Automation…. Automation……. Automation that is mantra we hear in testing world wherever we go today. Effective Automation is critical in today’s world where even weekly release to production is not fast enough. We can no longer have 1 month long regression cycle like the good old days and must have automated suites to achieve frequent releases. But do it incorrectly and it would soon spiral out of control. What I am seeing more and more is that to achieve quick Automation the quality is getting sacrificed. And an unstable automated suite is so much more inefficient than doing it manually. So how do we achieve Effective automation? Right tool & Framework: Selecting right toolset is probably the most important part and unfortunately where most projects fail. • Look at your application suite and try few different tools to find what will work for you. Don’t commit to an automation tool or framework just because it is the buzz word. • Setup the infrastructure with scalability in mind. Look into things like distributed execution, TDM tools, CI tool integration etc. • Spend good amount of time defining guidelines and framework for creating automation script. Test your Test script: Ensure your automation is not one-hit wonder but can ensure productivity for long period of time • Have exit criteria for automation. E.g: Don’t mark a script as automated till it has run successfully for more than 20 times which different data set • Have review process to ensure core reusable items are not corrupted • Track successful runs of scripts even after it is deployed. Use CI tools like Jenkins to track successful runs So what’s next? As applications become more complex so will automation. The next obvious stage evolvement would be integration with Machine learning and AI. It would be impossible to achieve continuous success at ever increasing pace of releases without infusing intelligence into automation. From self-healing scripts to predicting failure AI and Machine learning would be critical to make automation reliable and achievable