That is todays Mantra it seems. But there is no second chance to correct a first impression, if your rushed prototype was too rushed and no one gets the value beneath the flaws. It really depends on your product and market.
Of course, this is an MVP, so it has to work and be decent, and it also depends on the market and product. But when you launch your product, you're not really making a first impression on the whole world - only a few people will actually try it out. Having people try out your beta version helps you build in the right direction. Otherwise, you risk spending years building a product that nobody is interested in.
