Your question is Metric Comparison vs Statistical Test. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
You are reviewing a product metric change after a launch or experiment. The team sees a difference between two groups and wants to know whether to trust the raw comparison or run a formal test.
How do you decide whether to use a simple metric comparison or a more rigorous statistical test?