At some point in time you are going to have to cut the scope of your game. You are going to run into a resource issue be it time, money or people, or perhaps a certain feature took 10x longer to implement than planned. It's going to happen.
When scope cuts come, they feel like a punishment. "Everyone didn't work fast enough so we have to loose features!" But the reality is that people have a very, very hard time accurately estimating time to complete a task. This isn't saying "people in the games industry" can't plan well, poor time estimation is a inherent human flaw. And most people in they games industry are human. So from the very beginning of a project, you are going to have incorrect time estimates.
Step 1 - Scope Cut
The very first step in a design process should be to cut scope. You should trim away anything not truly needed for the game to be complete. When I say complete, I mean that the game covers all the core dynamics and compulsions truly needed for the game to succeed. For example, if your game is a dungeon diving game, you need loot to drive the replay compulsion. You cannot cut loot. But could you cut the feature that loot changes your player's appearance? Yes, you could. Or could you simplify "left-earring slot, right-earring slot, eye slot, nostril slot and head slot" to just "head slot"? Yes, you could.
Find the most basic set of features that completes your game. Why is this so important? Because these features CANNOT be cut. You need to define the set of features that if one of them gets cut, the game does not work. It becomes another game entirely and has to be redesigned.
A complete game design is like a recipe for cake. It will not be cake without all the ingredients. Adding more sugar and butter might make it tastier (but not good for you!) but removing sugar would make it something else. Biscuits, I guess.
Creep Happens = Cuts Happens
As a project progresses, new ideas and features get added to the game. At the time, a single item seems like a small addition, but multiple people making small additions over a few months adds up. Occasionally you should go back through your design and trim features that the game could live without. You will often find features that contradict or duplicate compulsions or dynamics you want to create. This way scope cuts don't feel like a punishment, but just a part of process.
Nearly every time our team cuts scope, we come out with a better product. It's more focused and streamlined and we create more time to polish and refine the existing features. The only time cutting scope is "bad" is when it cuts into the features of what is a complete game.