It is important to have a core idea for a game, its also one of the first things you'll do.
Your ideas' may be huge and plentiful, and thats great! Unfortunately you might not have hundreds of dollars to spend to have your full dream developed, and that's a shame. That's why it's incredibly important to sometimes start smaller and stay that size, if not you'll encounter what people call "scope creep" (when your idea gets too large to be fun/to develop).
Maybe if it's your first game or you're still beginning; you could make a game that exists, not necesarilly for publishing, but rather so you have a reference of what a final product should look like and so you can learn your engine (Your tools). Of course you don't have to do this. I didn't, but I wish I did.
Now if you're looking for something bigger, let's talk about that. Every game needs something to stand out. Ask yourself: "why would people want to play THIS game?" Is there a fun key feature? A cool art direction? An odd mix of genres (like a romance horror movie, but for video games)?
Id suggest that once you have your idea to not think of the specifics, once you get to prototyping the concept, you'll probably be making a skeleton of the game's inner workings and systems (Aka: no fancy stuff). Now open up your engine and get started on that prototype!