Bloggo back to the blog
Choosing the Right Strategy for Automating Mobile Application Testing-->
Summary: Mobile apps are fast becoming a part of core enterprise strategy. As the competition heats up for greater user experience, focus will shift to testing the mobile apps for a bug-proof application launch. What mobile application developers need is a strategy that offers the best solution for comprehensive testing.
Mobile app marketing is not just a buzzword anymore. Gartner predicts that 102 billion apps will have been downloaded in 2013 alone. Marketers worldwide are therefore in no doubt about the potential of and the need to launch a mobile app.
But before companies jump on the mobile app bandwagon, they need to make sure that their app is ready for the market. Nothing is more damaging than an app that has bugs. Automated testing is a guaranteed way to test mobile apps for bugs, identify errors and then resolve issues before upload. Incidentally, automated testing itself has become complex and time consuming, compared to desktop and web applications. Here is a brief update about the current scenario.
Testing Automation: Current Scenario
Today, Android and iOS are the most successful mobile platforms. With acquisition of Nokia by Microsoft, Windows mobile OS may soon catch up. There are several other fringe platforms that still have an audience. Multiple operating systems, and myriad number of handset manufacturers have made the mobile ecosystem extremely complex. App developers face one key challenge today.
How do you test an app for innumerable permutation and combinations?
The short answer is a comprehensive mobile strategy that takes into consideration only the relevant target device and network infrastructure parameters. App makers can improve their testing strategy by employing an effective combination of manual and automated testing tools to cover both functional and non-functional testing. The choice of tools is vital when it comes to getting mobile applications to market on time and within budget. App makers have to make the choice between Open Source or Proprietary tools for test automation.
Going Open Source for Test Automation
Open Source has come a long way, and is a highly beneficial option to conduct tests on mobile platform. Some of the prominent tools you could consider include Robotium, Appium, Calabash, MonkeyTalk, GUIDancer, and MonkeyRunner. At ITC Infotech we have extensively used Calabash to test mobile applications on iOS and Android. It’s a powerful tool when used in conjunction with others like Selenium, or Cucumber.
Going Open Source with such tools offers greater cost benefits and at times better functionality as well. Open source tools can be used with various programming languages such as Java, Ruby, Python etc, which provide better functionality for customization. Open Source is also great for testing applications where only the APK file is available and the implementation details are not known. There is a flipside of using Open Source in terms of support and capabilities. It can be difficult for test automation engineers to use Open Source tools. This is where Proprietary Test Automation Tools come in handy.
Selecting the Right Proprietary Test Automation Tool
The mobile test automation scene today is highly competitive. Multiple vendors, multiple choices, it seems great for the consumers. However, when it comes to choosing the right tool, too many options is not always an ideal scenario. It helps to have a well defined criterion for selection. The most important criteria while selecting a test automation tool is the technology employed by the tool. Based on this there are two types of tools app testers could employ. Object-based recognition, and Image-based recognition.
Using an object-based tool such as M-eux Test by Jamo Solutions, is advantageous when the application’s GUI is not stable. However, this type of tool requires the back-end of the application to remain stable. On the other hand, an image-based tool such as T-Plan Robot by T-Plan, makes it possible to have a platform-independent test framework, although it requires the app’s GUI to be stable. That’s why it’s important to know the status of application GUI before opting for Test Automation tool.
In addition to these, there is a third option of cloud-based tools, which offer manual and automated testing on real devices connected to their cloud. Some cloud-based mobile test automation tool vendors include SOASTA, Keynote Systems, Tricentis, TestDroid and Perfecto Mobile.
It’s important to select the options that best suit the requirement, and often a customized framework could be the best solution. At ITC Infotech we develop and customize Platform Independent Framework along with effective reporting mechanisms with ease of customizing as per the end user needs. We have expertise in using commercials tools like T-Plan Robot, Eggplant, Jamo Solutions, HP-UFT with mobile add-on and more.
To effectively mitigate the challenges involved in mobile test automation, you need a powerful testing framework. The best way to create one is by optimizing the selection of target devices with a mix of real devices and simulators. Other factors, such as support for applicable mobile platforms, script reusability, and the total cost of ownership should also be taken into account when selecting automation tools.
In conclusion, combining specific aspects of application testing with traditional best practices can be really effective way to ensure greater efficiency with optimized costs.
About the Author
Vijay Balasubramaniam leads the Testing and Quality Assurance Practice at ITC Infotech, and is responsible for creating a sustainable and strategic competitive advantage for the company. Vijay has a rich and diverse work experience, and has worked with organizations such as Deutsche Bank, Barclays Bank, Microsoft and Infosys, and has won several accolades for his transformational plans and strategies.
With over 14 years of experience in the IT industry, Vijay brings in special focus on Testing and Quality Analysis. He has taken industry standard models like TMMi® and has overlaid process model and experiences with the aim to drive practitioner-oriented process improvements.