But even if it is — you need professional magician(s) to operate it.
Automation is mostly about approach and teamwork. Sure, a lot of things depend on the tools used, but hey, tools won’t do the job themselves. In the end, it’s people who have to take actions and retrieve results.
I guess things are the same for most of industries. Tools — be it factory machines, business analytics software, transport or test automation solutions — can help you do the job faster and with less resources spent. But it is the task of the human operator to set the goal, launch the process, and achieve results.
The progress is going faster in some specific spheres. We already have self-driving cars which can take you around the city defining an optimal route without your involvement. It will soon be technically possible to message a fully automated car to take you home after a party. Perhaps, in a few years, we will have an AI that’s able to perform full-scale regression testing of web UIs. But today, the UI test automation task requires:
- A detailed plan,
- A team,
- And (finally) a tool which meets the requirements and project needs (refer to the #1 “Plan” part for details).
Creating a plan should, of course, precede all other stages. Assembling a team should also be done before the final decision about tools and environments and here’s why. It is common knowledge that professional, motivated technicians will do better even with limited resources than a crew of novices provided with an abundance of technical stuff.
Sure, this doesn’t mean you should keep your dream team under-equipped. In fact, the most efficient option would be to let them decide on part #3. Engineers should know what they need; and given they are a well-built team, they are likely to come to some agreement: an automated testing tool which would be really helpful for the project.
So again: people first; and proper tools will be discovered/created at a proper time. Set clear goals and objectives and make a point of hiring professionals — and they would find an optimal way of reaching this goal themselves. Good leadership would also be awesome. Combine all that and then do a small research together to learn about the current state of the global UI test automation market. Try different solutions, discuss, dispute, try again — and who knows, maybe a real magical wand will thus be obtained!