channels with the most app downloads. The calculation deduplicates multiple downloads by the same user from the same channel.
For example, if User A downloads the application once from channel C1 and twice from channel C2, the system counts this as one download from channel C1 and one from channel C2.
To track channel distribution for Android devices, you must generate and upload separate channel packages to the corresponding app marketplaces. These packages contain a field to identify the channel. When a user downloads and runs the app, the client automatically reports this instrumentation field to the server-side. For specific instructions on how to generate channel packages, see the channel ID settings in Based on the mPaaS framework.
Startup speed
This section shows the average startup time for the first launch and for subsequent launches, measured in seconds.
The first launch is the first time the app is started after installation. A subsequent launch is any launch after the first one.
Page traffic
This chart shows the traffic sources and destinations for the current page. Based on the pid and refer fields, it calculates the access sources for the page, the proportion of traffic from each source, the destination pages that users navigate to, and the proportion of traffic to each destination.
You can use the dropdown list in the upper-right corner of the chart to switch pages and view the corresponding data. The dropdown list contains the pages that you added on the Page Analysis > Page Configuration page.
User distribution by region
Based on activation instrumentation, this section categorizes users by the region field and displays the top five regions with the highest percentage of unique users. It also shows the user percentage for each of these top five regions.
For a map of China, you can calculate the percentage of unique users by province.
For a world map, you can calculate the percentage of unique users by country.