Smoke vs Sanity vs Regression Testing
Early in my QA career, I was assigned a bug to test, a minor fix in one of the key modules of our enterprise portal. I opened the application, focused only on the specific scenario mentioned in the ticket, confirmed that the issue was resolved, and confidently marked it as "Done"
A while later, my team lead reviewed the same fix but took a different path to reach the same functionality. Within minutes, he uncovered a new issue that had been introduced as a side effect of the fix.
He turned to me and asked, "Did you check this part?"
"Yes," I replied, "I tested it."
But what I really meant was: I only checked what I was told to check.
Watching him test it in front of me - navigating different routes, trying other workflows - I realized how limited my approach was. That bug fix had impacted a different module, something I hadn't even considered. That's when it clicked:
✅ Fix validation = Sanity Testing