As a Quality Assurance, for first time i usually analyze the feature to find which credential things in feature. Then, i give a feedback / suggestion about behaviour feature or design, so our target market will familiar with the feature quickly.
if all of those things done, i start to create test plan as our strategy for testing, when i find defect / bug, i will create and report the defect / bug with structured and neat writing so the developer will fixed as soon as possible.
When testing done, i will create report as documentation. so, in the future if there is an error we can learn and fix how can the error happened and take the action to fix the defect immediately.
Actually, when I test the feature I always think about an acitivty that is rarely done by user. Why ? because a possible defect / bug will be found there. Sometimes developers rarely think about an activity that people rarely do on a feature and only think about activities that people usually do.
What if I encounter a defect ? I usually reproduce the defect by executing the steps then see the feedback in back-end. If stiil doesn’t the troubleshooting, I always do comparing data. First data with correct condition and second data with condition that makes the system will produce the defect
How to choose the test cases ? when choosing test cases I am required to choose a test cases that is also related to the features we are testing, the goal is to see whether the integrity between these features can work well or not