‘Is manual testing going to go the way of the dinosaur?’
‘Are manual testers becoming obsolete?’
Just a few ways of phrasing concerns that have been raised a lot in the testing community in the past few months, possibly even years. With software delivery and test cycles becoming ever shorter, organizations are depending more and more on automated testing to deliver software that has at least an acceptable level of quality.
Testers not involved or even interested in automated testing (‘manual testers’, for lack of a better description) might start to worry whether their job is still secure. I would like to argue that it is.
The first and foremost reason is that automated tests do not improve the quality of a product, nor do they improve the quality of your testing process. Automated test tools are very useful for exactly one thing: quick and/or unattended execution of a number of predefined checks.
If those automated checks are of poor quality, you’ll get poor quality results. You’ll just get them faster. So while automated tests may speed up the testing process, they’ll never improve it.
However, testers do.
If your automated tests only check irrelevant things, your test tool won’t tell you.
Your automated test tool does not automatically generate tests for brand new software components (most of the time they don’t, at least).
Your automated tests do not discuss features with developers and business representatives to see whether the specifications are complete and unambiguous.
Your automated tests do not think ‘hey, what would happen if I do XYZ instead of ABC?’
I could go on for a while, but in short, there is still a lot of work to do for testers.
You won’t get away much longer with following the ‘old world’ process of waiting for specifications – writing test plans – writing test cases – executing test cases – reporting bugs – writing test reports. But you don’t have to fear that there is no place for you in the software development process.
And of course, you could always venture into test automation..