From an academic perspective, sure? I guess? But most things done in the name of academia are for the purpose of learning, so the goals are different. If your g

From an academic perspective, sure? I guess? But most things done in the name of academia are for the purpose of learning, so the goals are different. If your goal is to build a PRODUCT this is stupid advice. You really want to stop once your primary use cases go end to end. It doesn't even need to scale. Go make a sale or whatever flavor of "get users" your product needs to be validated. Most products don't get the benefit of being able to stand on their own. Every customer you add comes with their own unique last 10% that you'll need to work on to get them to sign up. So go find out what THOSE requirements are, or you'll end up doing the wrong "last 10%". Oh, and make sure you understand what the client is actually asking for, or you'll end up doing the wrong 10% and losing your next big customer, which can be a super demoralizing experience.