e-Zest members share technology ideas to foster digital transformation.

Why Manual Testing is better than Automation Testing?

Written by Shruti Mangalvedhe | Dec 30, 2014 6:15:54 PM

Most of the people from the technology world rate automation testing as more effective and efficient. I, however, do not conform to that view and feel manual testing is way more effective.

I personally think many of the challenging bugs can be found by manual testing. Thing such as ‘User Interface’ , ‘Usability’, ‘Installation and setup’, ‘Configuration and Compatibility’, ‘Error handling‘, ‘Localization’ and any other aspect that relies heavily on human judgment cannot be automated.

Automation testing involves testing the application using automation tools that reduce human involvement. Automation testing saves time, speeds up the process, makes regression testing easy, and enables use of the same test suit repeatedly. However, many people forget the fact that the goal of automation is to reduce the number of test cases to be run manually and not eliminate manual testing completely.

If the code or the requirement changes frequently, then automation testing is not suitable because the test case needs to be updated and the script needs to be rerecorded and replaced by a new script. Even a small mistake can lead to deadly consequences. A lot of testing areas are left uncovered in automation testing. Moreover, there is no human perspective in automation testing.

Database testing, a key task for testers, can be done only through manual testing. Also, automation becomes more complex with increase in requirements which makes test data preservation difficult. Additionally, automation is suitable for large projects which translates to many developers and testers so in order to have correct results. Also, the test case needs to be executed on a single machine which is very difficult.

There is no doubting the fact that automation tools are fast and enhance the testing process. But scripts are written by humans and not tools. Manual testing gives the user the required flexibility to do ad-hoc testing. The user can also spend more time with the application which results in greater odds to find real user level bugs. Users can think of more negative scenarios which tools cannot.

I agree I may sound a little biased about manual testing. Would love to hear your thoughts on this as well.

Reference:

www.softwaretestingclub.com