Should you Automate or not ?


The key aspects in Software Testing don't necessarily come from using the latest and greatest techniques (automation, unit testing etc). Here are a few best practices that are critical in our testing industry - 

Must Read :

  • communication (team and client)
  • culture (fun to work at)
  • engage users and developers and management (at different times)
  • reporting (do the clients want those reports or do they read them)
  • Automate tests that can be automated BUT make sure the automation is maintainable and offers a good ROI
  • Test as early as possible in the development process BUT be careful not to slow down the development process with comprehensive testing all the way up front.
  • Encourage developers to take responsibility for the quality of their code BUT don't forget the importance of independent testing

Automation gives you ownership over the testing of a product. It also keeps your coding skills sharp, in case you want to move to development. It ensures that regressions will be caught before they come back to bite you in the ass and might, every now and then, uncover an unknown issue. While developing automated test cases, you will inevitably end up developing a test framework of your own. With your own tools, your own processes, your own common libraries, etc. After that, whenever someone else needs to automate something they'll use your stuff and that will give you visibility and relevance.

Good Read : 

Now, the problem with automated testing is that very often test teams end up spending all their time maintaining their existing automation and failing to come up with new automated tests or actually spending some time with the product to manually play with it. Also, test teams who focus on automation usually end up designing their test plans based on how much automation they can do and not in how many features have to be tested and how thorough the tests need to be. 

And my number one test-automation pet peeve is definitely when some obtuse tester, in his never-ending pursue of a 100% test pass, sometimes goes and fixes (or even removes) a failing test case to make it pass instead of actually figuring out if there is a bug or an issue with the product. I know this sounds stupid and common sense should stop anyone from doing this but, trust me, I've seen it happen a lot. 

career growth in testing has to do not only with automation but with development of test plans, test tools, test frameworks and test strategies that will ultimately improve the quality of the products that the company sells and reduce the time spent fixing potential problems

Recommended Read : 

Want reach out to millions of readers out there?

Would you like to see you post published here ?

Send in your posts to . We will help you reach out to the vast community of testers and let the world notice you.