What Tests Should We Automate First?
It is rare that I get the question, “what tests should we automate first?” More often, folks seem to already know what they want to test. It is a forgone conclusion that they start with the biggest time-savers. They want to take the tests that take the most time to run manually, and automate them.
That’s understandable. Why not buy back your time testing labor with the highest cost tests? Why not get a quick ROI?
Many people believe the only reason we automate test scripts is to save labor and time. Done right, we can save manual labor, yes. Done wrong and you’ll multiply your work – paying people to manage broken tests everyday while continuing to do all the work you did before.
There are some problems with automating the longest running script first. Like the fact that you have to build out much more functionality to get one script to work. Like the fact that it’s probably the longest-running script because it tests many different things. Like the fact that it’s probably the most complex of all your scripts. It likely takes the most set up time, configure, and verify for a human. Imagine the difficulty of trying to program these complexities!
If you have these thoughts you’re not alone. If this is your reasoning, it is completely understandable. After sheparding many clients to success with test automation, I’ve found some ways to get test automation started fast and furious that may be counter-intuitive. Check out the recording of the webinar I did on this very subject.