Differences Between Mobile Web and Mobile Platform Tests
In this article, I will be talking about the main differences between mobile app testing and mobile web testing.
While running our functional tests, we can classify the main differences and effects between the two platforms as follows:
- Mobile Device and Operating System Diversity
- Launching
- Connection
- Complexity
Mobile Device and Operating System Diversity
There are many different models of devices and different operating systems in which mobile applications are used. These are also one of the most important factors affecting how our application works. It is important that the application works correctly on every model and operating system. For example, a feature may not work on Android 7.0 while it is working properly on Android 9.0 or a UI feature might appear fine on a 6 inch device while it might appear distorted on a 4.5 inch device. While running our tests, we should take these into consideration and run them to cover these situations. If you have more variety of real devices for testing, test quality will be better.
On the mobile web, it is very important to run tests on different browsers. Although the vast majority of users use the Google Chrome browser, we must make sure that the improvements made in web browsers such as Mozilla, Opera, Opera Mini, and Safari work correctly. Of course, the browser versions are important as well as the operating system version of the device.
There is also a good application in this regard when testing it manually. BrowserStack. It can make tests much easier by offering different operating systems and devices.
Launching
In mobile applications, we do not always leave the package. Usually, a new package is released every two weeks or once a month. We are sending release package in Trendyol Mobile Team every 15 days usually. Sometimes may be three or four week depends on the situation. At these times, the enhancements to the package are combined and the regression test is run. It is expected that both these improvements will work properly in the package and that the general functions will not be broken. Unless the user updates the application, they cannot see the new improvements (excluding force update).
On the mobile web, if the feature has passed the tests successfully, it can be sent to the production environment at any time. After deploying to the production environment, all users can see the improvement. It should be checked that the development and application made after it is released is working correctly.
We need to download mobile apps to the phone. For this reason, set-up and lift tests should be done. For example, when we send a new package to the store, it will be uploaded to the currently installed application as it will go to this user as an update. When the user updated the app, the existing situation should not be corrupted. Most importantly, if the user is logged in, the session should not be terminated after the update. For this reason, when a new package is released, the overlay test is of great importance.
Mobile web applications, on the other hand, do not require setting up on the phone as they can be used via Url. As in the mobile application, it is important that the existing user sessions are not terminated when a new development is made in mobile web applications and this is controlled.
Connection
Mobile web apps do not work when offline. When testing these applications, we look at how they work under different internet speeds. For example: How fast the home page loads at slow 3g network speed, and all widgets and images on the page are loaded correctly.
Mobile apps, on the other hand, can often work offline. At least some, if not all, features can be used. In practice, how it works in 3G or 4G and Wifi, how the application behaves when access is interrupted are important issues and should be tested.
Complexity
Mobile applications are more complex than the mobile web, since they have features such as pull, shake, scroll, etc. This expands the scope of testing and means there are more modules to test. For example, if we think Trendyol app and mobile web, it has not on the mobile web while the application has the swipe feature on the home page between tabs This is a situation that should be taken into account when testing.
In addition to these basic items, of course, there will be differences specific to the application or mobile website you are testing. Considering what kind of product it is and what it offers to the user, it should be considered how to test which modules and proceed accordingly.
Considering the items mentioned above, to offer an almost error-free product for the mobile application or mobile web both during the preparation of the test cases and while running the tests is the most important task of a QA.
In addition to these, there are some differences in service and event tests. When we can test the Mobile Web applications on the desktop browser, we can test the events thrown here on the browser console. In service tests, we can see which service with which requests were visited and which answers were returned over the network.
In mobile applications, the situation is a little more complicated. While you can test events in logs on the IDE, it can be checked by adding an Okhttp inspector in the debug application for service tests. We are using Chucker on service tests on Trendyol Mobile Team. It’s a nice library for that on the Android side.
Thank you for taking your time and reading. Have a good day.
