There is a some standard of user experience that people expect in most consumer (and to some extent, enterprise) products. This often comes to a head with someone trying to get their baby out the door. And I get it– getting it in the hands of users is the fastest way to get feedback. But if you don’t do build something well enough, then it will fail and you wouldn’t have tested the actual core value. Quality alone won’t keep people if it’s not something people like. Substantial value will, but if its problems eclipse its value, well– then it’s not all that valuable anymore.
How do you know that it’s good enough? Will you use it (if this is a product for yourself)? If not, will someone hand over a credit card in its current state? Or for consumer products, will someone use it and show other people your product? If not, you should probably learn what is wrong with it, and fix it.
The other thing I’ve seen most entrepreneurs trip over is being honest with yourself as to what the biggest untested things are, and figuring out the best way to test it. Once you’ve proven your MVP though (or more likely, it fails), be okay with starting from scratch if it’ll improve momentum. If continuing makes sense, do that.