The pain points of those app testing that have to be said

Case: After the iOS version of a carpooling software was updated, there was a shocking bug that caused the car to be unable to be called due to a positioning problem. Subsequently, the official took a series of remedial measures, including: texting an apology, giving away vouchers, and changing the homepage heat map to a new car-hailing entrance. More seriously, due to Apple's Appstore review rules, even if the bug is discovered and solved at the first time, the official can't release a new version at the first time, so this problem will not be solved until a week later, it is conceivable It is known that a large number of users must be lost to other carpooling software within this week, and this accident has also severely dampened the confidence of old users, and has also brought far-reaching negative effects to the official.

In apps, such serious bugs are not uncommon. Why do these giants spend a lot of manpower, material resources, financial resources and energy to develop products that still have so many serious problems. The reason: the internal test of the app is very difficult to do! Pain points of

APP testing

Difficulty in distribution: Take iOS applications as an example, due to the limitations of the operating system, it takes about 17 steps and about 20 minutes to install an internal testing application on the tester’s iPhone, and the installation of each mobile phone will take about 20 minutes. Repeat the above process. Cumbersome distribution steps pose a serious obstacle to adequate testing.

Highly fragmented operating devices: For a long time, the fragmentation of the Android platform has been criticized for various reasons: devices are complex, brands are numerous, versions are different, resolutions are not uniform, and so on. Not only did these create huge barriers to development, but they also brought high costs to testing. The reason why the iOS platform was popular with developers before was because of the unity of its devices and systems, which brought a lot of convenience to developers. But today, to develop an iOS app, you have to take into account everything from the Apple Watch to the Apple TV and countless iPad and iPhone sizes and systems ranging from iOS 6 to iOS 9. Therefore, the fragmentation of the iOS platform has also become a major problem in developing high-quality applications.

Complex running scenarios: In addition to operating systems and devices, apps also run in various complex scenarios. For example: WiFi, 2G, 3G, 4G and other network status, moving or stationary and other motion status. These complex scenarios are not faced by traditional software, and these have also become difficulties in testing.

Highly complex functions: An app may handle complex information such as gestures, GPS coordinates, acceleration, camera, push notifications, other interoperable devices, cloud storage, network interaction, mobile payment and other complex information that traditional software has never faced. Processing increases the complexity of the software and also increases the difficulty of testing.

Flashbacks are difficult to track: Generally speaking, if an app fails to gain stable user growth within 3 months, then the app can basically be sentenced to death. Then the early user experience is crucial to an application's ability to stand, and the worst user experience is crashing. If a user encounters two crashes on the first day of using an app, the user generally chooses to uninstall the app. And those flashbacks that occur in user scenarios are often difficult to reproduce and track. So crashing is a lingering nightmare for developers and testers.

Versions are difficult to track: Apps in the internal testing phase generally iterate frequently, and new versions are frequently released. The management of these numerous versions also requires a unified solution.

Problems are difficult to feedback: modern operating systems generally can only run one program in the foreground at the same time, so the internal test users need to save screenshots when testing, and upload or send the corresponding questions to the developer, and this feedback process is also cumbersome .

Difficulty for internal testing users: A problem that developers often face is that after the product is officially launched, there will be various serious bugs that have not been encountered in the internal testing process. The reason is that there are too few internal test users and the test coverage is not comprehensive.

To sum up, during the internal testing of the app, it faced many problems, such as difficult installation, complex running scenarios, highly complex functions, difficult to track crashes, difficult version control, difficult feedback of problems, and difficulty for internal testing users to obtain. Therefore, it is not surprising that the serious bugs of the above-mentioned carpooling software appear.