I've seen the thousand yard stare that gets triggered when you start talking to a game developer about all the obstacles standing in their way. Trying to get a player through your marketing, into the install, through the tutorial or first level, and past the first month of play can feel like a horrible escort quest; there seem to be a thousand monsters waiting to take your player away, and they seem to have awful pathfinding... stumbling through the game flow you imagined for them.
Unfortunately, you can ignore it, you can complain about it, but you can't just wish it away. Followers of the blog will know I'm obsessed with friction. I believe there are many ways that we as experience designers put silly obstacles in our players' paths. The key to doing something about it is prioritizing it so that you don't try to boil the whole ocean at once. (Honestly you shouldn't try to boil the ocean at all... it's not a very nice thing to do, but you get the point).
I suggested to the crowd at Casual Connect Asia that you should ask yourself two key questions to help you figure out which points of friction are critical, and which you can live with.
How beneficial or harmful is this friction?
How avoidable is this friction?
Tech obstacles like slow load times, and poor connections don't help players and don't make you any money. They're worth fixing.
If you missed the discussion in Singapore, I wrote a recent Gamasutra article that can help you think through the topic.