The technique I am finding that works best for me is to not want the players to do anything. Don't get me wrong; if the game is set up with the premise that they are a group of heroes with the mission "Rescue the Princess", I expect them to try to rescue the princess. And I have players who generally get on board with the story, particularly if I inform them in my descriptions why their character might be motivated to do what I expect them to do.
What I have mostly stopped doing though, is trying to predict *how* they are going to save the princess. Assuming ahead of time what the players will do easily leads to railroading, or the players getting stuck because they can't seem to figure out how YOU thought they should rescue the princess.
So what I have been doing more and more, is to just write situations or stories and then let the players blunder into it and see what happens. I just make a few rough notes about the NPCs that include their motivations (what they want and why they want it) and some notes about any important locations and the current situation, and I'm pretty much ready to go.
The important difference here, is that I know what will happen if the players do nothing. I can not know for sure ahead of time what the players will do, so I have stopped trying to predict it. If the players decide to just sit in a bar and drink for the whole session or just blatantly ignore any plot hooks thrown their way, I know what will happen. If they try to solve the adventure but fail miserably, I know what will happen. If they come up with some ingenious solution I did not think of, I don't get thrown off my game because I did not have a planned way for them to solve it.
Basically I make adventures like this now:
1: Put problems in front of the players
2: Throw a bunch of tools and materials around the area
3: See what the players make of it
Disclaimer: No princesses are rescued in my games.
Edited by Ralzar, 05 April 2014 - 03:41 AM.