The designer doesn't know what the final game will be.
This can be easily tested with a few questions. If the game changes by every question or suggestion, there's no focus.
This can be due to the fact that there is so many things coming to the game and the core is not yet nailed.
The designer doesn't talk to others or the communication isn't clear.
Designer should talk daily with the other team members. I recommend using feature presentation.
If the communication isn't clear, then is something where the designer needs to work on more. I use paper and pen a lot on daily bases to describe how things work or should be. It's also a matter of listening to others feedback and taking in the parts that are best for the game. It's also a matter of respecting others views and encountering them. Designer is not the king but more a funnel taking in all the good ideas for the game and these can come from other people too!
As a designer you should not be afraid to fail and to present something others won't like. It's part of the job to find the best possible solutions.
Now most of the games you'll be doing might not be for you. Also, what works on one platform can be different on the other platform. So study the platform you will be working on. Don't assume how certain people play, but study, playtest, ask and learn.
Farmville was an extremely popular game. On Facebook. When they came to mobile, the version of the UI was quite hard. They had not paid enough attention to the UI on the different platform as it was directly ported from Facebook to Mobile.
Then came HayDay with really nice way to harvest the crops.
Also designer is not the dictator stating how things will be. Within a team you take in others feedback and ideas to improve the game. You're like a funnel for the all possible ideas and you pick the best ones that make the best possible game.
So often I hear "Player will get it." Well, they won't. You have to be obvious and help the players.
You cannot rely on the players to read the texts or really paying attention to the tutorial. Aim to make your game understandable without text and tutorial.
Think of a screenshot from Angry Birds. You can probably get the game from any screenshot.
Try a game called Peggle. Analyse the gameplay and then how they reward you.
The game is pretty much random. You shoot the ball and just wait where it goes.
Ok, now take away all the feedback, text, particles, effects and the lot.
Ask yourself would you play the game anymore?
You cannot overdo rewarding, can you? And you should rather way overdo it than underdo it.
Rewarding is extreme important and you should add it as a feature to the backlog to have it.