In part two of our four-part series on the Future of Automation, we are going to look at how RPA positively impacts the second question that IT professionals face – Did I break anything in my last release?

Regression and Smoke Test Automation with RPA

As we discussed in Part 1, Robotic Process Automation (RPA) is quickly becoming the one the hottest topics for the IT Professional for its ability to provide relief from manual repetitive tasks. In this article, we are going to focus on RPA and the role it can play in what is usually the most manual of all testing processes – regression and smoke testing.

Regression testing is focused on ensuring that a new release does not break anything else within the application, while smoke testing is focused on ensuring that the new release made it to production OK and did not adversely impact any other system in the production environments. Traditionally, these types of tests when assisted by tools such as WinRunner or Selenium would almost always require a developer whose sole task was to keep the scripts in sync with the application. But RPA provides similar scripting to these tools with the advantage of not requiring a developer. Hence, reasonably technically proficient testers can be trained to develop their own scripts.

The only drawback here would be in terms of reporting – since RPA does not offer the advanced reports and in-depth analysis that other dedicated tools provide it is best to combine RPA with the existing tools available to the testing team. The addition of RPA serves as a tremendous productivity booster by taking the manual processes away and offering easy to maintain testing scripts to the test engineer.

Similarly, RPA is proving to be very useful for smoke testing activities for production deployment verification or system impact analysis. Traditionally, these tasks are almost exclusively manual as they are usually complex steps performed across multiple systems. RPA through its user interface and recording abilities will allow the test professional to orchestrate very complex testing scenarios that touch multiple systems that can easily automate well beyond the traditional tools that were available.

Check out part three of our ongoing series on the future of automation, where we answer the question “Is anything going to fail?”