When deciding whether you want to include any functionality / feature then you have to link it back to an actual business requirement.
When planning a site the first thing I do is spend time identifying specific, measurable business requirements for what the site needs to accomplish.
Having done so then you can come up with feature requests and ideas of what you want to include in the site. Having this big list of ideas (such as your splash screen) you then need to link those ideas back to the business requirements, and if the feature you have come up with can produce a measurable benefit on a requirement then you're good to go with it. If it's not going to provide any measurable benefit then really you're just going to be (at best) wasting your time and money implementing it, and at worst actually detracting from the goal of the site.
UX isn't about coming up with fancy devices that people like looking at; if you can't measure any benefit in doing so then it's not worth proceeding with. It's good to come up with all these ideas up-front regardless of how crazy they may or may not be because they very well could provide a business benefit. But equally they may not.
If 'amaze user visually' is something you can measure, and it has a direct measurable benefit to the business requirements then go for it. But that's going to be something hard to measure, in my opinion.
