1. It shouldn't be boring. Find the boring spots, and either take them out, simplify them, automate them, or compress them.
2. It shouldn't be too hard to do over and over again. If you had to do something 100 times in a row, how would you make it easier to do? Can you make something happen with 1 mouse click instead of 3? You would save 200 clicks! What if you had to do it 1000 times in a row?
3. It should be easy to remember how to play. Human short term memory has limits, and those limits are part of what you are working with when designing a game. If people can't remember the rules, especially in a board game, it's probably too complicated. Think of Chess or Go. What makes those games good is not the complexity of the rules, but the complexity of the emergent strategy.
4. Honestly evaluate whether or not you have fun when you play it, or play different sections of it. If it isn't fun, it needs changing, even if it is clever or well done. Fun is one of those things that isn't about 'balance' or strategy, or game mechanics, or sophistication. People have fun chucking rocks into water, for example.
5. Listen to what people say after they have played your game. Most people aren't clear about what exactly they didn't like, but they can tell you when and where it occurred, and this is useful for leading you to trouble spots. Usually they will give you ideas for solutions that don't fit your design, but they know what needs work, so pay attention, and come up with a good solution.
-Jim DuBois
Arcane Journeys