Is the crash rate of iOS higher than that of Android? Data from Tingyun tells you the truth

The rapid development of mobile Internet and smart phones has attracted many developers to join the entrepreneurial wave. The emergence of different types of mobile applications has added convenience to our lives, brought us joy, and helped us learn more knowledge. We are increasingly relying on applications to perceive the world and communicate with each other.

According to the latest statistics from AppFigures, an application data tracking company, the number of applications and developers on Google Play Store exceeded that of Apple App Store in 2014. The total number of Android applications on Play Store reached 1.43 million, while the total number of iOS applications on App Store was 1.21 million.

However, among the millions of applications, how many mobile applications will enter our smartphones and become essential software for mobile phones? In addition to the design and functions of the mobile application itself, the performance issues of the application will directly affect the fate of the mobile application.

Generally speaking, due to the existence of various factors such as network environment, device model, development environment, etc., there are more than 107 million combinations of application performance problems (5079 different models * 1172 operating systems * 18 operators). The frequency and types of application performance errors are beyond imagination. Factors such as connection timeout, flash back, freeze, crash, black and white screen, network hijacking, poor interactive performance, CPU usage issues, memory leaks, and poor interfaces will have a huge negative impact on mobile application performance.

Recently, Tingyun released the first application performance management report in China, "2014 China Mobile Application Performance Management White Paper", which covers more than 350 million terminal devices, more than 240 million daily startups, and more than 10 billion real user requests per day through Tingyun App monitoring. It describes the current status of mobile application performance from five important dimensions: crash, error, network request response time, interactive performance, and operator network response time, and four data standard intervals: excellent, standard, minor hidden dangers, and serious hidden dangers. The report shows that more than 74% of users will choose to remain silent, endure, or leave in the face of performance problems, and nearly 5% of real users will give up using the application after a performance problem in a mobile application causes a 10-second delay in response.

iOS crashes more than Android

In terms of mobile application performance, the impact of crashes is the most serious. Mobile application crashes are mainly caused by the operating system, which refers to the forced closure of the application during operation, thus interrupting the user's ongoing operation experience. Application crashes can cause key business interruptions, reduced user retention, poor brand reputation, and reduced life cycle value.

Figure 1: Mobile application crash rate standards proposed by Tingyun

The report statistics show that when the crash rate of iOS exceeds 8‰ and the crash rate of Android exceeds 4‰, the number of active users shows a clear downward trend, and the average crash rates of these two systems in 2014 exceeded the standard values, indicating that there are generally minor hidden dangers.

Figure 2: iOS system crash distribution

In the iOS system, iOS5.X and iOS8 are the operating systems with the highest crash rates. Compared with other systems, iOS7X has stable overall crash performance and a crash rate of less than 10‰. From this, it can be seen that mobile developers must pay attention to the support of mobile applications for the latest operating system versions and the continued compatibility with old versions.

Figure 3: Android system crash distribution

Compared with iOS system, Android has a lower overall crash rate. Only Android 2X system has a crash rate higher than 10‰, which is poor. After Android 4.X version, the crash rate is stable at 2‰-4‰.

Figure 4: OS crash rate comparison

According to statistics, the average crash rate of iOS is about twice that of Android. Why does the closed iOS system crash more than the open Android? Tingyun summarized four reasons. First, Android 4.X has been significantly improved compared to previous versions, but the latest iOS 8 version is unstable; second, in terms of update strategy, iOS has a longer update push cycle, while Android will push updates at any time; third, due to the particularity of language/system architecture, OC needs to directly face the underlying API, which has a higher possibility of error, and is affected by the OS version. The impact of hardware differences is greater than that of Java; finally, it is precisely because the iOS system is more restricted, such as memory, background, API restrictions, etc., which is also one of the reasons for the high crash rate.

When an application crashes, a crash report is created and stored on the device where the crash occurs. This report describes the conditions under which the application crashed. Usually, the report also contains a complete stack trace of the currently running thread. Mobile developers can use the stack information to infer the cause of the crash and make corrections.

Other performance indicators

Mobile application errors refer to the summary of HTTP errors and network errors that occur in the production environment of the application, which will cause the application page to fail to display and affect the user experience. In terms of overall errors, connection timeout, connection failure, 500 error and unknown host are ranked high in mobile application errors, accounting for 6.3‰ of the overall error rate, of which request timeout accounts for *** of network errors. When HTTP errors occur, mobile application developers should prescribe the right remedy based on the error code, and network errors require analyzing the cause of the problem from the network level.

Figure 5: Mobile application error rate standards proposed by Tingyun

The mobile application request response time will directly affect user retention. The number of retained users and the retention rate directly reflect the quality of the application and its ability to retain users. After monitoring the host request response time of more than 50,000 mobile applications, Tingyun found that when the request response time exceeds 2000ms, users begin to churn.

Figure 6: Tingyun's request response time standard#p#

Interface layout, view loading, stored procedures, database query, program operation and other factors can directly cause slow loading and lead to user loss of patience. According to statistics, when the application interaction execution performance time reaches 400ms, the application enters the stage of slight performance risk. By querying the front-end UI thread or the background working thread, the application interaction performance bottleneck can be found.

With the development of 2G to 4G networks, users have put forward higher and higher requirements on the network performance of operators. However, under different networks of the same operator, the network response time in different provinces is also very different. Since China has three major operators, China Mobile, China Unicom, and China Telecom, as well as different networks such as 2G, 3G, and 4G, there are many combinations. Mobile developers need to consider the impact of regions, base station coverage, population coverage, and network standards to ensure a smooth user experience.

Mobile application performance management is key

Tingyun R&D personnel said: "Compared with user churn, mobile application performance issues will cause more losses to users. For example, when the application crashes or errors occur, it will cause key business interruptions and revenue declines, which will further affect the product's life cycle value; if the application request response time is long, it will lead to a slow end-user experience and a decline in user retention; if the application interaction performance is slow, then the page elements will load slowly, causing freezes or incomplete layout chaos."

From this point of view, it is not enough to just develop a mobile application. How to ensure the performance of mobile applications, improve user experience, and stop user churn is a problem that every mobile developer needs to consider. Tingyun App focuses on helping developers monitor and manage performance issues after the application is launched. By embedding Tingyun App SDK in the application, synchronizing the real user access experience, and promptly discovering problems such as crashes, connection timeouts, and memory leaks during use, it helps developers stop user churn as soon as possible. Through real-time, multi-dimensional and stereoscopic performance data display and automatic analysis, it can prevent problems before they happen, reduce the maintenance and iteration costs of the App after it is launched, and directly improve user retention rate.