Should We Automate Everything? [Video]
Automated testing does speed up processes and free up man hours, but it's not always the right choice. Learn about when manual testing is still best.
Join the DZone community and get the full member experience.Join For Free
I am meeting a lot of people these days - trainees, test leads, test managers, and customers - who have been brainwashed that manual testing is slowly dying and automation testing is the future.
With such flawed understanding comes unreasonable demands:
Releasing manual testers, otherwise technically very strong, from the project.
Pushing for 100% automation even where it is not possible or not required.
Pressuring testers to learn all sorts of automation tools in a short timeline without no future plan to use them.
Testers writing automation code for one-time testing activities with no respect for RoI (Return on Investment).
Trainees reluctant to learn anything other than to code. This results in ignorance of fundamental testing techniques like boundary value analysis, equivalence partitioning techniques, Data Flow testing, etc.
Clients hoping to reduce team size quickly using automation.
The root cause for this, I think, lies in lack of understanding about what automation can and can not do. In such a scenario, it is very important to have an Automation Expert on the team. The Automation Expert can be consulted for all automation proposals and his advice can be useful to make go/no-go decisions. His knowledge and experience can also be leveraged to deal with difficult customer demands.
The Automation Expert will also help to understand the limitations of automated testing, and hence, the team will come to appreciate manual testing and not just throw it out the window like antiquated furniture.
This presentation from Angie Jones resonates with this article:
Opinions expressed by DZone contributors are their own.