It’s tempting to spend disproportional amount of time on the most visible parts of the product (that’s where’ you’ll get the most comments as well) like visual design.
But it doesn’t really matter how good feature’s visual design, interaction or animation, if no one is going to use it. If it doesn’t address big enough need for a sensible group of people. And if it’s not clear which feature would do that, it’s time to prioritise some research.